 Hello, nice to see you all. Oh my god. I'm amazed many people here Yeah, nice to see you. I'm a little bit nervous because it's my first word camp Europe and Yeah, it's my first big talk. So But I heard that the WordPress community is here to support and not to seeing me fail. That makes me a little bit Relaxing more. So First I would love to invite you all to our living room so assume that's your living room where you live and Now I have a question for you. So which table is longer? What do you think and? I would love to do it with you old-fashioned like and I would love to see your hands up So I will ask you now who thinks table a is longer. Please hands up Okay Who thinks? Thank you very much. Who thinks table B is longer Okay, some people say also table B, but less and who wants to say nothing Hands up. Okay, some people don't want to say anything Okay, so it's interesting because yeah, most people think Table a is longer and the aspect ratio is Three to one from table a and one point five to one for table B but And the truth is they are the same size and when I first saw this I was like in a restaurant drinking a coffee with my friends and a friend she showed me this example and she told me Oh, we will have a look. What do you think is longer? I said of course table a and then she said no they are the same size And I was like what now I cannot believe it. So I took a spoon and I measured it there And yeah, I also brought you something to measure it to prove it to you because also in this picture I think it's not that true But I think the stages is way too big to measure. Let's try you know, but it's really the same size Okay. No, I okay. I cannot do it It's it's too. It's too big. I never expected this No, but yeah, they are the same size and so The reason for it why we think this is because of the table legs and because that of the alignment So our brain is playing us a little trick here and that's also because of illusions So if you look at this picture for me, it looked also that the green one is moving a little bit and So but yeah, it's it's like it is it's just a illusion and Yeah, we are and our Perceptions can mislead us sometimes so the orange dots are also the same size and Yeah, that was also for me. Ah unbelievable and That's why I started with this for you with this is examples because sometimes it's also important in instead of just guessing to test and to measure and That's what we did and That's my topic testing instead of measuring My name is Viola Gruner. I'm from inside and a web press agency and Today I'm just going to talk to you about the goal of this presentation. So what's also my goal what you're Going out with Then also and I would love to explain you the workflow so I created with my team a workflow and I want to show you one example how to use this workflow Then some examples from my experience. So I made some growth test and I want to explain you these and Then my learnings also and the learnings of my team And finally you also will get the workflow So I even have made your download link So every one of you that you can have the slides the workflow and some examples how you can use it so yeah at the end you get this and Yeah, the goal of this presentation is How to make the right decision? So if you're going out that you know how we can you make the right decision for example and How to structure your leads also and not your leads how you structure your and decisions better and your ideas and then how to grow with low budget and This talk is For I think for everyone. So if we are having here some freelancer who is a Freelancer, maybe a few. Okay nearly no one hour. There are some okay. If you are an agency who is working for our agency Okay. Wow Most of you and who is Working for a product Okay, and nothing of these also some people Okay some okay just for interest. No, but I think every one of you can get something out here and can use this So I would love to start to talk to you about the contact form so the contact form is for a marketer or something very important and We had a problem. So the sales team told me viola We need one more form field at the contact form because we really want to structure our leads better and Yeah, they wanted to have the monthly volume inside of the contact form and To yeah structure them better and that was for me a big. Oh my god. No We cannot do this First because it's a big risk. So normally if you're adding one more and form field You're here also in university. I heard always. Okay, and the conversion rate will drop So also for example nail petal they increase the conversion rate 26 percent by removing just one form field and image scape they even and Increased also their conversion rate hundred twenty percent by removing a lot of form fields So yeah instead of guessing I said to the sales team, okay We will test this but my God feeling said no, it will not work But yeah, let's test this and it was a long conversation, but yeah, we did this But how is here the question and yeah with my self-created workflow? I created this also together with my team and We have here different steps in this workflow So first is to share and pitch our idea then to rate the idea Then to start the experiment and finally to draw a conclusion and to implement this test into the website So I would love to bring like with this example. I brought to you with the contact form I will lead you through this different processes So first it's share and pitch the idea So that's the document. I know it sounded very fancy the workflow. It's just a simple excel feed and So here we were and collecting all our ideas and we were structuring them So all the ideas are having the same structure and the same way how we were Doing them so we have consistency and You can see there are different ideas and we have the item name we have the hypothesis and the key metrics and If I would like to implement this test, I'm talking to you about and adding one more form field to this document I'm just putting