 Juan, thanks for joining this webinar today on the IEA. We will wait still five minutes to make sure all the people that registered have the chance to connect. And then we will start with the presentation and the exchanges today. Thank you very much. Please, you can use the chat or the question and answers bottom to provide comments or questions so that you might have on the presentation or what we are discussing. And then if needed, we can also allow some people to talk live depending on the time we have. And if there are people that wants to share experience or ask a question directly by voice. So let's say to start at 11.35, Paris time. I see we are reaching already a critical mass. We have more than 20 participants already connected. We're expecting some more, but I think time has cast to keep going. Actually, we're reaching 30 now. So already a very good number. Anyway, I wanted to make sure that this section is recorded. If you have any problem, let us know. And then a video recording of these will be shared with all of you. So in case you have lost a part of the workshop or if you want to share with some of your colleagues, you will have the recording for you. But let's start with the quick questions on Mentimeter. Please connect to www.menti.com, as you can see in the slide, and use the code 4894 6111. So we can start with some questions and answers that you can all see. The results on the screen is quite interesting. We start with a simple one that is, where are you from? And let me share the screen on the Mentimeter slides. You can use your phone or your computer going to www.menti.com. And the code is the one you see here in the slide, but you can also see it in the new screen. I'm going to share up in the right side of the screen. As you can see, the first person that answers is from the Philippines. Welcome. You are from the Philippines, our colleague. And once you start answering, we'll see also where are the other participants from. I'll let you the time, because firstly, to go to the site and put the code text a bit of time, but then we'll go quicker for the other questions and answers. We'll need to change the code only once during this presentation. So I think it will be quite quick once you are done. So we can see there is a participant from Sweden, from the KTH Energy Systems, from Kigali, Kenya. We are from Rwanda. We have a generic from Africa as well. I will not answer, but I'm based in Paris, but I'm Italian, as you might wonder, from my accent. And we have also known today with us that he's from Cote d'Ivoire and works at the IAEA and is based in Paris. Cap Verde, Zambia from the capital, Mozambique. We have already two countries that speak Portuguese. I hope my accent is understandable also for you today. Maputo, Mozambique, Cap Verde. I'll let you sometimes, so this is the first question. And for the others, everyone will be already set up with the main team. We can go faster with the other questions. Ghana, that's nice. We have a good representation of Africa. And also, we have the Philippines from Asia. That's great. So we can share experience from different continents today. I have two minutes more, and then we can go with the next question, actually. Ghana, I have 11.40 on one minute, we change. So Mozambique is becoming bigger. This means there are more than one participants that are joining us from Mozambique today, same from Cap Verde. OK, let's go to the next question. This is more like a quiz. It's a bit funny how it's done. I'll let you answer this question. It is quite simple to understand, I hope. We are just asking, what is the number of people without taxes in sub-Saharan Africa? We'll have three options. Please answer the question. There are eight minutes, but of course, when I see many people answered, I will stop the chronometer. We'll see the answer anyway during the presentation. Eight people already answered. Let's wait for 15. I think there were 15 participants in the main team. OK. I think we can let the answer. The many passcodes didn't see this up here on the screen. It's 48, 9, 4, 6, 1, 1, 1. Everyone has voted, so we can see. Most of the people answered correctly, that is 600 million people in Africa today without access to electricity. It is quite high share of the population, as we all know. It's almost exactly half of the population. Now, let's go back to our presentation. Here we are. And let's go to the first slides. So the first slide is just about the IA, who doesn't know who we are. We've been funded just after the first oil crisis in 1974. We are an intergovernment organization that now don't deal only, of course, with oil crisis, but has been funded to keep member countries of the IA in a secure energy environment. This means that, for example, IA members need to keep a stock of their oil demand in their national borders or in friendly countries in order to be able to release this stock if there is an oil crisis and enabling the price of these products to stay more stable and don't fluctuate as much as it could be in an event of crisis. Recently, during the war that is happening and the energy crisis that in 2022 shocked a bit the European markets, especially for gas, the IA had done some work with its members to try to keep the prices to don't fluctuate too much. So as I say, we are an intergovernment organization. And now we cover all fuels from fossil fuels to renewables, energy efficiency, and all technologies. We are based in Paris. We are about 350 people working here, but we are quickly growing. And our main objective is to ensure the world is a secure, affordable, and clean energy supply based on all different energy products and technologies that are available today. One of the most important things we do as is written here down is that we provide data that work with government partners and private sector to provide a global view of the energy sector. So we cover all countries in the world. We have energy balances for all of them, but we also go beyond energy balances and we cover other indicators as, for example, access to electricity. And I need to thank most of you that are joining today that always cooperated with the IA in building better access to electricity statistics. The IA work in Africa has been expanding recently. We have a partnership with the African Union, UNECA, the African Energy Commission. We have collaboration with the African Development Bank. And we have many countries that are today association and some that will become association soon. For the moment, you have Egypt, Morocco, and South Africa, but you're working with other countries that showed interest to become an association country of the IA. We have been organizing some meets with African energy leaders in the past years as you might have joined. And we release many data and analysis and publications that cover the world energy sector, but also the African energy sector as for example, the African energy outlook that we released last year. Let's go to see some results from the World Energy Outlook 2022. We have seen last year analyzing the data and doing some estimation of what was the situation after the COVID crisis and the energy crisis that happened in 2022 on new connections for access to electricity. So we have seen a strong slowdown in many countries on new connections, but also a strong impact on households income that prevented them to be able to connect to new forms of energy, but also to consume energy by having affected their available, let's say income for energy. What you've seen last year, we estimated that in 2022, for the first time in decades, since the IA started at least tracking access to electricity in 2000, we have seen that there's been a rise most probably of the number of people without access in the world. To reaching, again, the levels that were in 2019, in 2022. And this rise happened mostly in Sub-Saharan Africa, as you can see from this graph here. Yeah, we see only before the crisis, but I will soon show you what happened according to data and estimations starting from 2020, so when the COVID situation started. And after the COVID, we also estimated that unfortunately in Sub-Saharan Africa after some years of decline that started in 2013, as you can see from this graph, the situation of the number of people without access started to reverse. And again, we have an increase of the population without access in Sub-Saharan Africa since 2020. And we estimated that in the worst case scenario in 2022, the number of people without access in Sub-Saharan Africa reached again the peak almost, reached the peak of 2013. So we practically erased some years of strong improvements because of the population kept increasing in these years very quickly in the continent and new connections had been unfortunately slowed down quickly. But let's go back to Menti. So we'll keep the same code that we are using before. And you can answer a new question that is related to what we have just seen. Some of you already answered it. This was how the COVID for you has impacted and affected access improvements. And that's how do you expect this will impact 2023? I've seen seven people already answered this, please continue to do it since then most of the people believe or have the feeling that 2020, 2022, 21 was the worst impact on access improvements, very strong actually in 2020, still very strong in 2021 and 2022. And it seems the feeling for 2023 from these seven people that answer is that the situation is better. The impact is reducing from what you've seen in the last three years. But I give the time to someone else to answer since only seven answer are ready to this. So it's still Menti.com with the same code as before, 48, 94, 61, 11. And let's see what is the feeling that you have on this. On our estimates, of course, is very similar to what we are seeing now from your opinion that 2020, 2021 was very difficult for different reasons, changing priorities of the use of budgets, supply chain disruptions that was very difficult to get the material needed for, for example, grid extension, et cetera. And we see like everyone agrees on that. And still 2022, the situation didn't improve because of the energy crisis kept many problems. And also the COVID, the still was very active in China that is one of the main suppliers of electrification material was affecting the supply of these materials. Very interesting to see that. If there is someone that want to share in the chat or later by voice their experience in their countries on the impact of COVID and the energy crisis on their electrification projects, please feel free to do it in the chat now. And maybe we can also take a time to exchange on this interesting and important topic later. So let's go back to the slides. And as I was saying, please feel free to share your opinion on this impact and what was the most impactful, the most impactful effects from the COVID. Okay. Now you are back to the presentation. So let's go to see exactly what we are gonna cover today. What is the agenda? So we'll see what are the objective of the guidebook that we released recently on improving access to electricity statistics that it'd be shared to you, but we'll share it again soon if needed. What are the key concepts that are in this guidebook? And what are the key concepts of improving access to electricity statistics by the use of supply data with one overview of what is the process for tracking access to electricity from the collection of the data to the dissemination or publication of this data. So we'll see all the steps. In the guidebook is explained much better, but