 Hey guys, I'm Abhishek and I'm a senior product manager at booking.com based in the Amsterdam office. So today I'll be speaking a little bit about my journey of applying some of the product craft skills from some of the previous domains I worked in to what I do today, which is marketing. So and hopefully there are some insights for you guys. So I believe this should be relevant for the experienced product managers who are planning to move to a new industry or also for early stage PMs who want to get a perspective around product craft while working in marketing industry. Yeah, so I'll start by giving a brief on the product journey I've had so far and I think this is quite relevant to the topic itself. So I started working as a product manager almost nine years ago with a fintech company in Mexico. So we are building mobile payments and mobile banking solutions. So as you can tell, there is a big component of B2C because we are building these services for the end user. So there's a component of building the checkout experiences or the payment experiences. But then there's also a large component of platform, which is about integrating with the banks, integrating with the payment service providers or integrating with the telco partners. So I operated primarily in the B2C and platform space. Later on, I moved back to India, which is my home country and started working for a healthcare company, but again, focusing primarily on printech. So I was building a marketplace payments. I was building the insurance services there is still part of it. Again, similar to the previous experience, there is a component of B2C, which is again around payment experiences, how people pay for, let's say medicines online, how do they pay when they book an appointment for the doctor and so on. And then there's again, huge component of platform, which is integrating with the services, integrating with the insurance providers and so on. And four years later, I moved to Netherlands. And I've been working here since almost four years. And as you can see, things changed quite a bit since then. So right now, I lead the product stack for performance marketing, primarily for the search advertisement business. So the general term that we use for this is at tech, which actually encompasses all the different marketing channels, but we use the term at tech. People also call it marketing technology or mark tech. So either way, essentially, what we are doing here is that we are building technology that support and scale the marketing activities that your business or the company wants to do to achieve certain business outcomes. So that's the core problem here. And then depending on what scale you are operating, there can be a component of machine learning, which is basically how do you use machine learning to optimize for the content you want for your ads? How do you target your customers better? How do you optimize for the budgets, let's say? So this is the domain that I'm currently operating in, at tech and machine learning, which is very different from what I've done before B2C and platform. And that's what I'm going to cover a little bit today. So yeah, so let's touch upon the agenda for today. So I will be broadly talking about two areas. First, we will try to define a broad landscape for this domain, the tech domain itself. We'll talk about what the industry we are in, what are the typical business goals for this industry, and we'll try to get a sense of product stack. And then we will go on to analyze three core craft areas. So we'll talk about product discovery, we'll talk about product strategy, and we'll talk about people management and how that changed when you move, let's say, from outside marketing to within marketing. There could be other craft areas, which are critical for a product manager, but I've chosen these three because I feel these are the three that are most impacted specifically when you move in marketing. For example, there is always a big component of product delivery or product execution, which can change. But in my experience, I felt the learning there was not that high. It's basically how you engage with your development team and how you build the process around it. So yeah, we'll limit our discussion to these three. Yeah, so first about the industry itself, right? So it's like I said, it's called advertising technology. What we are essentially doing here is to scale marketing activities for particular business outcome. You see there are different channels within this whole domain. So organic marketing is always very critical, and there can be a big component of tech in that, how you enable that. Then you have search advertising, which is primarily what I'm doing, brand marketing, email marketing, and then there are different channels within social media marketing like Facebook, Instagram, and so on. So we will not be talking about the specific product strategy for each of these channels, rather we'll take a more general approach and try to build a general framework on how do you approach either of these, right? So how do you adopt the product mindset needed to target any of these channels without going into the specific customizations of these channels? Yeah, so the key product goals, right? So if you take, let's say a B2C product in mind, some of the goals could be around increasing the number of transactions, let's say if you're an e-commerce product, or maybe increasing the NPS score, or improving the customer satisfaction, and so on. But in a marketing universe, typically you see there are goals around how do you drive growth, which is how do you get more customers? Then there could be specific goals around retention. If your company wants you to focus on marketing activities to get existing users to transact more, or maybe engage with the product