it into the backlog and there you have the test the form field test and Yes, so you put it inside of the document and You can see also. Mmm. I linked this Part and it's linked to a pitch card and the pitch card is just a simple Google Docs and I will show you how we were Like or what metrics we had here So I'm now showing you the pitch card So it's divided in two different parts So we have the part before testing and we have the part after testing So it starts with the hypothesis Yeah So if you are adding a hypothesis you should write down What is going to happen and it's also good to write down a number So we were saying if we are adding one more form field our conversion rate will drop 12 and percent that means from 7% to 6.1% and We had this number also. It's out of research. Okay, it's not just by God feeling It's it was a research. So if you build your hypothesis, please try just to write it also down measurable Then we had the objective. So why we are going to do it. Yeah, because the sales team told us and They wanted to make their leads more manageable and structure them better. So you just write about the why and There's also good to sometimes have also the numbers. Normally you try to improve your conversion rate at these tests And you have also that is a sign inside of it And so here you can just take a screenshot for example if you see at a Competitor something nice you just take a screenshot you put it there because the pitch card. It's just a one-pager It shouldn't be too much work, you know, just keep it short snappy and simple and then the duration so In the pitch card are a little bit more metrics than here I'm showing you now because it would be a little bit longer and but it would be for this test 12 weeks So I added you also a link to this a B testing tool and you type in in this left left side for you and Different numbers for example the conversion rate and you are having at the moment the conversion rate You're expecting so normally increasing it and then also how many unique visitors you have and you see Yeah, it's a very high number of visitors to make it Significantly that means that we really can prove it scalable and yeah, if we are having more leads this number will not normally not change So yeah, and here we see also the duration and Yeah, that's how we did it and The next step is to rate the idea and to rate the idea we use the ice model and I called it ice model and it's also called ice model that means impact confidence and ease and It's also here in the tool So we are going from the backlog to the rater and There we had our different team members and they were also all written down and Everyone is counting every voice is counting the same same. So we were rating from zero to ten and That means for example for impact we were saying, okay How high is the impact from zero to ten zero means there is no impact at all ten means there's a very high impact so you can see everyone is allowed to put the numbers inside of it and The thing was here also that I saw that Anne-Sophie and Martin they gave both two points So they said oh my god if our conversion rate drops. It's a very very Sorry my hair It's a very high impact But normally giving a ten it was one of our first tests. So we still weren't that used to Give numbers here. Normally we never had this high numbers because you know having a ten impact means it's a very worse Test or a very good test. Yeah In the next year and say your confidence. So there we also used our God feeling. So we were saying What is my God feeling saying as it's saying more or less. So you can see also the whole team thought, okay? I think the conversion rate will drop for sure. So we were all at the same page here and Then the east and that's also important thing because often we think okay with what could we start? This is important this just have a look how easy it is. So to add one more form field It's not that hard. So it's also a very easy test so out of this you're getting the priority and If we are now going back to our backlog we see here the priority So you see different tests, but we saw oh my god, okay we need to start with this test because it's having the best number for now and Then you put it into life and In the live and part so if you run a test especially an important test and Conversion rates and so on you should take care that you are not going to run another test and What's testing the conversion rate? So a social test for example a LinkedIn test is okay, but you know tests can influence each other So take care of this And yeah, we started the experience experiment. We waited 12 weeks and we did the AB testing So that means the visitors are coming and they are getting so one part is getting coming to contact from one And a and one to contact from B. So with the monthly volume inside of it and Yeah, here we used and Hubspot and Google Optimize and There are a lot of different tools you can use for it. Just I made good experience with these tools and And the option a was seven percent so the conversion rate was seven percent as we expected So we thought it's seven percent and it was and now the option B I have now the question there. We have the one more form field what you think you think the marketing team was right Like the conversion rate was worse or you think it was the same or it was better So now I'm going to ask again for your signs Who thinks the conversion rate Was worse so like the marketing team said, okay, we have Some signs, but not that many. Okay. Who thinks it was the same hands up Okay, it's nearly the same like the first one and who thinks it was even better Hands up. Okay Let's okay here in front to people decided another thing So the interesting thing about it was the conversion rate was even better and I couldn't believe my eyes Because really it was for me No way and also for the team that The option B was the winner So but it's not significantly Better, you know, it's significantly the same but still Yeah, the winner was option