today we, some graphics maybe it will be interesting experience to see. And then we'll look at the IEA templates that we use for data collection. Some of you already know because there have been filled and the data will be submitted. Thank you very much again. And we'll introduce a quick exercise that we would like you to take. I don't know, today you will not start doing it actively in your screen, but we'll send you later the Excel files, the solutions of this exercise so you can practice. This exercise is based on some data that we invented for tracking access to electricity and is based on filling the IEA questionnaire with this data that will then provide access to electricity indicators. So let's start with the objective of the guidebook. The first objective is why we start this guidebook. Is that universal access by 2030 is a critical sustainable development goal. As we know, there's the G7.1, but we are failing behind that, as you've seen in the figures before, space and in sub-Saharan Africa. Well, the situation got worse unfortunately because of COVID and the energy crisis. And to accelerate progress, we think that data are the most important tool for starting any electrification planning. So if a country or government relies on quality data and data that are easily available and timely as on the time, so up to date, their planning efforts will be much better. This data starts from a simple tracking access rates at the national level to having like a jurisdiction village level or GIS, so geographical, the design aggregated data on what is the access situation today. And also we've seen for planning other additional data are needed, but today we don't focus specifically on this. As you know, the idea has been, I don't know if you know actually, but we have been the first international organization to produce a global database on electricity access information. We start in the very early 2000 and actually in charge of this was our current executive director, Farty Birol, that was in charge of tracking access to electricity and started this project many years ago, now more than 20 years ago. Like today we are official COCUS audience for all the SDG7 indicators with the World Bank, the UN, IRENA and the WHO. We just released, as I was saying, the guidebook for improving electricity data statistics. And this is meant to be a guide for government to improve access to electricity indicators by the use of supply side data. What do we mean by supply side data? We'll see better after, but we mean by using information from energy services providers. So this can be utilities, distribution companies, can be mini grid operators, it can be solar system distributors, or it can be other national institutions that track the supply of electricity in the country. Like for example, an energy regulator tracks from all the different distribution companies in the country, how many new connections have been done, what are the total residential connections, for example. Use of supply data is very accessible and cost effective, so it's ideal for tracking progress on electrification because you can use this data when the system is in place to track even access by month. So knowing last month or even like last week how many people were connected to the grid or to other kind of systems. And combined with household service, they provide a very high quality and timely estimates of electricity access. We'll see later that supply side data have many advantages, but also household service have very important advantages. So providing information bottom up from the consumers. And the two sources should be used together in a sustainable and a strong data strategy. The guidebook also provide a methodology to estimate the access from all grid technologies, namely standalone system or solar system, for example, that are very important and need to play a very important role to achieve universal access in the next coming years, at least for the near term before the electrical grid can reach most of the population. The guidebook also slides how to start using this information for having a geographical disaggregated access information system. For example, we all know that many utilities or distribution companies know which villages they connected, where are the people that paid their bills. So the information is there, but sometimes it's not unfortunately well used. There is not in place a system that permits to track, record, report this data in a way that is easily usable, but the steps to achieve that are not very complicated. They will not look this in focus, but we'll have other sessions of course and other work with our partner countries to look at this important geographical disaggregation more in details. Let's move to the key concepts of access to electricity tracking. First of all, we need to have definitions. What is the definition of access to electricity? The definition of access to electricity is accordingly to what we wrote and has been peer reviewed by many countries and many other organizations that the household is access to electricity when it's connected to a grid or of grid source. And this grid of a grid source needs to be able to provide a minimum level of energy service that needs to be defined. This minimum level of energy service and its definition is very key to then understand which are the systems that can provide access to electricity. So what are the connection and households that can be counted as having access to electricity in the access rates indicators, for example. An example of this is if you have a minimum level of energy services that includes the user, for example, of lighting, a television and a fan, probably the smallest of grid solar system like solar lanterns cannot be included in here. But if your definition is a bit different, some systems could be included. As we see later, I will show you what is the definition that the IA proposing is a guide book as the first point or the minimum level of energy that can be considered as access. And you see the definition is very interesting because also include the fact that the household needs to be able to improve and increase their level of energy. So their consumption, their demand and their use of electricity services. The access to electricity rates represents the share of the population that benefits from access to electricity as defined before. So it can be calculated as the population with access divided by the total population. Once, when we do that on a top-down approach from the utility data, for example, on connections, we count how many households are connected today. We need to estimate how many people lives in these households, for example, using the household sites data that can be different in different jurisdiction, different states, regions or areas. And once we multiply these two, we can achieve to understand or to estimate the number of people benefiting from access that divided by the total population bring us to the access rate. So existing data and surveys can help also track the different household sizes, for example. Now let's keep the definitions and let's see how we define different connection types. Here we'll define mainly three different connection types. The one is households connected to their main grid. The grid is a network of transmission and distribution lines connected to electricity generators that can be centralized or decentralized, this doesn't matter, but they are connected to this main grid. Then we have mini grids that are small electric systems that comprise a generation unit or units and distribution lines that are linked to households and other console. Sometimes mini grids can be connected to the main grid. And when this happens, reporting of the connection must be done to avoid double counting only by the company that is selling to the final console. So for example, if a mini grid was operating before in a village and was serving some households and then the grid arrives, the mini grid connects to the grid, if the mini grid operator continue to supply energy as the official distributor to the households, we'll count this connection as a mini grid connection. In the case the grid will take over the business and the mini grid operator will sell, for example, their infrastructures, we'll count this now as a grid connection. This is very important to avoid that there is a double counting and we still count this as mini grid, but elsewhere we are counted as a grid connections. So each connection will be counted by default in our guidelines by the company or by the type of the company that is distributing the electricity directly to the households. And then we have standalone system that are not connected to any grids and they serve one household only. This can be solar systems, it can be fossil generators or other types of technology that include rechargeable batteries, micro hydro, wind, et cetera. These things are generally small. For example, for the off grid solar systems, we'll have solar systems that start at 10 watt peak and we'll see later how much energy services can provide and then we'll have other smaller type of systems. But let's see the IA definition of access. What is the minimum level of energy services that I was mentioning before that the IA consider as access? So here we see on the green, let's say, the green bar, we can see the energy services, lighting for charging, radio, TV, fan, refrigeration. And in the columns, we have put different types of technologies that provide the electricity service. These are start from grids and large generation sets, mini grids, standalone systems. So for example, we have a solar systems bigger than 100 watt peak, standalone system bigger than 50 watt peak and smaller than 100, et cetera, et cetera. And we can see that each different technology can provide different types of energy services. This if we follow the column and what kind of energy services, so green bar, it touches. For example, a solar lantern or a dry batteries can provide lighting normally, not more than that. While a multi-light system, multi-light solar system can also provide phone charging. So it's a bit bigger, have a bigger photovoltaic panel, a bigger battery and it can also, besides having some lighting bulbs, can also provide the possibility of recharging your phone. And once we move toward the right of this diagram, we can see that energy services increase. For example, for a solar on systems of 10 watt peak and more, we can also add to the phone charging, the utilization of a radio, especially if the equipments are of high efficiency. When you see the green bar that becomes with stripes, it means that this is possible, this energy services is possible with this technology if the appliance system devices uses of high efficiencies. For example, a very high efficient radio, LED lighting bulbs, et cetera. And this of course will evolve as the efficiency of appliances will increase, we can provide the energy service with less electricity consumption and power, so with smaller systems of course, as we can see from this arrow that we designed. And if you go all to the right, we see the grids of course once, when they are stable and what they are serving because in many countries, unfortunately this is also the problem that the grids are not able to provide and supply electricity all the time and with a good frequency. But in theory, a grid can provide lighting for charging, radio, TV, a phone, refrigeration and other larger appliances. Ours can do mini grids when they are well designed and also large generation sets. The minimum level that