more, then there could be specific goals around revenue. So this is beyond just getting more customers or getting the customers to do more transactions. This is about getting the customers over food, generate positive return on investment. So you bring in more revenue for the company by using marketing channel. And then there could be specific strategy goal, right? So how do you use marketing channels to differentiate yourself from the competition? How do you use it to launch for a new market or differentiate yourself into a new market? And there could be platform specific goals, right? If you want to drive that option of your app or get more customers to book through app and so on. So this is basically, this could be a very high impact area for the company, because you can operate into these different areas of business outcomes. So let's speak about a product stack, right? So once we have a good understanding of which industry we are operating in and what are the boundary of business outcomes that you can operate in, as a product manager, your job is to identify the product stack, what is the ownership that you could have? What is that you can influence in terms of product? Because compared to a B2C product, you might not be able to influence everything, right? You need to operate within the boundaries of your product. So you need to draw a rough idea or rough boundary for what those products stack look like. So this is how I see marketing technology and Act Tech in general. There is a base layer of data, right? So and this is, let's say, even if you're launching a B2C product, you need to track the metrics, right? So you need to get the data in. In similar fashion, once you're running marketing campaigns, you need to be able to measure those campaigns, right? You need to be able to measure the performance and the success. And if you're doing that at scale, you need to develop skills around how do you work with data engineers, how do you work with architects, right? So how the storage of data works, how the governance of the data works, because in a privacy-centric world, data governance is really critical. So all that can be really like a domain in itself. But definitely the base layer for a product stack for any marketing technology product. Then comes the reporting and tooling, right? So imagine you have all this data and you have figured out a way to successfully store and manage this data by leveraging your engineering team. You need to be able to provide access to this data to your users, internal users who are marketers who are actually implementing those campaigns, right? So and you will basically user-friendly access for this data to whoever is going to use it. So that's where the reporting and tooling aspect of the products that comes in play. And this is where you can potentially apply some of your previous skills around UI and UX. And you can really shine in here, right? So this should come as a second nature if you have done a lot of B2C work or done a lot of platform work. How do you basically build experiences for the users who are then taking some decisions using that data? Now, depending on what scale you are operating it, this could be limited to just these two where you're just leveraging data and building some tools and reporting around it. But it can also go beyond that, right? So for example, how do you automate some of the core user actions to run any marketing chart, right? Which is about creation of new campaigns, which is around optimizing for particular languages, optimizing for particular geographies, getting access to some of the insights, right? So then you can really drive some of your product thinking around automation. And lastly, production, which is where the ML part comes in play. So how do you basically leverage the data that you have to make some predictive analysis? So let's say if you want to get an understanding of potential return for a particular set of customers, particular set of traffic. In here, again, as a product manager transitioning into marketing, you need to develop certain skills on how do you work with machine learning engineers, right? And this is a really good opportunity for anybody who probably does not have that experience. And when they move to marketing, they can potentially earn that experience, right? So in here, you basically get an opportunity to define a business outcome or define a business problem using the capabilities that you have at your disposal. So this is broadly how a product stack is shaped up for any marketing channel. It could vary a little bit if you're, let's say, working for brand, then you also have a lot of huge component of media buying and creatives working with partners in a different way. But by and large, you can see how this spans out from the base layer data to the advanced level of predictive analysis. Cool. So keeping that in mind, I'll try to touch upon three craft areas and see potentially try to draw a comparative analysis, right, between domain. So yeah, so the first thing, like I guess the starting point for any product manager when you move into a different industry is to develop an understanding of product discovery, which is to understand who your user is, how do you gather feedback, how do you gather new insights, and what is the process of iteration, right? Even if it's not there, you have to develop some framework from scratch. So see if in a typical B2C framework, you start out with a business problem right in a particular industry. So let's say you want to solve problem in travel industry. So you go out into your market research, you try to understand the feasibility of that problem, size of that market, the competition, and so on. And then you take a step further and understand different user demographics, different user types, and probably