Subscribe
Share
Share
Embed
We follow the journey of Faith Forster as she creates an AI native tech startup & product.
Hey. Welcome back to Building Out Loud with me, Randy Silver and
Faith Forster:Faith Foster.
Randy Silver:We've been documenting the journey of Faith building this new app, but we haven't seen that much of it, and we haven't checked in on progress in a while. So, Faith, you were saying before we hit the report button, you had big news yesterday.
Faith Forster:I actually now have a live MVP, which is super, super exciting. Once I pushed it live into production, it's then been a huge process to debug it. So I've spent a lot of time going through it, just making sure everything works exactly the way it should. And also making sure it's all secure, multi tenant setup and the data's in the right place. It's been fun, but it's actually been really interesting because if you remember back in one of the previous episodes, I was saying I had really low confidence that I can actually get this to work.
Faith Forster:The agents will actually collect the feedback and show proper feedback. It sounded correct, but the links weren't working. I can pinpoint the exact moment where I'm now confident it will do that job. And that was on June on the January 12. I got an email from customer support at Replit confirming that the advice the agent had given me within Replit was true, and the way Gemini was working basically isn't supported through the direct Replit integration.
Faith Forster:And so I've had to set up a direct integration into Gemini separately, and now it's working exactly as it should. And so I'm now a lot more confident that this will actually do the job of collecting customer feedback from external science and competitor intelligence in a way that's reliable. I'm confident enough that we can get there in terms of making sure the product delivers value and does what it needs to. The process of getting feedback from other product leaders and the development cycle is so quick. But the one thing I was really worried about was that feasibility question.
Randy Silver:That's really exciting. And I think it's hilarious that you had to deal with support and a real person who hopefully knows what they're talking about rather than just asking a different version of the LLM, the same question you were asking. But you're gonna show it to us today. Right? So we're gonna take a look.
Faith Forster:So this is now the MVP. So at the moment, all it does is draw feedback from external sources, review sites or comparison sites. As part of the sort of next phase, once I get people testing this, I will then create the integrations so we can then draw feedback from internal tools as well. This is the product dashboard. This gives an overall view of business impact, customer value, and at the moment it's empty, but you'll also be able to add custom product metrics once you can connect your internal tools.
Faith Forster:We've got a summary of our competitors and adjacent products and a profile of each of those competitors. I still need to debug this bit. We also have some updates on any recent changes that have been announced to our competitors so we can track that. This is not part of the MVP. I've put it in as a placeholder.
Faith Forster:I'm going have a section where probably bounce around with feedback and competitive information and work out what that means from a strategic opportunity perspective. What we do have is the roadmap across product teams. You can see both the timeline view and the Kanban view here as well. So where I've really focused is on two things that we talked about as part of the scope of what the MVP should be. One is the core workflow at a product team level and making sure that works.
Faith Forster:And then how that feeds into the data in the product dashboard to give some of the insights that as product leaders, we should really be monitoring, like return on investment on product work. Actually in reality, it's just super hard to get the data and monitor that properly. So I've used Dext here as my demo product. Within settings, we've got the product teams. We've also set up the goals.
Faith Forster:So we've got some goal themes. We've also set up the metrics that will be commonly used across the overall product dashboard, as well as product teams for their targets. And we've set the goal period as well. We then have our business goals. It uses those themes.
Faith Forster:You then put in your goal that you've agreed for that year, and you put in a target as well using one of those metrics that we've set up. Within team goals, the team can put their North Star metric, what their goal is, as well as their target, and then we can link it back to the business goal as well so that we can see that displayed here on the product dashboard. At this point, it's just showing the planned value realisation. So once they deliver the work, and when we think we're going to start realising value from that work and how that tracks. And so overall for the product, you can see how the different teams are contributing to your business goals and when that will be realised, whether or not you're doing enough to hit your targets, and if you're on track to hit it by the end of the year.