the IA defines us as a first access is the basic bundle of electricity. And this is the one that we see now in red in our diagram. This includes so having lighting, our radio and phone charging. And as you can see our definitions also implies that the level of service is capable of growing over time for this also. So today's connected with the small solar system that provide this, but the plan is that this also will move up the larger of energy services gradually and reach more and more energy services as times pass. Then we also define the special bundle to provide us, so let's say a target for later, okay? We can have a rich access with a basic bundle but we want that all households in the country at a certain time will reach higher energy services. So the essential bundle also include a television and a phone to keep the household cool. There is a definition that is more detailed that tells you like how many hours per day, the phone, et cetera, but I think we don't need to cover this today. In the guidebook all the information is done and that's where the extended bundle that we also add are refrigerators and larger appliances. This is once we reach almost complete level of energy services by the household and we see that we will need all very big standalone system. So in the villages that are very remote this will require that the household is equipped with a large solar systems and today we know there are problems of affordability. So this shouldn't be for all households the target for tomorrow, but for the day after tomorrow, yes. So in the long term we need to also keep in mind that we need to be able to grow access level over time. But for counting and access to electricity, the IAEA proposes in the guidebook to use the basic bundle. So lighting, phone charging and the radio to keep informed is very important the service that the radio can provide to people to know the material, to know the news, et cetera, et cetera. And let's move now to the next slide. Let's compare the minimum energy levels with other organization. This column, let's say in the different colors represent the World Bank multi-tier framework tiers that many of you I believe knows. So tier zero is no access, tier one is very basic access tier two is increased, et cetera, et cetera. The World Bank is a very good definition of that. And I use that to put together all the different IAEA definitions and that's what some others. As you can see the World Bank, let's say first access will be tier one and this consider very small solar multi-light systems or free VAT peak. The IAEA basic bundle is a bit higher than that as we have seen before. We won't have to include the radio in the first access or the minimum level of service defined, but the basic bundle will still fell in the tier one of the World Bank. And we can see also here the essential bundle we've seen before where we have the television and the fan. It sits in the tier three and the extended bundle of the IAEA that has includes the refrigerators. It sits between tier three and tier four. We know also that several organizations now are advocating for greater ambition in access provision. For example, the modern energy minimum are the organization that is called energy for growth hub and they propose the modern energy minimum that is a minimum level of consumption of electricity to provide the increased level of quality of life. This minimum they define as 1000 kilowatt hours per person per year of which 300 kilowatt hours are for residential use. We can see that this as compared to the extended bundle of the IAEA is not much different. So we are in the same page. The only thing we want to say again here is that this is a very good goal but it shall not be the goal of first electrification because many households are not today looking for this very high extended bundle of energy services because they cannot afford neither the energy, the electricity or the appliances to consume this. So they can start at the basic bundle but keep in mind that this needs to grow over time. But the important point here is that harmonizing electricity access definition is a very important task that still remains challenging. We see that many countries includes different type of connection or different type of definitions in their data they share with us and we hope the guidebook we publish will help having some harmonization but what we stress all the time besides harmonization is transparency. So once we have metadata and information and documentation that clearly explains how an indicator is calculated for example saying our access to electricity rate includes only grid connection for example and then there's much more we can say on how it's calculated without the sources. This already provide a very transparent information for the data users that are comparing maybe with other countries or they need to know how to better start for electrification planning in the country or a specific region. And let's go to the next slide where we can see the difference of sources for collecting access to electricity data. The one that we cover in the guidebook is the supply side data and the other is demand side that we have seen or mentioned before. So supply side data are the data that are come from distribution companies and for example, give us the number of residential connection of customers or the consumers that they are supplying electricity with. While demand side data are coming directly from the consumer. In our case will be from the households and they're like most often coming from surveys or censuses where the national statistics statistic office for example, ask households if they have an electricity connection or what type of connection, et cetera and gather directly the formation bottom up. But let's see what are the benefits and challenges of both of these sources. Advantages for supply side data sources is that it's less costly, easier to implement because there is no extended also survey to put in place but you can directly contact the number of companies serving electricity. And so the high frequency, the data can be available. I can ask the utility, how many people have you connected last month and if there is a system in place at the utility level they will be able to tell us. The demand side data source however, provide also very good advantages and this is on bottom up information that permit us to avoid double counting that can come from supply data. We can more easily count off grid connection and inform a connection because directly asking the household will answer you if they have electricity in their place. And ad hoc surveys, so a bit extended surveys like the MTF for example, can provide also much additional information on how the household is consuming this energy, what is the quality of the energy they are having, et cetera, et cetera. And so this can also be as you can see in the guidebook we published some of these indicators additional can also be estimated from supply side data. What are the challenges however? For supply side data, the challenges are that it's difficult and needs to put in place some connection, estimation or discount factors to count for informal connection or submetering for off grid systems and for the use of grid system as backup for example. Because this will be a double counting if we count us now so that's both a soleno system because the grid is not very stable and we count both the sale of this soleno system and the grid connection will count this out, so us too. So this needs to be taken into account to avoid we are overestimating our access to electricity. What are the challenges of demand side data sources? They are costly and they cannot be run every year. So they are run most of in many cases between five and 10 years surveys or censuses. The sample design in remote areas is complicated to be representative sometimes because remote areas or rural areas in the countries are very different depending on the region. So having a representative sample is not very easy but I'm sure in your country's national statistics office are very prepared for that and can help you in design good surveys. And so there is a subjectivity or the response of the household. When you ask an household if they have electricity one household can answer yes. For example, if they have just a flashlight because yes, I have lighting, so they answer yes. While another household will answer yes only if they have a grid connection, for example. So for a service is very important to train the survey or the people that go surveying households and also to provide good information to the households that are respondents or to the respondents. The main message here is that a strong access to literacy tracking strategy means to combine these two approaches. Supply side and demand side data to build on the synergies and the advantages of both of them and mutually reduce the challenges that we have in both of them. For example, the use of supply data reduce the fact that surveys cannot be run every year and so provide a more timely information adding up on surveys. So a service can provide some more precise information on the number of people that are connected so avoiding the double counting as I explained before. But let's see an example of an access to literacy data strategy that I was saying must rely on the complementarity of demand side and supply side data sort of. Let's start with this example. We have a country that, I don't know if you see well, I've seen the thing moving, but let's see a country that had a survey in 2015. So we can see here on the DS means demand side, so will be the survey data. In 2015, this country had a survey where they've been counted the number of households that have access to the grid to mean grid and to solar systems. The same year, if we look at the supply side data, so coming from the companies, the distributors, we see that there are many differences actually. For example, we had 3.6 million people connected to the grid while the survey says that there are 3.1. Who is correct or who is not? This is another question, but it's very important to keep in mind that surveys, if they're representative, you can provide a very good information to avoid like, for example, to take into account informal connection, informal connection, subnetoring, and the supply side data normally for grid access should be also very good, but we see in many countries that there are issues, for example, on the number of meters that are serving multiple households or in some countries we also heard that one household could have more than one meter. This is what, quite surprising, but is happening in some countries that works with us on tracking access to electricity. So we can see there is a difference here. So what we can do is calibrating our supply side data on the surveys or censuses results to avoid the double counting, the connection that we missed, et cetera. But we need to ensure that the survey is representative. Then what happens in the next years? We don't have any more surveys or censuses till 2021. And so we need to rely on supply side data from our distribution companies. As you can see, like the data in 2016 has been, in this example, calibrated based on the census that happened in 2015. So we have a correction compared to the number we had in 2015. This something is happening in many countries today. I believe some of you are here connected today. We exchanged about this exact problem that a survey that just happened or a census are providing new information that will require correcting and calibrating our