zero down on one user segment. So let's say you want to focus on solo travelers, right? And that's where your product building stage comes in. Once you understand the pain points of solo travelers, you build an MVP, you define the success metrics, you develop some sort of feedback loop, so you understand how those users are using your product. And then finally, you develop a framework to iterate on it. If you want to scale it or tone it down or take a different direction. So that's a standard lifecycle for product discovery. Let's see how that changes, right, for tech. So in terms of market research, there is very limited scope of direct market surgery, right? Because you are primarily dependent on the ad platforms you're working with. So at an advertiser, you need a platform like Google or a Microsoft or a Facebook to run your ads. So the research or the road mapping that you do comes from those partners, right, on what those partners are building, what are the capabilities that they're offering. So it becomes really crucial or as a stepping stone for you to build those relationships with that partners, right? In a similar way, you might be engaging with some research agencies to do the research. You can probably transfer some of those skills, but in a way, this is about engaging as a product manager, taking the lead and influencing the planning of your advertising platforms. User segment analysis, again, in marketing universe, there might not be a direct user level targeting always, right? Again, it's a privacy centric universe. So a lot of the analysis you do is at a data level, anonymize data level, not always at a user level. You understand the user from the perspective of your product. Let's say if I'm doing marketing for travel, I do understand the broad landscape of who my travel, who the traveler is, but may not always be from the marketing perspective, because not all channels are content where user is telling you what exactly their needs are, right? So there's a lot of understanding that you need to develop around the user, which becomes a challenge itself, right? So how do you leverage that, right? So rather than doing user segment analysis, you do capability analysis. So which is to say, how do I run my campaigns more efficiently to get the outcomes I need, right? So let's say if you're working with Google and you're working on text ads, what are the new capabilities that they're building that I can test and maybe try it on my customers or my potential customers? So let's say, can I extend my ads to add more information, right? Can I try and experiment with different budget strategies, right? Can I adopt different strategies for different geographies? So your discovery process should involve understanding of those market capabilities, and that's how you take a next step around building the product. So in terms of MVP, yes, again, the difference here is that in a B2C, you have a product workflow, right? So your user is taking a particular action. So let's say e-commerce product, your user is coming to book a product or like buy a product online. So you have a life cycle of that product, but in marketing, you don't necessarily have that, right? So your roadmap or your backlog will have certain projects that you carve out based on the capabilities that you want to try, right? So that's again a shift in mindset, rather than working on the same workflow and iterating on it, you are working on different projects. The thing to keep in mind is that does not necessarily translate you into a project manager. It's more about building a backlog of initiatives that eventually make up your product stack. Defining success criteria should remain the same, always the same, because you do not want to drop your product mindset, even if you're running projects, you have to have specific success criteria, specific metrics that you can measure that against. You still need to develop the feedback loops. So how do you measure those projects? How do you measure the user interaction around those products, let's say ad campaigns? And finally, iteration, the only caveat here is that you iterate with partners. Similarly, the way your research is revolved around the partners. If you test out a new capability and you want to make any changes, you have to influence the partner rather than doing directly, because again, you are dependent heavily on the platform you work on. So hopefully that gives you some sense of how the product discovery could change a little bit when you're working in marketing. Yeah, I would also like to touch upon a little bit on product strategy. So here I'm just have a few high-level insights rather than a framework. So what usually I have tried based on my understanding of on the tech stack is that the first thing first is you have to hold the foot, which is you have a default or the base set of product stack, which your users are using to generate these campaigns at scale. So your goal should be to maintain and evolve that. Other than trying a lot of new iterations, your first goal should be to make sure this is running very well because at the end of the day, you need to empower the marketers who are actually domain experts. So somebody could be an expert in a particular language or a particular type of content. So there has to be a constant effort involving that stack. And that's where you really leverage your engineering team and your skills around team management, your skill around development processes can really shine and give you an edge here. Second thing, I think I briefly touched upon in the last slide as well, that you need to develop some skill around road mapping without direct user input because not all marketing is direct content. So there's a lot of push because you're trying to capture the market, you're trying to achieve