Subscribe
Share
Share
Embed
Honored two years in a row as a top tech podcast by The Webby Awards, CRAFTED. is show about great products and the people who make them. Featuring incredible founders, innovators, and makers that reveal how they've built game-changing products — and how you can, too.
What trade-offs did they make? What experiments did they run? And what was the moment when they knew they were on to something BIG?
Hosted by Dan Blumberg, an entrepreneur, product leader, and public radio host with chops as both a technologist and as a public radio host. Dan has founded startups and led product releases and growth initiatives at LinkedIn, The New York Times, and as a consultant to big banks and startups. Before getting into tech, Dan produced and guest hosted WNYC's Morning Edition, the most listened to show on the country's largest NPR station.
Listen to CRAFTED. to find out what it *really* takes to build great products and companies.
Janna Bastow final episode and transcript
===
[00:00:00] Remember when companies used to do April Fool's jokes right back when it was a little bit cool and we came up with this joke product and we said, using AI, we have now developed the auto magical roadmap, you no longer need your product manager, it's just going to read your feedback and gather all the ideas you've put in there and it's just going to put your roadmap together and communicate it to your stakeholders and people wrote in saying, I love it, where can I access this?
And I'm like, no, this clearly does not exist. Fast forward almost 10 years to the date we got access to GPT 4. And within the week we had implemented our first piece of generative AI into ProdPad.
That's Jana Bastow. And between the software she's built and the conferences she's run, Jana has done a lot to level up product managers.
Janna is the co founder of ProdPad, which helps product squads create better roadmaps and deliver better outcomes for their customers. Janna is also the co founder of Mind the Product, one of [00:01:00] the biggest product conferences and communities out there. Janna has strong opinions about what good product management is.
And on this episode, we'll discuss the evolution of this hard to explain profession, and how, with the rise of generative AI, the role of product managers is changing again. So in the future, product managers aren't going to be the same project delivery people that we have been in the past. We're going to be more like curators.
Welcome to crafted a show about great products and the people who make them. I'm Dan Blumberg. I'm a product and growth leader. And on crafted, I'm here to bring you stories of founders, makers, and innovators that reveal how they've built game changing products and how you can to craft it as produced by modern product minds, where I advise companies on product discovery, growth, and experimentation, learn more and sign up for the crafted newsletter at modern product minds.
com. Crafted is sponsored by Artium, a next generation software development consultancy that combines elite human craftsmanship and artificial [00:02:00] intelligence. See how Artium can help you build your future at artium. ai.
So for the past decade, you've helped product managers build better roadmaps. And I would love to just start with what you consider to be a good roadmap. And I assume that's a Gantt chart with lots of fixed dates and deliveries. You said you were going to be asking provocative questions and you start off right on the gate with that.
Um, no, obviously, you know, a Gantt chart is a bag of crap as a roadmap. It harks back to the old school way of doing road mapping. You know, I think 10, 15 years ago, that's what people thought was a good road map. And, you know, we've been in this fight against bad road mapping for that long and we're starting to see the tide change.
Um, but really good road mapping is about making sure that you can communicate the direction that you need to be taking. The product in order to meet your product vision it's a strategic communication document [00:03:00] it's not a release plan it's not a list of all the features you're going to deliver and when you know these old school Gantt charts grew out of old.
Project plans and this need to keep a control on you know what features are going to be delivered and what day they're going to be done and how they're all tied to the I. T. project costs and working that way just ends up making the teams deliver less with more bugs and with less satisfied teams and so ultimately.
We're seeing teams take a step back and say, well, what are the problems we need to solve in the order in which we need to tackle them? And so new lean roadmaps speak to that. Much better these days. So take us back to when you decided to found ProdPad and sounds like some of these frustrations must've been at the heart of that.
What made you lean in and launch a better way of building roadmaps and communicating that vision? Ah, well, the, the [00:04:00] order of that was actually a little bit backwards. So when I first started building ProdPad originally, I just needed tools to do my job as a product manager better. And back then my way of roadmapping was Gantt charts.
I didn't know any better, right? I wasn't surrounded by lots and lots of product people and best practice and all that sort of stuff. I was just doing what I thought my bosses thought was best. And it wasn't until we started sharing those roadmaps with other people, this really early version of BroadPad.
And the first People started coming back and at first they loved it, but after a few weeks the feedback was I love being able to take my ideas and put them on the road map and you know, see where they sit in timeline, but I want to be able to take everything from this column. This first bit and just move it over by a month.