data for the years to come. So for the years to come, till the next survey is done, supply side data will provide very good information and we can continue track our grid, mini grid and solar system or off grid connections as we can see from here. And then you see this counter in 2022 as another survey or another census. And we again have a new information that will then calibrate again our supply side data for the years after till, for example, 2023. But also we can, this is very important when possible correct if needed our historic time series at least going back some years to be sure we are having a good continuity in the data and analysts and data users will be able to use this data for their analysis in trends. So survey sensors can also help recalibrate historic data. And now let's move to the overview of the complete data process. First looking at what are the fundamentals to set up a sustainable and functioning and efficient access to electricity data strategy. The first, let's say fundamental is identifying all key stakeholders that work on data. This is not only the data providers like the utilities, the mini grid operators, the energy regulator, other agencies in the government, but also data users. So what are the main data users that we can and we want to put in the discussion to include in the discussion, this can be policy makers, planners, energy modelers in your ministry or there can be from other ministries or other government agencies, of course, but can also be from external agencies as donors that are present in your countries that help you working on electrification planning. If they're also included in the discussion data in the data discussion, they can provide helping, highlighting and identifying what are the data needs because before starting any strategy, we need to know what are the data we want to collect and what are the data the data users need. So once all these key stakeholders, data providers, data users are identified, is very fundamental to establish a data working group that will meet periodically to discuss on the data needs or challenges on timeline, like for example, when data needs to be provided, submitted, et cetera, et cetera. Providers organize training for data providers, et cetera, et cetera and training also to data users for use this data. The second principle is the legal foundation. This is key to have strong legal foundation for a ministry or another agency that is tracking access to be able to ask for the data that they require for. So for example, having a mandate from the government to collect this data, but also if there are some policies and laws to enforce data collection, like for example, the total electricity suppliers are required to provide connection data to the ministry or to the energy regulator, et cetera. This helps a lot in making sure these companies works in the data exchange process. Of course, as I said before, establishing a data working group and including data providers is already a very good step because the data providers will be part of the discussion and will feel like this is also their projects. It's not that they are receiving a summary cast of data, et cetera. But then the legal foundation are key. As we have seen in many countries, the countries have a strong legal foundation for data collection. They achieve to produce much better data than countries that are lacking today this. Second is defining a PR plan for the data work. This means that all the stakeholders that we and the T5 should agree on a realistic timeline and scope for the data work. On the data collection tools, for example, do use an Excel questionnaire that we send to the utility. They'll send our back or we put in place an online platform where utilities automatically submit us data on new connections, et cetera. And also what are these aggregation of the data? Like this can be done gradually. We start, for example, national level. Then the utility needs to start providing data based on geographical definition. For example, by jurisdiction, by region, et cetera. This also require, sorry, agreeing on writing down documentation procedures or from this tax having an agenda like a calendar of data collection to data dissemination and publication. The fourth principle is resources. We need to allocate financial, human and IT resources to data collection because if there is no trained staff, if there is no money to do data projects, if there are no computers that are good enough for doing this work, there is no data. So resources need to be part of the plan and they need to be secured on a long term. Not only like for one year for a specific project, it's better aiming for a long term engagement for a budget for data collection. Generally on energy, but also on access to electricity. And access to electricity can be part of the general country energy data strategy. So resources can be also shared in there and make it more efficient. The fifth principle is training. We need to provide continuous training to the staff that work in our ministry or agencies that process the data. Continuous training to the data providers, to the data users and to other statisticians. Training can be, for example, as participating today to a training provided by an international organization, but also internal training like of more senior staff that trains junior people that arrived in the ministry or the ministry itself that train data users in the utilities or in the energy distribution companies on the data process. And to all these, like especially for training external people, the first principle of having an established data working group is very key because it will very much facilitate all these exchange. Because also the utility ones can come to the ministry and train the ministry on their way. They use data so the ministry best they understand or their challenges with data available and together they can. Better plan. And then last but not least is to always make improving data coverage and quality as a priority. Not sit down on, okay, now we are doing enough. We have a national access rate, but still with a working group plan to cover new and relevant indicators. So if there is new needs and priorities you might want to expand the data collection you are doing, you can aim to have a better geographical or technological desegregation for your data, et cetera, et cetera. And now let's see how it looks like the process of data working flow. Yeah, we design in the diagram the national entity tracking access to electricity and there are four main steps for data production, the data workflow. The first is data collection that is done in cooperation with data providers and external data providers that are not giving connection data but maybe also sites or other relevant information. Then we have the data users on the other side that also provide input as we've seen before in our working group to know what data we want to collect, et cetera. Then we have data validation, these are important steps. When we receive or we collect data we need to be sure this data are of good enough quality and there are no mistakes, errors or problems in the data. So if any problems is spot we need to go back to the data providers and try to solve the issue and have a better quality or more correct data. After we have data processing so the raw data that is being collected are put together to create the indicators we are looking for, for example, the access rate. And again, at this stage we can have further validation because when we calculate the access rate we can see again that there are some problems in the data we received for example, an access rate of 120% in a specific area don't make sense. So we go back to look at the raw data and we can go back to the data providers if there are any problems. And last but not least, data dissemination. Once we have done all this work to ensure we have very good data we need to make sure these data are easily available to the people that needs them for their work and so we need to disseminate them properly. This can be in different forms, et cetera and easily accessible, of course. At this stage again, we can have other validation because for example, our data users that receive or download the data from your website see something that believe is not correct and can help us in improving the quality. So at this stage. So the data validation we see is very central. We are in a continuous data improving mode. Like once we find a mistake on the data validation process but also at the other stages we can still continue to highlight or identify problems and so keep improving our data. Here I showing the same, exactly the same process but with more information this is a figure that is in the guidebook. I don't think we need to go more in detail. See here is very much the same we have seen before but for example, when you look on data providers we see what they include. We see from supply side we have a little utilities, energy regulator, solidification agency, industry association, energy distributors and from the demand side we have also service censuses. So for example, national statistic offices, et cetera. Then for data collection we put down some steps to establish data collection from the methodologies definitions that are for example, the one that we are proposing in this guidebook to be published, what is the framework, et cetera, et cetera. I let you take the time to look at this diagram and all the text that goes with in the published guidebook because today we want to see a big overview of everything we don't want to focus only on the process. So let's start now with the first step, the data collection and let's go to see what we need to collect. What we want to collect at first is connection. So we want to count how many households are connected through grid, mini-grids and standalone systems. That's meet our definition of minimum energy service as we have seen before. For grid, what are the main sources, as we have seen many times are electricity distribution companies, energy regulators and others. What are the challenges for grid data and connection? One is the informal connections or households that are connected but not officially sub-metering or backup supply. So when there are households that are using two different types of connection. What you can do for going farther on grid connection, we can look at the level of consumption. For example, utilities know through building data how much households are consuming. For example, the number of households in different consumption tiers, how many households are in the social tariff? For example, that goes up to, I don't know, 100 kilowatt-hours per month or 50 kilowatt-hours per month depending on the country. And so they know, for example, X thousands of households consume less than 50 kilowatt-hours per month. X number consumed between X kilowatt-hours and X kilowatt-hours. This is a very important information, the level of consumption because can help in estimating a potential demand of new connections or make easier planning for the future electrification. The geolocation of customers where households are located that are connected, many electric distribution company knows that already they just need to process the data better to make them usable. Sometimes they just know where is the node or which village is connected. Sometimes they know like more information on this, especially now that or in countries or in regions or areas of the countries where meters and smart meters are taking more share of the connection. But this is still a slow process, the