Faith Forster:Within the product teams, the way we get to that understanding is they have their own dashboard where they can see the value realisation broken down by initiatives. You can also see sentiment and a couple of other metrics. They can decide which of their competitors are relevant to their product scope. Where there was real heavy work is done for the product team, we have agents that collect feedback from review and comparison sites on competitors that are relevant to this team. Here, by default, it's shown them grouped as themes, but you can also see the raw feedback as well.
Faith Forster:And it has that link, the platform it's come from. You've also got the feedback on our own product here. From either of those, you can manually create an opportunity. This can be done automatically, again, by another agent or a team member can choose to create an opportunity. If a team member creates the opportunity, the agent will still gather the evidence for that in terms of competitor insights or your own product feedback.
Faith Forster:So within these opportunities, you can pop in your impact hypothesis and your goal contribution, how much you think this will contribute towards your team's target or goal. So this is obviously a hypothesis. It's a guess at this point. The team can also vote on their level of confidence. It will gather the relevant customer feedback and also some insights on how the competitors have solved for this and links through to their Help Center documentation or resources.
Faith Forster:There is another AI agent who can help come up with solution ideas but the team can also manually add those themselves and comment against those. You can favourite them and there's also some recommended actions which you can choose to have the AI agent do for you or the team manually adds those. And you can have some discussion about all of that within the team before they choose to add it to the plan. And then if we go to that team's roadmap, you can see the activities either again by Kanban or Timeline. And the idea is that these will then have an integration with Jira or alternative tool that you might be using so that this always syncs with what's happening in Jira.
Faith Forster:But you can see here the focus is very much on realising value. So we've said when we think we'll start it. We've got an indicative time investment we're willing to put in. The team should be agreeing that and with the product leader, and that then calculates when we think we can start realizing value. So it assumes it's at the end of the investment time that can be changed and you can start realizing value earlier.
Faith Forster:You can also use AI to then generate your PRD or user story based on all the context that's been gathered within the product already.
Randy Silver:That's a lot of also's for an MVP. So is this really an MVP? Is it a version one? Where do you stand on that?
Faith Forster:Yeah, it is a really good question. I think that idea of an MVP is going to change within the context of AI because I have already built so much in this and I've been working on this for, say, eight weeks. So there are features in here. This counts for the number of new items since the user last logged in. I remembered my tech team at DEX saying to me, it would take eight weeks just to build that.
Faith Forster:It's actually technically quite a difficult thing to build. And so in that context, the timeline view, I was debating with myself whether just to have the Kanban and keep it simple, or do I add that timeline view in as well? It literally took seven minutes for it to be built. And so why not?
Randy Silver:Sure. But there's the other side of that of you've built this all out. You've fallen in love with it, obviously. And I can see how you take a CPO gig at a new company, so a new job, and this is how your mind works, and it is the operating system for how you do organization. What you haven't done though is this is the minimum viable product, which is there to, in theory, prove a hypothesis, Prove that it is valuable, that's feasible, that's usable, but no one's used it yet.
Randy Silver:So what have you proven?
Faith Forster:Yeah. Yeah. Totally. I have a a list of people who will help test this. I'll start reaching out to them tomorrow to get them to play around with it and give me feedback.
Randy Silver:What's the guidance you're giving them? What's the hypothesis that you're trying to test with them?
Faith Forster:There's definitely some open questions I have around the goal setup. Does that match reality across lots of different types of organisations? Because that is something that's quite complex and quite varied. It is something that worries me about whether or not this is too simple or too complex for certain organisations or just doesn't fit their process. So, that's something that probably will change.
Faith Forster:The Product Vision section I really want feedback from other Product Leaders and multiple Product Leaders before I even look at what that section does. The core workflow in terms of getting feedback in, understanding the option needs associated with that, using that to then work out what solutions we build, add it to your roadmap, that is the core product life cycle.