And had I you. Just listen to my customers outright. I would have just built like a multi select drag and drop feature, something to just pick up everything and move it over by a bit. Now, my assumption back then was that people would put stuff [00:05:00] on the roadmap and they would build it. I know that I didn't build everything on my roadmap, but I just figured I was just kind of a mediocre product manager and everyone else was a great product manager.
The truth was when I talked to other product managers was that no one was building. The stuff on the roadmap, right? I mean, they're building some stuff, but most people were leaving stuff behind. Everyone's roadmap was a lie. Even the really good product managers we were talking to. So we took a step back.
I'm like, if everyone is having to move everything on the roadmap, then what is the point of this roadmap? So he asked these five whys, we got down to the core and realized product managers aren't actually doing their roadmaps. They're not making roadmaps with the idea to do everything on it. They're, they're setting up roadmaps to please their bosses and then just moving everything off, pushing everything forwards.
And so we realized if we don't have these timelines, This timeline at the top moving forward, what could a roadmap look like? And that's where we came to this concept of the three buckets, the now, the next and later originally it was called current near term future. [00:06:00] Can you fast forward to today and share more of what you started with, with roadmaps and evolved into this now next later framework, what, what, what does prod pad do today now that it's been what a decade or so, right?
Since you launched. Yeah. I mean, when we first launched that now next later roadmap, the original page that it was on was called, you know. The roadmap page. And we weren't even sure that we could call it a roadmap, right? Because it wasn't, it didn't look like a roadmap, a roadmap to us looked like a Gantt chart.
And you know, what we've really done is we've reframed the roadmap. We've reshaped it into something completely different. And over the years it has taken on different shapes or, um, we've, we fleshed it out with new capabilities so that you can do more with it. So, you know, we've, we've done that. linked it to your OKRs.
So it's an objective focused roadmap. So you're not just saying, here are the problems we're looking to solve, but here's the why, right? We're solving this problem because it's going to reach this, help us reach this objective for the business. So what's going to help us reach this objective? These things on a roadmap.[00:07:00]
gives you that opposite view, which is really helpful. Um, but you can also expand it down. If you're saying, we're solving this problem in order to reach this objective, well, how are we going to do that? And you can break it down into the ideas. Or what I like to think about as, um, the experiments, the things that you might try in order to solve that problem.
And I like to think of them as experiments, because an experiment is a really freeing term, right? An experiment could mean things like, let's build this new feature. Building a new feature could be a way to solve the problem and reach that goal, but it could be things like changing a feature or even removing a feature.
A lot of teams don't think about how they could take features away in order to do stuff. They think about the roadmap as a list of features to go add, but also an experiment doesn't even have to be feature related. An experiment could be, let's change the pricing, let's change the packaging, let's change the proposition.
So that way the roadmap isn't just a list of things that the development team needs to code [00:08:00] next to add to the product, which just makes the product heavier and heavier and heavier. And instead turns it around into, well, here are the problems we need to solve as a team in order to reach our vision and why we're doing it in order to reach our objectives and how we might as a whole team think about using our resources to solve those problems.
It's a much more holistic view at how you can use product management to solve problems as a company. That's a big shift to go from list of things to do and instead prioritize problems to solve. The analogy that I sort of like is, it's less of a map and more of like a Google map, where you know, where there's, we know the destination, we don't know quite how to get there.
You know, like there may be traffic that may come up or road closure, and we're going to go around that, but we, we have a clear sense of where we're going. And how we get there, you know, that's the thing where, if you want to, you know, really practice your craft, you hope to be on a team that's empowered enough that that has that problem to solve.
And you're not given a solution to implement. What are you seeing in terms of teams and their freedom to chart that [00:09:00] path? Yeah, I mean, this is why tools can be so empowering. Because what we've designed here is a tool that actually helps you become a better product manager. It helps you become a better product team.
I mean, if you give somebody a blank page, a completely blank canvas. There's a thousand, a million different ways they can go use it. And many of them are going to be the wrong way to use it, not particularly helpful ways. If you give somebody a template, a clear guidance on a way to do something, a better way of doing something, they're less likely to fall off the path and go the wrong direction and more likely to follow, uh, good healthy habits.
Good, healthy processes that are going to help them ask the right questions. And so we've designed prod pad to be like your coach, a little bit of a sidekick to sit with you and say, make sure that you are putting things that are at the problem level and not the feature level, think about things in terms of hypotheses and talk about your target outcomes and your actual outcomes.
Connect things back to your objectives, break [00:10:00] things down into multiple different experiments. Don't just put features on your roadmap. It's. It's pretty difficult to just put features on a ProdPad roadmap and assign dates to it. The template makes it easy to create a lean roadmap that guides you in the way that I was talking about.