B and that's also This example why I brought this example to you so often we read about something and While I was testing together with the team we saw hey often the things are Different because the target group is different, you know, because your website is so many different factors Things are sometimes different. So that's why I think you should test or why we tested a lot Yeah, and finally we are coming back to our pitch card and To draw the conclusion. So it's also important to write down the conclusions after you are running the test To have them summarized So we are putting it in closed and there you have a lot of different Test you were running so I brought you some tests here with me And you have always the conclusion with the pitch card here And yeah, so you have later access to this tool So you can decide how you want to structure them better in the future. You can use the date also you can add things and Yeah, as you wish and Yeah, we were writing down the outcome So there was no significant difference between and the two contact forms and That meant we can implement this new contact form to our website and The learnings for us as a team. It was yeah, it was different than we thought and It's also nice if you are in different markets. You could test first for one market So you are having less risk So we tested it first for the German speaking market and then we tested it for the English speaking market So I recommend you to do this as well if you can and Some side effects of the workflow It's the team spirit because ideas are getting collected of everyone You know not the head of marketing is saying you do this and you do that No, you take as a team a decision together and you you are seen in the team and Everyone is allowed to make something and often, you know, we lose Our ideas we say hey, let's do that, but we never start with it so if you write them down and you collect the ideas you make them true and Yeah, everyone every idea gets seen like I said and also the equality of the team so to taking it together, it's really great for the team spirit and You learn together. So for us We had every second week a growth hacking meeting and everyone in my team came up with our idea and with a pitch card and we did this and Yeah, we learned all the time together and now I brought you also some examples and And how we improved the conversion rate with some tests and Yeah, first of course the contact form there you can do a lot of different things So if you have a contact form and you can for example reduce and add necessary fields For example here the phone number we changed it to optional for example that can change it a little bit and Reduce and add a new form field So you can have also less You can have a expendable form field. So if you type something in It's going to open, but that can also be you know, some people get angry So it can also be bad for your conversion rate And then also you can use a testimonial. Sorry Alex. He's now sitting here I just took him as an example. It's not his quote I gave it for him and we improved also in the previous company. I worked for here our conversion rate Just a little bit 2% Yeah, but it was interesting for us to see but just at the German and French market and not at the English speaking market. So also here we see the market difference You can also Change your call to action. So for example instead of contact us and so on you can say book a free consultation and Try a little bit with this. This is really these are some simple tests and you can have and some big Change here and now I bring brought you also the best test we did with the most increasing conversion rate and that was so We were Building a landing page for one for our keyword So we were reaching where we are getting a lot of new Visitors through which keywords and we were building as specifically for this and keyword a landing page So you put in the headline for example the main keyword you're having you're putting some benefits inside of it Very simple. Okay, it was not that big paid with a lot of information just short and snappy Then we had also the contact form also very short and Then we had again Alex you sorry and then we had what our clients say and Yeah, just something the people can trust and we improved our conversion rate 18% for some keywords But not for all so please don't think okay I use this test and you'll have said our conversion rate will increase 18% no because also you need to test to to try it out and Yes, the learnings out of these are testing instead of guessing of course and And also, yeah, remember every target group is different Know your numbers. That's also something very important if you want to test significantly and Don't be afraid to fail test So I said fail because there is no failing if you test Because every test you do is a win and it's a learning So every completed test is a succeed. So success Sorry, so yes, really it was for us when we finished a test We were oh wow and we celebrated it even if the numbers weren't better or even if they were worse We were so happy that we did this test and it helped us often to make decisions and Yeah, no What should you do now? So first you should download the tool if you want and yeah, you have it for free I'm not going to measure anything. So don't be afraid that I'm collecting data there and research different Possible tests so you can look at a competitor or at websites. You really like and check it out What do they have what do they do? Then evaluate the test so even if you're a freelancer and you're not having a team You can evaluate it for yourself with the ice model and then start the test and Yeah, congratulations. You listen to my presentation Thank you very very much Yeah, thank you so much here. You can download the workflow in this little here I'm so confused because of this nice group and yeah, just download it You can reach out to me whenever you want With your questions and you can write me you can make now some questions