smart meter. So now we need to rely on other kind of information. We can work on affordability. So knowing the tariffs, how much the kilowatt-hours cost versus the income of households in this region that can be provided by other agencies in the government. And also look at the quality of supply through already existing indicators like the side fee, the side and others that give us an idea of, for example, how many hours per day electricity is available, how many cuts, how long they are, et cetera, et cetera. For mini grids, who are the sources of data? For mini grids, we have a licensing authorities. This will be the best approach. So if the country have a licensing process, for example, the energy regulator that needs to validate a mini grid, this entity can track all the mini grids that been licensed. But it's important also that this regulator or a licensing authority keep tracks of who, which mini grids are still active because we keep track only on when they've been licensed, but we don't know if they're still operating, we can't come on counting issues. Then there are the mini grid operators themselves. So in some countries, there are not many mini grid operators so directly discussing with them can also be interesting because they can provide also the number of people also they are connecting. Donors, their programs or other programs the governments are launching with mini grids, industry association and other. The challenges are a bit similar to grids, but the most important one, I think is the non-operational mini grids. As a mini country, some mini grids have been put in operation and some years after are no more operating for different issues or bad planning or that was not considered before and the operation and maintenance of the mini grid or that the income from the mini grids were overestimated and these come back to knowing how much we expect that the sales of electricity will be from for example, the level of consumption can help in better sizing a mini grid so making it more profitable and no lose operation. Then we have double counting, of course, some mattering, et cetera. And for going farther again, we can track geolocation for mini grids will be easier because we know we can easier know where the mini grids is located and normally mini grids lines don't go very far. So if you know already where the mini grids located is already a good information in knowing more or less where the customers are located. We have the level of consumption of course that we know from mattering and building and again affordability as for grade. Important to note again is that what we want to know is residential connection. Sometimes grids or mini grids operators can provide the total number of customers but what you want to know for tracking access to electricity in this case is the number of households that are connected. So we need to ask for these and most operators knows for standalone systems again, we can have a licensing authorities that license standalone system operators or so long system distributors for example to operate in the country. And what we suggest and propose in the guidebook is that these licensing authorities work with the government to put in place some regulations that requires these companies to also submit data on their number of customers or sales or systems in order to operate in the countries. So this will be in the ideal case or something for the medium term where we established this licensing framework and then the licensing authority can directly collect the data with a legal framework. So requiring the companies that are licensed to operate in the country to also provide the data. Then we have the companies themselves but sometimes in some countries the landscape can become very crowded with many companies so it's being complicated. It can become complicated and we have industry associations. What are the challenges for the solar system as standalone system in general are the boundaries so which products we include what is the minimum sites that provide the minimum energy services. In the guidebook we've tried to select which are the products that can provide the minimum level of energy services we propose in the guidebook that is the basic bundle of the IA. And for example, in the case of solar system this is the system that have at least 10 watt peak so they can provide lighting for charging and a radio. Other challenges are the backup use, the free market like the companies that are selling directly to companies repeat sales or also that are buying more than one standalone system and end of life because if you just keep track of sales you need to know that these systems don't last more than five years especially the solar systems and so at a point there will be no more operative and they need to be discounted. But I think the most important issues here are the backup use that in some countries the backup use of solar on system is very, very big to because the grid in many urban areas is not very reliable so many households are buying solar on systems to keep continuity of the service. I see this number is very big if we count if you double count the solar system and the grid connection we will be starting having some errors in our data. And going forward again we can try to track geolocation this is easy for example pay go contracts when where the solar on system companies knows where the system is located because they track the consumption of the households and they charge them every period with some the charging period and they can also stop remotely the system if the households don't pay. But for the sales is more difficult because once an household boat when a customer go buying a system if it's installed directly but if it's not installed by distribution company knowing where the system is it's more complicated. Then the level of supply so this is come from the sites of system keeping track of ranges of sites of the systems that are in operation is very important to know what are the level of services that households are receiving. Of course on 100 watt peak solar system provide much more services than a 10 watt peak system this 10 times smaller. And then affordability of course keeping track of the cost of the systems for pay go what are the installments that households needs to pay. But let's see some of the problems we just mentioned with some diagrams. Up here we can see grid informal connections as you can see some people are connected officially to the grid but some other can be connected informally. Informally can mean illegally like they really go to the distribution line and put their cable they go to their house and use the electricity or informally they connect through another households directly. So the grid company will not see them as customers and so they cannot know if they are connected. The recorded approach that we could take is to exclude the informal and legal connection and work on policies and programs to ensure every households can be connected officially and formally. So for example, reducing or incentivizing the connection fee as many countries are doing for example, every cost is providing small monthly fee for the connection fee instead of one upfront cost installment for the household. So spreading the cost over time is proving that making upward more affordable to pay these over time than only one time is increasing the number of households that are officially connected. To solve this problem and so survey data can be used alongside for example, the non-technical losses from the utility. So knowing how much electricity we are losing that is not coming from technical issues or technical losses for example, lost us hitting the lines and this can give an idea of how much energy is for example, stalling or it's not paid for. Then we have some mattering that is when an household or a small company provide or a landlord provide some mattering to other households. In many cases in this situation, the grid company only know of the main household connected or the main meter and then this meter is split in different households by other service providers that can be private or not. And this makes that we are missing some connections that are happening in reality. So disaster are connected but the grid company data are not taking them into account. So how can we do that? The best approach will be to legally requiring the providers of some mattering services to report this number of connections to the utility but also as in this case, household service are very helpful to estimate what is the real number of households connected to the grid as compared to what the grid is reporting. And then we have double county connection I've seen before we can have also connected to the grid they use for example, solar on system as backup. Or we have also that have multiple standard on system. For example, they have a generator, they have a solar on system or they have two solar on systems, et cetera, et cetera. Here as well, we could try to require distributors or standard on system to collect data from their customers so when they sale, they can ask, is this for your primary energy use or it is for backup energy use? And that's we are again, survey, we see again that surveys are important to calibrate supply side data can help estimate a discount factor for a specific country or area to understand what is the backup use of, for example, solar on system. This is just an example. We have seen from some publication from some peer agencies that this number can change from 70% to 3% only. So there is no standard default value you can use for backup use. You need one that is representative of the different areas of your country. Let's see now a bit how to estimate standard on system connection or solar on system connections. We have two type of connection. One that is coming from electrification programs on the government and one that is happening in the free market. And then we have Pego contracts that as I explained before is selling a service practically. And another one is the direct sales of direct distribution of assistance. So an also buy a system or the program just provide directly a solar on system to the outside. Why the Pego contracts is when there are monthly installments that they also pays for using the solar on system. Under electrification programs for Pego contracts what we need to collect is the active customers contracts and these require companies to periodically report the number of active customers. Okay, so this can be written in tenders for example and do a tender for the electrification programs you can require already the solar on system companies that they will need over time to provide how many people are still connected. For example, in the next 10 years every month you need to tell the ministry how many people are connected in the villages that are covered by a specific procurement or tender. Then we have under electrification program direct sales and distribution and here we need to discount the systems that are no more operative. So we'll need either in the best case scenario to require a periodic verifications of system that are still in operation. So like having some third parties going to the sample of the household that received the solar on system and verify that these are still operative. And the best practice will be to include in the electrification program operation and maintenance. So there will be someone going to the household that has been provided with taxes to also provide the operation and maintenance services. And at the same time so tracking if the system is still