certain business goals. So you try to roadmap by leveraging your ad partners, you try to roadmap by leveraging other industry experts. So basically your roadmap process could not be entirely internal. It has to evolve into influencing people outside your company. And lastly, since your backlog is not purely product workflows, but a set of projects that you identified, you need to be able to focus on experiment velocity, because not everything might work. So if Google launches a new capabilities, it might not work for your customers. So in order to bypass that or in order to do that more successfully, you need to build a process around experiment velocity. What is the framework that your team has adopted, but you to be able to test on these things faster? What is your turnaround time if you want to test a new ad feature? So that's where the product manager skills come in. This will not purely be done if you're just letting, let's say the marketers or the techs running the show in terms of the course stack. So you have to focus heavily on driving experiment velocity and also you have to be heavily driven by metrics. Only then you can make the right suggestions to your partners. And lastly, people, I think this is probably one of the most critical aspects that I've realized in terms of the mindset shift. So as a B2C customer, the idea of a product owner really resonated with me, where once I take ownership of a product, I take ownership of the workflow or the product lifecycle end to end, which is right from understanding the customers, right from influencing the behavior, right from measuring the data. But that might not be the case when you're operating a marketing channel because you are working with, again, marketing experts who are domain experts, specialists, business leaders who understand the channel or the finance leaders who are determining some of the goals. So you have to understand your spirit of influence, right? What are the aspects that you can directly influence, right? And what are the aspects that you can't directly influence? And what are the aspects where you need to leverage other teams or empower other teams, right? So in the slide where we mentioned what products tag, you need to do that exercise for yourself or whatever channel you operate in, right? Whether you have full ownership of the data, can you actually make upgrades on the way you store data? Can you make changes in the way you do productive analysis or the content, right? So really critical to understand your spirit of influence, you cannot assume that you will own the entire lifecycle of the user or the channel or the business activity being done. Second, once you understand that, you need to own that. You need to own what you influence, right? So you need to own data into it and you need to become an evangelist for that or the other stakeholders internally you're working with. So this is one of the key learnings, I realize that not everyone would probably understand or have a high level view on the tech stack and the product stack. People might be operating into their specific spears of influence themselves in which they are operating let's say for content, maybe for images, maybe for a geography and so on. So you need to own the product stack that you have for all those users for all those activities and wherever you can't own, you need to influence, right? So let's say you're working with a marketing expert who has a who's a domain expert on a particular type of creators, right? You need to be able to influence the strategy that they adopt of the scale content, right? For example, in product stack, be leveraged to scale that or help them out. Again, if you translate this from a B2C to a mark tech perspective, as a marketer, you are only owning the incoming traffic, right? You're only owning the inbound traffic. Once and your goal is to make sure that your user sees an ad wherever it's relevant and you get the click at a positive return, but you can't really influence what happens to that user afterwards, right? So once user clicks on an ad, lands on our website, you are dependent on the other product teams to take that journey forward, right? To show them the right experience to make sure that they convert, right? So you don't always have a direct ownership on the rest of the product funnel and that's where you need to influence. And lastly, manager, like I said, being a product manager in a mark tech would be misleading for several people, especially sometimes even in the leadership level, right? So for example, you are by default subsumed under the tech team and you might be, your focus could might very easily devote into being a technology leader or being a data manager and so on. So you really need to make sure you identify your spirit of influence and you basically convince the leadership in order to make sure that you are adding the value at the right places, right? So because you will still need the tech team to maintain and grow that system, but you also need product managers or the business leaders to understand new opportunities for the same business outcome. So managing up is really critical and could be a, especially when you transition fresh into the domain, it's something that you should be careful about making sure that everyone is aware of where you can add value because you do not want, it can, there might be a case where you, over time, you are translating yourself from being a product manager into something else into some sort of marketing manager or some sort of technical manager, right? So you have to be careful about that. Yeah. So I feel these are some of the insights that I had to share with you guys today. Hope it was insightful, hope you learned something out of it. And I hope to see you guys next time at Prague School. Thank you.