and I also wanted to say thank you So much to all volunteers here. I'm so impressed Really, it's the sound check behind the scene the volunteers. So thank you so so much to you making this possible It's it's amazing. It's amazing job. They do they put a lot of work. I'm so thankful Yeah, and thank you Fiona, but you're not ready yet because maybe some people have questions Yes, and then we have two mics one there and one there so if you have a question, please line up at one of the mics and We also have an audience that is maybe online and watching this live stream So if you're on the what live stream and you have questions, there is a feedback button and If you put a question there if everything works with internet I will get your question and be also able to ask it Yeah, we would like to ask everybody to give feedback about this talk On the website of WordCamp 2023 there's a feedback form you can give feedback to every talk you have been We appreciate your feedback to improve WordCamp every time we have it. Okay. I see a first question over there Please go ahead. Yeah. Hi. Hi. Hi, Viola. Thank you so much for your talk It was really interesting about AB test is always interesting to me I actually have a few doubts about AB test which I want to to forward to you so you can you know Convince me about no, just kidding. No No, just that try to understand how much AB test could be effective for everyone So for example, first question is AB test can be done also for small website I have a blog that I have like 2.5 K user for each year Is it okay to do like like a AB test for this or it could be Just a two small Group of people second question it so I would start answering you the first question if this is okay so that's a very good question and Yeah, so I just can say you to have really significant proof It's Great to AB test but there are different steps, you know, so for example AB test you can imagine Yeah, like I don't know. So the first thing is you can do asking experts Collecting their data then in the next step you can also measure more, you know, it's also good and Thing to do and then AB test is really high in this pyramid So if I do you think it would be work it no because so for sick being significant I really recommend to have the data and to use this AB testing tool but there we had a very and high number of Confidence so 95% so you can even drive tests with lower ones And for example also for my master thesis I had a target group for the experiment of 220 people so it depends also a little bit on the test so you can research it but for being Significant for the website. I recommend you to have this high numbers Okay, so for Be rude, but could you do the next question? Maybe after the next person because in line behind? No, otherwise It's just I I Hope I answered your question. I was really like Can the next person in line ask the next question and you can always write me asking the questions Just contact me if you want Sorry, he kind of actually asked my question, but I wanted to know how you get into the math of the Statistically Significant numbers so if you have a smaller sample size, how are you doing the math to know what is significant? So if you are having a smaller target group, I recommend you just to research and then you can calculate it There are different formal so you can use and also with this AB testing tool I showed to you is a really, you know Valid tool so it's really high numbers But if you are having smaller numbers, it's also good to Try first to get traffic on your website and you can do this differently for example SEO SEA and so on and then you can test better But I know many websites they need to run a test for years to get significant data So I recommend you also to research it and sometimes you see a tendency. Okay, if you're a be tested with a smaller group But for being significant You need very high numbers. I hope that answered your question. Thank you too. Hello. Hello I was just wondering if it's really important to consider what in the in the Contact forms that you showed us if it's very important to notice what kind of extra field We add to the form or not so if it's a if it's a field that the Visitor would like to be asked to input or if it's something that really don't really don't add any value So for me a contact form should be important for you So, you know, if you say that's important for me That's important to measure it better to structure something so it should be for you But of course You can also if you want to make them feeling good your clients you can also try something else I hope I Think you didn't understand my question Okay, I Was just asking what kind of input field If it's important to notice what kind of input input field we are put or Remove from a contact form like if it's a necessary Information that we need to get from the visitor or if it's not something necessary that we need to get from the visitor So I think the contact form what kind of information I cannot say it depends always on The things you're working for I was asking if it needs to test Different kind. Ah, okay. No. No, I got it So it depends on you if you're happy with your conversion rate No, you don't need to to change it, you know, but if you want to improve it You could for example remove some fields. That's normally the market saying hey if you're removing fields You're making a charter. It's normally bringing you more conversions. So yeah, it depends if it depends on what you're selling Yes, it depends on what you're selling so you can research here also So it depends always for example in logistics. They have and the normal conversion rate around six To four to six percent so you can research this what's the normal and conversion rate in your market and Depend on this you can say okay. My conversion rate is really good. Are you saying we could test this, you know I hope that answered your question. I see also on the other side. We have somebody at the microphone. Ah Yes, thank you for