operative or not. If we put that in the electrification plan and the service and the tenders we know that we are planning for a more sustainable electrification because we've seen in many cases when you just give for free a solar on systems sometimes the household either stop using it over time or sell it to someone else maybe that in the urban areas that are already connected to the grid for the cap use. So you lose this, but if you plan for operation and maintenance you also ensure that these also will have a long-term access to electricity. The same can be said for pay go contracts under electrification program. If you plan for a long-term electricity provision to the household instead of just a new connection you will get more results, more long-lasting results. Then we have the free market that of course is more complicated to track active customers. The best solution will be to create a licensing framework as you have seen for companies and requiring them to report active customers. This is easy for pay go contracts as you have seen because the company knows who are the household that are still paying them so that they have an active solar on system. But for the free market and direct sales is more complicated. So again, we need to discount and estimate the systems that are no more operative. But how do we do this discounting when it's needed to be done? First of all, we need to discount for the end of life. We need to use an assumed lifetime for the different type of products. For example, for solar on systems of time but above we can estimate that they will have a lifetime or more or less five years or less depending on what are our information. Then we need to discount for the backup use. This is very important we've seen before. This is especially true in urban areas, for example. And as I said before, it changed a lot depending on areas due to these digital countries. So it needs to be based on specific information. And then a discount factor for rapid sales as well. This is less important than the other two but still needs to be estimated. In the IA template that most of you received with which we recover and collect the data from your country, there is a small tool built in that permit you to discount for the end of life for backup use. So we just need to enter what is your estimated lifetime for the specific product. You enter the sales year by year and then the tool estimate how many systems are still in operation in each year. You can also enter a backup use or the percentage of systems that are sold that are used for backup if you have this information. And so this tool also correct for this and we provide you the number of systems that are operative and so the number of operative connection to solar on systems. Now we finish the data collection side. Let's go to see a quick overview of data validation. So the steps before I will ensure data are of quality. First of all, we have four type of data validation checks. One is validating and checking coverage and definitions. The second one is internal coherence. We have consistency with other sources and last but not least, plausibility. But let's start with the coverage and definition. Under that we have different kind of checks that can perform. One is boundary and perimeter that verify that connection data includes only, for example, the residential sectors or other boundaries can be that we include only the systems that can provide the minimum level of energy services that we define on our tracking. For example, for the IA, if you are speaking of the basic bundle, a minimum service will need at least a 10 watt peak solar on system. The smaller system will not be included in the access rate. The timeframe is correct. It's exactly the year we are counting. Are the units correct? For example, I received a connection are millions, thousands, am I using correctly? And also geographically, is this for the national level, is for a jurisdiction for a village and this is exactly what I'm using it for. Here an example of coverage definition checks. If we receive connections from a utility and we have total connection, we need to be sure that we include only the residential connections or the bluer. So we're going to further ask the utility to provide only residential customers and not all customers. Internal coherence checks. What are internal coherence checks? Verify relationships among data. This can be, for example, a arithmetic check that verify that data all adds up to totals. For example, verifying that total connection are equal to the sum of connection by type. For example, grid plus main grid plus standalone is equal to total connection. Or by area, for example, rural and urban, the sum of rural plus urban connections are equal to the national connections. And then we have time series checks that for example, ensure that there are no discontinuities in the data. No breaks in the time series that are not justified by something that happened. And here I will have a quick example that is also in the guidebook, where you can see, for example, that in 2022, the sum of rural and urban is not the same as national. You can see in the violet dot up there. And also we can see that there is a big break in the time series for rural connections. So what needs to be done here is first verify that this break in the time series in rural connection is justified or if it's just a wrong data manipulation. If it's justified, for example, it could be a big program on solar and systems that in 2022 delivered a lot of households with access. And in this case, also the national connection need to be corrected accordingly. But if the problem was in the time series, rural connection need to be corrected, et cetera, et cetera. So spotting errors is the important party and internal coherence. The consistent with external sources, this is very important, a check, because ensure that the data collected are in line with other comparable data from other sources. For example, the IA access to electricity data differs from the World Bank one because we use different sources and methodology, but we know that. So we can compare them knowing that there are specific differences and so being able to know if there is a difference that cannot be justified. Then we have, as you can see here, is the difference between the IA access rates with the World Bank access rate for different African countries that are represented by the blue dots. And then we have the plausibility checks that ensure that the final data are in line with reality. So this can range from checking obvious things like a simple like a reality checks, like is the access rates higher than 0% and lower than 100% or can be more nuanced evaluations that rely on the data checker or the expert knowledge of the sectors. Examples are confirmed that there are non-negative values in access rate. As I said, we verify that access rates are no higher than 100% but also that reported, for example, reported zero star reality if we have zero connections in, in for example, in solar system, is this true or is just that we don't have the data? In the case we don't have the data instead of reporting zero, we can have a qualifier that tells us that these data are not available or a metadata or the documentation that go with the data that will tell these data doesn't include standalone systems that are not available at the moment. For example, this will help a lot data users. And that's to verify the data falls in expected ranges. For example, if there has been a very strong electrification program in the latest year, are we seeing the access rate increasing? And if not, why is it correct? Is the program that is not effective or is the data that is not effectively tracking access? So once we do the verification, once we spot mistakes, we always go back to the raw data source and contacts and we try to work with the data providers to find solutions. Sometimes it's easy, sometimes it's more complicated and some mistakes might be left to be solved for next years. Some will require quick estimation for this year and more solid solution for later. But always keep track of all the mistakes, how it's been solved, et cetera, to have a good log of documentation. Then let's go to the data processing part, how we process the data. Let's see the example of calculating the access rate. We count connection of the scene here in the figure in blue. We estimate how many people are connected in each household. So we multiply, for example, the household sites by the number of connections. The best it will be done by area. If we know the household sites by a different area and we have the connection, but the same area so we can use this because as we know in some remote areas of the countries there can be a very different household sites. So this can make a difference when you calculate the access rate and then calculate the access rate by dividing the total connected population by the total population. Many countries use a basic formula, but the IA has a proposed in the guidebook a recommended formula that practically propose the principle of calculating access to electricity indicators by technology and geographical areas. The recommended formulas calculate the access rates of bottom up by location and by system and can also provide the access rate by system. You can see in this example we can see grid, mini grid and standalone system connection by different areas. For example, the east region, the west region. And then we have the total by region, the total connection in the east region, the total connection in the west region and that's the national total. The collected, once the connection will be collected by technology in the area, if we multiply them by the household sites of the area, we obtain the population with access, as you've seen before. So how many people are connected to grid in this example? Then if we divide this by the total population data that are probably coming from the national statistic office, we obtain the access rate in this region. And we can see that we can do that for each region and each area. Each area and each technology, as for example, this 80% of standalone system in the west region. Then we can go and bottom up calculates regional totals and national totals. To calculate is better not to use the national level, for example, the average outside, but since we collected the data bottom up, it's better to use directly the total population with access that we estimated bottom up before. And so we'll have, for example, in this example, 38 million. Then we can divide by the population of the national access rate. And of course, we can do this for each technology, the national level, for each technology, for each area in the country. We can also use definition of areas instead of geographical location as rural and urban, but this will require like, to have a strong definition of rural and urban and that also data providers align to this definition of rural and urban when they collect process and provide you the information on the connections. And now let's go to the last part of the data workflow that is the data dissemination. Data dissemination, as you've seen, is when we publish the data, okay? We have the first principle is the data need to be relevant when we publish accurate and timely. What does it mean? In terms of relevance, this means that the entity in charge of tracking access need to align the data production and dissemination to the evolving needs of the data users. For example, including a split of technology when it's needed. And that is ensuring that the data