Randy Silver:The fundamental question I have about this because when we first started talking, it was about that. It was about bringing the comparisons and helping to manage ideas and rank the value of the ideas. But there's a lot more to it right now. Putting in the vision, putting in the teams, putting in the contribution to profitability and success and value, all that, which is a lot of also's to what you first described to me. So the question I've got is there is a thing here that you said you wanted to prove that you can do, you got the confirmation yesterday that it could do it yeah and it is working, it's doing what you want but it's doing a bunch of other stuff that's context and kind of makes this the question I've always had is, has this fit into the life cycle or to the governance and operational cycle of a team?
Randy Silver:Because right now, there's things that exist in other places like vision, like an integration with Jira and other things that makes this potentially the dashboard for how you manage a team. So there's quite a lot here. You're you're saying this is kind of the hub, but that wasn't the original MVP ideas. Is the information that we're finding from competitors? Is that useful?
Randy Silver:And how would you use that in your day to day operational management?
Faith Forster:Yeah, that's true. One of the early episodes we did talk about, like, just getting this live and getting people on it and testing. And I did say that, yes, I couldn't just focus on that corp gathering and use the feedback. But actually, the initial conversations, the problem validation type conversations I'd had with a number of product leaders, what came out of that was, yes, that's valuable. It could save a head count.
Faith Forster:That's nice.
Randy Silver:And
Faith Forster:actually, what really matters is making sure that the product teams, the investment we're making in the product org is delivering against the business goals, that we're delivering value. And actually, that's why product leaders exist. That's their core job to be done. And so, yes, we want to gather that feedback, but actually making sure that what the product teams are then doing with it is driving that business goal, to me, is that ultimate value proposition. But to get that data, to get that input of things that are normally really difficult for a product leader to find and monitor, like return on investment and customer sentiment, how are we tracking against our competitors, and are we actually gonna deliver our business goals in time.
Faith Forster:You need the input of that core product workflow to get that data.
Randy Silver:So what are you looking for from this round of feedback? It sounds like something that it's not something a product leader can use by themselves. They'll need other people to be inputting information into this and using it. They'll need to integrate it into operations to a degree. Are you looking for them to say, yes, we're using it.
Randy Silver:This is how we're using it with these are the things we need and the things we don't need. Are you looking for them to say, just I found it really interesting? What does success look like and what is a good conclusion from this round of feedback?
Faith Forster:It's the step before. So I will get product leaders and some very senior product managers that I know just to play with it themselves and give me feedback on how it works. Does it work the way they wanted it to? Is it what they expected? And how would it fit with how their organisation operates?
Faith Forster:So for the moment, I just want that feedback. And then I'll use that to make sure that it is working the right way and has the appropriate level of flexibility. Once that's done, I'll hopefully also get agreement from some of them at least to then start using it with their teams as an alpha once those updates are done. So I think there's still one more round of revision before we actually get a whole product team or product organization using it.
Randy Silver:How long do you think it will take for that feedback to come in? What are you budgeting?
Faith Forster:Two weeks. I've got a hit list of people I'm about to reach out to. It's just a question of their ability.
Randy Silver:Fantastic. And so what are you gonna do while you're waiting for that feedback to to come in? What are the next steps?
Faith Forster:There's still some tidy up. I hope that I'll get in the next twenty four hours, but there might be some things that I still need to do afterwards while I'm organizing those conversations or giving them access. The next big thing to crack from this product design or build perspective will be setting up the ability to create those integrations. That's technically not easy and something I need to work through how to do that. But also from a go to market perspective, I'm going to use that feedback to revise the content on our website.
Faith Forster:I'm going to get these podcasts polished up and out and start playing that drumbeat of how AI is changing the world for product teams and what that means and why tools like this that can streamline that whole decision flow on are we building the right things are gonna be critical.
Randy Silver:Fantastic. Look forward to checking in on how it's going over the next episode and then coming back and starting to hear how feedback's coming in.