And so quite naturally, teams gravitate towards working in this way. And it's not a hard way to work. It's quite a natural way to work once you actually do it. Think about it. You know, some people ask me for feedback on how do I convince my boss to move from here to here. They're going to flip their lid.
And honestly, asking for forgiveness is easier than permission. Most people are able to just get into Prodpad, start a trial, set up an easy first starter template, and then just bring it to the meeting. If you just show them that you've made a better roadmap, they'd mostly just get on board. You don't get some screaming boss being like, where's my old roadmap?
They're like, oh, okay, cool. Well, talk to me about why this problem is more important than this problem. And talk to me about why you think it's connected to this objective and not this one. [00:11:00] Great. Work from that as a base, as opposed to what you had last year. See, you founded Mind the Product also around a decade ago.
I want to tie what we were just talking about, the state of product management to Mind the Product and the kinds of talks that would be featured on the main stage over the years, and I'm curious how that's changed in the last decade, like product management, when you and I first met, which I think it was in like 20, 13 or 14 or so.
I was a pretty junior product manager at the New York Times at the time. Product was a thing that not a lot of people were sort of aware of. It was hard to explain. It still kind of is hard to explain, but at least people have like heard the phrase product manager. I'm curious, like as product has grown and awareness of who we are and what we do, what are the hot topics right now?
And like, where were we like a decade ish ago when mind and where do you think we are now in terms of like, What new product folks, mid career folks, like, what are they most eager to level up on? Such a good question, because what we could really see was the growing up [00:12:00] of product management over the years.
So, when we first started Product Tank, which was the precursor to Mind the Product, the conversation In the room was quite literally stuff like, what is a product manager? Am I a product manager? Is it different than a project manager? By the time we started the mind, the products conference, which the first one was in 2012, so just over 10 years ago now, the conversations were inspirational, but some were quite practical, like how to prototype.
In various spaces, right? How to prototype in a hardware space or, uh, how to think about some of these more practical product concepts. But I remember looking back on this a couple years ago and realizing that nobody talked about things like psychological safety or product culture or product leadership.
Until like five, six years ago. Right. So the conversations now are about how to grow into leadership positions, how to manage a team, [00:13:00] you know, the world has changed. It's become much more difficult to manage a team. You know, back then the conversations were around, how do I manage my product? And now it's like.
Great. I've got a product team of 200 people. How do I figure out who sits in which roles and how do I manage those types of people? And how do I grow in this continuously? And how do I make sure that people feel safe speaking up and we're continuously learning and outperforming our competitors and attracting the right people to our team.
There's a completely different sort of level. Up, that's needed. So you don't just need the practical of how to do good product management. That's expected, but it's also about how to create the right environment for good product management to happen. I know you sold mine, the product a few years ago. Uh, if you were still running it today, I imagine there'd be talks on, on generative AI and, and, and such topics.
It is sort of the topic of the year. How do you think, and what are you working on in product pad ways that generative AI is going to change what it means to, you know, create roadmaps, create backlogs, do the job of a [00:14:00] product manager. Remember back when companies used to do April Fool's jokes, right? Back when it was a little bit cool, and we don't do it now because it's become infinitely uncool and just a bit dangerous.
But it was April 1st, 2013, we came up with this Joke, product, far future thinking. And we said, using AI and natural language processing. We have now developed the auto magical roadmap, right? You no longer need your product manager and prod pad to make your roadmap. It's just going to read your feedback from your customers and gather all the ideas you've put in there.
And it's just going to put your roadmap together and communicate it to your stakeholders. And we had joke copy that it had written saying, you know, your, your ideas made it to the roadmap. Congrats. Your idea has not made it to the roadmap and had some sarcastic copy as to why it hadn't made it. And we published this, thinking this is about as, you know, clearly a joke as Google's toilet paper internet thing.
And people wrote in saying, I love it. Where can I access [00:15:00] this? And I'm like, no, this clearly does not exist, right? Um, so we had to update it by before the end of the day, being like, this is a joke. This is obviously an April Fool's joke. We don't have this. But hey, you can actually get, you know, ProdPad's full on roadmap.
Here's what you can do with it. Anyways, fast forward 10 years. It was almost 10 years to the date that we got access to GPT 4 API. And so within the week, we had implemented our first piece of generative AI into ProdPad. And the first piece was doing things like. Let's remove some of the grunt work from people.
We know that product managers, um, often have blank page syndrome. You have an idea. It's probably written on a post it note, or it's a scrap of an idea that you've written into ProdPad. Let's take that gist of an idea and have it write out prompts for the rest of the idea, right? So it can help. Describe it.