are reproducing reality on the ground with a good degree of accuracy and reliability. And this is very important to implement, as we've seen before, verification and revision of the data process to make sure the data we publish are in line with reality. Then we have timeliness as well, so it's very important that the data are available on time and they're up to date to make sure that the users can use the best and more recent data available on their analysis and tracking of the effectiveness of access rate. For example, you know some countries that are present today like for example, Mozambique and Rwanda are able to produce very timely data for their data users that you will be on the annual level. Then we have accessibility and clarity of the data. This is very important. If we publish very good data that are relevant to curate and timeliness, but no one's can find them because they are nowhere outside of the ministry servers or they are hidden in a website that are very difficult to access, this is a problem because no one will benefit from this very important artwork. So after creating quality data, we need to make them available and accessible to all users. For this reason, we recommend that there are clear release calendar so we know when data will be published, where will it be published and now we can access the data. For some more disaggregated data, for example, there is a need of requiring a profile to be subscribed or the access be restricted only to specific entities or people, but in general, it's very good to have this data and the persons, at least the more important data users to know when the data will be released and where, but also to the most large public to have some data available for free and easily that are maybe more aggregated. Data can be and must be released in different forms to be more clear. For example, different formats like Excel databases, they can be also in reports that have graphs and figures and tables, et cetera. But in general, a combination of all these different formats is the best because they can convene for different users. There are some users that prefer or they need a report for the work, some users that prefer to have a complete database in Excel or in another formats or from a webpage, for example, that they can use for their analysis. So it's very important also to have different type of formats available. And then we have the last that is Corelence Comparability and Transparency. So the use of metadata. As you know, there is always difference how countries, regions, even companies in the countries treat data. So it's very important that we keep track of data about data, this is called metadata. This is important because people will have then a transparent record of how this data may diverge from standards, for example, or what they include, like they include all type of technologies, connection, et cetera, et cetera. Where the data sources, what are the sources are from surveys from a combination of surveys and supply data, only from supply data. We have good examples in the guidebook that you can look at but publishing together with the data itself. Documentation on how the data is being calculated, what are the methodology and definition, supply data can help a lot of data users. So release data must be accessible, relevant and transparent. But now let's go back to some questions for you and let's try to go back. Sorry, I think we lost the screen sharing but I'll start it soon. Let's start again, a main team. So let's go to main team.com.com but now we need to use a different code. I leave you the time, so the code here is 49111595 that you can see at the top of the shared screen now. I leave you the time to connect again and then answer. The question is what is for you in your country the main barrier that you encounter for tracking access to literacy? Again, you can write down the things that we will see the different answers that we receive. So the code again is 49111595 and the site is menti.com. Can go with your phone or not but I see no one is answering now. Maybe when we wait for people start answering that we see, I can look at some of the questions and answer where we have had in the question and answer. Okay, there is a question on mini grids saying that since they have distribution lines should this be considered as grid connection? The answer is no by definition since we are defining here grid and mini grid separately. Since mini grid often are separated for the national central grid that has include transmission lines and we will count as grid connection only the ones that are part of this bigger national transmission grid. For mini grids, we can then mini grid is if they have only distribution lines that are very short distribution lines. Normally there is a generation unit close to a village and the lines just go to the different households in the village. There are no longer transmission lines. So this is why we separate them. Sometimes when the grid arrive connects to the mini grid. So the challenge arrives there, do we count as grid or as mini grid? And as I tried to show before is that in this case we need to track the connection as coming from the entity or the company that is selling officially the electricity. So it's still the mini grid operator is telling the electricity. We can say this connection is mini grid but this is the grid taking over the distribution business. We can count this as grid connection. However, it's important to track also how many mini grids are connected to the grid because this is very key. There is also a question on how the IA can help in improving access to electricity tracking from our colleagues in Zambia. The thing that we can do actually is more on the training side, we provide many trainings and also not only this workshop but we'll have some more in-depth training happening this year and also again, there is a training happening in Italy in July. We have a small number of participants that can participate but we ask you to please send your application because we'll consider all of you and we'll give priority to the motivation of people to the relevant of their position if they really work on tracking access, et cetera. And also we'll need to give priority to the countries and people that are cooperating the most with us on the data collection process. And this will be, for example, a training that we are looking for sponsorship and the people that come will be sponsored to join this training. This is an occasion that, of course, will be for selected participants because it's live but there will be many more things that we'll continue to organize as after the summer, both online and live as times come. And that's where we can help your country rise the message of the importance of access to electricity data and so include it, for example, on energy transition planning. And once energy transition planning includes required resources for data development, this means that you can rise for donors, for example, or for the government to put your, to give you more resources and budgets to do that. So I think like many people already answer, we see that reliability is one of the most difficult things to track. Corruption, harmonization of data is a difficult thing. Then we see the reliability of the data. So this requires more data checking and more work. I see many of these problems can be solved in a medium term once we establish this data working group with all stakeholders and we work with them periodically to overcome all these problems like reliability of data, et cetera. Institutional setup, these are so very important, but again, the first, let's say, fundamental that we see for a data strategy that says to put together this working group can also solve this because once you start speaking with all the entities working in the countries where you can achieve all these availability data, analyze the data, yeah, this analyze the data means there is no time probably to analyze the data. It is time consumers and these come back to the reporting part, to the resources part, sorry, not the reporting part, sorry. I got distracted with your question. This is very interesting, Cloud of Word, I'm very happy, we have this, I will take a picture. Location as well, location is very important because knowing where the connection or the people that are getting access are is key and fundamental for planning and many times we don't know that, not the facilities, et cetera. Then let's see another question, I think a last slide probably, on the workshop itself. We didn't finish with show quickly as some Excel templates after this, but we'd like to know from you if the workshop was useful, what additional content you would like to see in this kind of workshops or a future workshop, et cetera. Please, I let you answer this when we start seeing all your feedback that will be taken of course into account for this. Feel free to have a positive feedback and negative feedback and what you'll like to cover on our next workshop, et cetera. Positive feedbacks are very appreciated, but also constructive feedback, of course, because you want to make the most impact on this. There is an interesting question from Kenya for our colleagues about the fact that Kenya already have a national ratification strategies. How can the data part can be integrated in an existing policy? That's unfortunately, I'm not an expert on how this could work, but if it's not possible to include it in the existing strategy, it's very important that the people that are in charge of implementing the strategies rise the importance of tracking and establishing a link to the ratification strategy, data strategy that can serve as a backbone to provide the information to the ratification strategies to track, to monitor, and to also plan again once we have seen after some years how trends have evolved if we need to change direction, et cetera. So it's very important we rise this up but fortunately, I'm not able to tell you how you can integrate this in an existing policy. But maybe we can have more discussion as if we have some people from other countries that already have been able to implement some modification to an existing policy and they want to share their experience. Then I have a question on the minimal energy services from our colleagues, I think in Ethiopia that are asking from tier one solar lantern products if they are considered electrified households. So the IA propose that only the solar systems that are of 10 watt peak and above are considered as access to electricity. Solar lanterns normally are smaller than three watt peak and they can provide only one light point. So we don't consider they provide a minimum basic bundle of service that we described. So we don't include it in the access rate. However, we suggest that this is tracked separately. So you know the number of also that use solar lanterns because it's a big market can provide an information of how many households are receiving let's say more quality lighting than previously. But of course it's not providing an energy service that is life changing because that's the one light point of the solar lanterns doesn't. It can permit of course people to study at night when the light is over, et cetera. So it changed as for the quality, but still I feel and the IA proposes in this case to exclude from access rate and not electrification of course, but to count them separately separately. Thank you very much. We have already 15 comments. We will read them more offline and come back with proposals, but for example in reading, how can we track access