your presentation first and the question is What was the difference between the conversion rate from country to country because you said only for Germany I think yeah, and what was for us it seems It should be different and that's the interesting thing about this you mean about the 70 18% Yes, we improved Germany. It was increased, but yeah in Germany. It was increased So that was I think might I left this company at the after this test So I didn't do the test for United States, but my this team told me afterwards. Hey viola the test was really Yeah, the conversion rate was not that good, but for another keyword. It was good, you know, so yeah, but That's yeah the answer. So it's also good to try to test the different markets To do a B test and also for those different fields Thank you. Thank you very much Thank you First of first it was an interesting presentation and it's always interesting to analyze these conversion rates like a topic of research My question is have you considered it doing in in more scientific way? So like maybe doing logistic regression as these are coral categorical categorical variables so maybe you can Like set a new hypothesis and refuted the probability level of how to refute that hypothesis like in more scientific way Not only a B split testing like another level because always a B split testing from my experience because I worked in this statistic and research field It's sometimes can lead to to erroneous results and Regression it's it's more scientific and like right way to do this. Have you considered doing in that way? Thank you Yeah, thank you very much. So because we had this amount of Group as you saw it's we had a lot of visitors It was like the tool you can check it out later the a B testing tool. It's really looking that you are having Significant Proof but as you say you can always test even better But you know if you are a B testing it shouldn't be too much workload to just run this test I think because with this a B testing we made very good experience in my past and with this kind of numbers So we saw the difference after these tests and they were not changing and also with this a B testing tool I showed you you can also matter Measure afterwards all the numbers. So yeah, that's my answer to this. I hope yeah Thank you very much. Thank you. Yeah, we just have time for two more question Yeah, yeah, it's just my question is really fast that you're majoring. Yeah, again, you are again. Nice to see you again Nice to meet you too. Always great content My question was this and it was simple. So imagine for a second you're doing An a B test. Yes You end your a B test you get your answer from the a B test You do the edits to the to the site to the service with the app or whatever to impact and to get the The update to get to have the version of the one that they have wind and You notice that it's not Doing is not performing like it was expected, but you will do at that time. I Will wait the weeks I said before so even if you see okay if the test is really bad, you know One time we said, okay, let's Stop it without waiting. You should wait normally, but normally never we did this So if it's a really important number and you say, okay, there's coming no conversion rate through it And the test should be for three months and you say after one month, okay We did this once but normally I say wait really because in my experience You need to wait to get significant data and yeah, so I recommend you to keep calm so if you do the test and at the end of the test you get a solution that if you update that you Add to the site as official one. So they will stop the a B test and you see That is that is not giving your the performance that you expected you just revert I know it never happened because we a B tested before so we knew okay It works, you know, that's the reason for testing Okay, because you are making the decision based on the numbers and then you get the results and that's the good thing You so you can sleep well and you know it will work because you a B tested it until it's Significant and that's really how it was we never removed Good test afterwards just when you are in the testing period and you say hey, it's so so so bad Yeah, but normally you should weigh the the time and you see when it's significant and if it's significant It was always afterwards the same. So yeah, thank you so much Viola I think you're welcome subtitles being that is a Trying to subscribe our things. Bye. Bye. So for the last question, please. Whoo last question Oh, so the main goal is sales. So maybe you didn't know the impact for the Leads qualification for the sales team. Yeah Yeah, and the impact that's a good question because and I also told the sales team Hey, you should measure this what impact it is having but they you know It was just for splitting the leads to the right person. So if you got a bigger company It was handed to Kevin and if it's a smaller to Luca. So, you know, it was for them So they told me we all I know we cannot give you any numbers for the impact But normally it's great to have here a number as well to say, okay In terms of lead quality, for example, you say we measure it in HubSpot How if the leads are getting better? So if you can change a lead from Lead to a client you can use different steps here And and then you can measure here also the numbers and that's also so important Like you said to have a look at the lead quality But yeah, I had 30 minutes. I couldn't talk about this as well, but good good question also If anybody has any more question, I'm sure Viola is willing to talk to you. Yes In a few moments, she will be outside and then she can answer all the question It's almost time for our game card question Thank you so much. Thank you very much. We have a small token of appreciation from working Thank you very much that you took this effort and that you can remember us when you're at home Great. Thank you. And thank you to everyone who came and supported me. Thank you very much