with GIS? We'll have sessions on this in the future. One will be the one physically in Italy this summer. The sum of few will have the chance to attend. Please if you didn't do apply to the training because I'm sure you received the email for the application. If you didn't receive the information for applying, feel free to contact us and we can make sure that you have the chance to apply. But as I said, there is a very short number of people that we'll be able to attend. So we cannot unfortunately bring everyone to Italy this time, but there will be other trainings on this, even some offline, online, remotely. And we'll read all these interesting comments later. So thank you very much. Let's go now to continue our workshop. We have, if you can stay at least 20, 30 minutes more, we can finish today with our workshop. I wanted now to show the IA template for tracking access to electricity. So the IA template for tracking access to electricity as some of you knows very well, looks like this. It's an Excel file with a first page that includes instruction on how to use it and some definitions of the access the IA use. And then there are some information to fill. First thing to do since this template questionnaire includes macro is to enable content. If not, it will not work. So you'll have to click here up in the yellow line, enable content and then we do to fill the data after reading the instructions on how to fill it, to fill the first information. So you choose your countries. Yeah, I just put other English because Italy's not there. Then we write the name of the person that filled the question, Gianluca. And we can also put the institution, for example, is a very important information, even if normally this by email, so we know already who we are, but the date, oh, 23, I'm still in 22. The data source, this is important. This is not an open question, but provide you, are you taking this from distribution companies, from also service or for both of them? So let's say we take it from both of them. Then the boundaries of technology, do you include only grid connection, grid and mini-grid or grid, mini-grid and standalone? Let's say we cover all of them. And then a brief description of the definition you use, blah, blah, blah, blah. You can even put a link to, for example, metadata if you already have them, et cetera. We ask us to link to the national electrification plan or other related documentation. If you don't have one, you can just put not available. The IA is not an electrification plan. Your targets, you can put a link to the official targets or write down as, for example, our targets for electrification is 100%. By 2030 in line with the SDG7, for example. And then we ask an open question to describe what we discussed before the impact of COVID and energy crisis on electrification. After everything is filled, even if you cannot fill a portion, you can just write not available or not applicable. And then you click start. Once all the blue cells are filled, you can click start and you go to the tables to be filled. Here we see we have four tabs. One is for the main data entry. One is for the secondary data entry that is on time series access rates. One is the tool for estimating the standalone system access. And one is for just doing your calculation, copying some data in and putting links to your sheets. The first, if you read the instruction, we'll know that you can start here if you have data on connections. If you don't have a data on connection on 420, for the moment you have only the access rate directly, you can directly go in time series. But once you fill here, the time series will be filled. What you need to fill here, so the number of connections in millions of households by grid meaning grid and different solar system sizes for the different years. Then what we need to fill is the population. Normally when we give you the question for your country, we pre-fill this population with the data we have at the IA, but you can put your own population if you have different data and write in the comments, for example. Why? The comments is very important to write some information on the data you are providing. Then there is also sites. Here we let you enter national and urban also sites to be used. And then the rule of also sites is automatically calculated with population data. And this will automatically calculate after with the formula that is here, the access rate by multiplying the connection by the also sites. So we give you as a population with access and divide this by the total population data that we entered or that are already entered. And this is done by technology, by area and then at the total level. Once this is filled, as I say, will be also automatically filled in the other tab that just report more aggregated only the access rate in percentages. And if you don't have, as I say, connection data is written here as well in cell row two, you can fill directly time series with access rates. So for example, here you can say we are at 90% access rate. Okay. And normally when we send you the questionnaire directly, we pre-fill also the last year data from the IA and you will see some differences from the previous year IA data. So you can also connect on that, comment on that. Then there are some buttons that permits you to put the default values or go back to different system. Once the question is filled, we'll provide you some graphs on the access rates. For example, if we had like 80% grid, 4% mini grid, and so around systems to have 90% or 6% we'll also have some other graphs here. You see like a column of access rates by technology. And then the estimating tool for access with standalone systems, you can enter the lifetime you have for your system. That can be five, six years, three years, depending on what the information you have in the system that you provide is used. Five is a good proxy for solar arms systems. Then you choose if you have information on the backup use. For example, let's say you know that 5% of the sold solar arms systems are used for backup to the grid in your country and then you adjust to fill the sales of delivery. So how many units have been sold of these sites in 2000? Let's say we sold 1,000 units, then 5,000 in 2001, 500 only in 2002, 100,000 in 2003, and just adding different numbers. 15,000, 10,100, a lot in 2000, no, this much. As you can see below, the tool is already calculating the systems in operation, estimating them using the lifetime and the discount factors. Since we put a lifetime of five years, this means that the systems that we sold in 2000, they will not be more available in 2005. And so on, 2001, they will not be more available in 2006. And so on and so forth. So this permits you to just put the number of sales that is more easily to collect and to have a quick estimation of how many households have access to these systems. This is important because if we still keep into account system that has been provided or sold many years ago, we are probably overestimating our access rates. Now let me go quickly to the exercise that we'll send it to you in Excel form. And also we'll try to send you the solutions of either side that you can do with dummy data from with fake data to train on filling the IA questionnaire. Let me just open it because I haven't not opened it yet. And then I will share it and quickly explain it, but we'll send, as I say, the exercise. And we are planning also to record the solutions to have a step by step on how to solve it. And this will be also very helpful to filling the questioners in next iterations that we might have in the future. So, still a couple of minutes and I am coming. Here we are. So the exercise that you see here is an Excel file. There is a tab of start here where we explain exercise. You are in charge of traffic accessibility in your country. We have some data sites provided in the tab. The blue tabs, grid data, many grid data, surround systems, and you need to use this data to fill the IA questionnaire that is here in red. So for grid data, you have a free distribution companies that provides you different data. There will be some problems and some questions to answer as well. The questions that you answer will help you solve than the exercise because our question will make you reason on what are the issues that are happening here. The same thing is for mini-grid. You see, for example, here a problem is that the company fat didn't provide the units. So what we need to do, et cetera, et cetera. And there are other tricky parts there for mini-grids. Similar, you have different information. Some are coming from the energy regulators, but then when you go through the exercise, you can ask data to the mini-grid operators and have more interesting data to fill your questionnaires. But if you go through the different guiding questions, this will be easier. Similarly for solar home systems, you have a lot of data here that you will use to fill the IA questionnaires by also using the tool that is built in the questionnaire that we have seen just before. And then you have additional data that are just the population and the national outside. This exercise is at the national level. So it's simplified. But I think it includes some challenges that you might encounter tracking access to electricity. And also will make you familiarize with the IA questionnaire. And so to not only submit the data to the IA, but also use it eventually as a tool for your own access to electricity indicators calculation. I think I'll stop it here today. If you have any other question, please write them in the chat. So maybe we can have the last exchange. And then we can close it up and please keep in touch by email, by the WhatsApp group. If you're not yet part from the WhatsApp group, please send us an email. We'll provide you the link because in this group, we'll provide more information on a coming training or a coming publications from the IA, but also you can share with the other colleagues all the work you're doing, challenges you have. For example, you can ask a question, is someone already resolved the challenge of implementing a data strategy, blah, blah, blah. So I have two questions yet to answer. How to join the July 2023 activity? So to join that, we need to apply. I don't know if you received the email, but please send an email to myself or to no one and we will make sure that you have the information to apply. As I said, unfortunately, the number is limited. The slides will be shared, of course, after the presentation and that's what we'll share the video recording. We'll put this video on YouTube and you'll have the occasion to watch it again or to share it with your colleagues so they can also benefit from this discussion that we had today. We'll also share the exercise, as I said before. Thank you, Leon, for asking for the slides to be given in advance. Unfortunately, I had to finish the slides last minute because it was a bank holiday in France, so I didn't have the slides finalized before, but it's a very good comment. But anyway, we'll send the slides in PDF format together with the video, et cetera, to all participants in the coming days or whenever we are ready to do that. Thank you very much. Let's keep in touch and very happy of the big participation that you have had today. The work you're doing is very important on tracking access to electricity and please let's work together to find a better technical solution but also try to look on how we can support you trying to find the best support from your ministries and the country agencies. Thank you very much, everyone.