It can help give you ideas as to how you might, um, come up with target outcomes. It'll outline, you know, the problem you might solve. It'll help flesh out the description, but we also gave it a little bit of an opinion. We wanted it to prompt you on [00:16:00] things like, well, what risks and challenges you might run into that you could skip that stuff.
Yeah. Just, you know, Delete it from your idea, but we did get it. People thinking about, you know, whether there are safety concerns or privacy concerns or other things like this with the ideas that are putting in or, you know, user stories came out the following week, product managers hate writing user stories.
We saw a huge uptake on this, like increases in usage by hundreds of percentage points. points, people generating user stories because they are bland to write. And what this will do will actually brainstorm the different user stories. So you can just say, here's all the user stories you might think of, select the ones that are actually relevant to this particular thing, hit go, and then send those onto development to go get built.
Generating key results from the objectives. And we again, gave it an opinion, so it didn't just generate any key results, but they were actually in a more of the leading outcome focused key results, giving you better templates to work from, but we didn't just want something that was just generative. So we also started applying it to help.[00:17:00]
Be a little bit more of a sidekick. We always envisioned ProdPad being like a bit of a coach and this helped us reach that vision. And so what we started doing, one of the first implementations was you could take it to say, well, I've written this idea. Is this idea any good? And it would compare it against your vision in ProdPad and tell you whether it's aligned.
One of the first ways that we knew that this is going to be really powerful is when we took our own roadmap and ran it through, this is just in the pre production. This is in the. That using test data before we even put this into the app, we just used the underlying prompts and tech to test this, whether this work.
We said, well, let's have it judge our own roadmap and give us feedback on it. We had it judge our roadmap. And it came back and said, yeah, your roadmap is pretty well aligned with your strategy and with what you're trying to do here. But you know, this one, this one, this one, well aligned this one here.
Can't really see how it's aligned. What's going on here? And that one I could read right and say actually I could see why it's not really well aligned. If I give it a little bit more context, you know, it'll [00:18:00] help explain as to why we're actually doing this thing. That helped. And then this other one, it said it's not very well aligned.
I looked at it, went, it's right. That shouldn't be on a roadmap. So we were able to make a strategic decision to remove something from our roadmap based on something that took, uh, you know, it's implemented as a one click thing now that you can just use to sense check as to whether you're on the right path or not.
So these are all things that we've done in the past few months using AI. And we've got a lot more under the hood right now as well in various stages of beta or alpha. That's very cool. How else are you using generative AI tools now? And how do you envision product managers, product squads, more broadly speaking, we'll be using them, you know, now or in the near future.
The next stage, I think, is that not just product managers, knowledge workers, all of us, we're going to have AI sidekicks, right? And I don't think it's going to be like one AI sidekick, like a Watson following us around, that we, you know, tell it to do things. We're going to have one per [00:19:00] job, most likely, right?
And, you know, for the product management job type thing, we'll be able to just ask it to do stuff, right? Write this user story and, you know, help me come up with some, you know, ideas. Key results and, you know, grab me that report. Can you tell me what's on my road map right now and tell me, you know, how much feedback I've got?
And can you summarize Dan's feedback from last week? You know, this is all stuff that the product manager has to do, uh, right now in their job. People come to the product manager and they're expected to have all this knowledge. So we've built that. We have an AI chatbot that we have built that you can talk to, and it has knowledge of all of your ProdPad.
It also has knowledge of the same stuff that GPT does. You can ask it for product knowledge, but we've primed it on good product management practices. And it'll give you advice, and we've trained it to be inquisitive. So if you talk to it about a tricky stakeholder, it'll ask follow up questions like a coach would.
So we're calling it the AI product coach. So it knows everything that's going on in your system and I expect that over time we're going to have an [00:20:00] AI tool that just knows everything that you know all your documents all your stuff that you've got going on and you'll just be able to talk to it will do a lot of this hard part of your job that you don't even realize is hard but just weighs on you because you've got to keep it in your own head like product managers are miracle workers sometime the amount of stuff that they know And the amount of people who come to them with questions all the time, and they just somehow need to know both the breadth and the depth at all times.
And this will take some of that weight off because the AI will know a lot of this stuff. It'll know the nuances about what feature was launched and when and how well it did. And you know, what's coming up on the roadmap and what that has to do with that strategic initiative that somebody is asking about over there.
Yeah. And yes, it can pull you a report on it. Cool. Like give us some time and we'll get there. There's, there's always been this. As new tools come into play, it is for decades, right? It's always like, oh, it'll make folks more productive or they can take more time off and it never ends up working that way.
You just end up taking on more stuff. So I'm [00:21:00] curious, what do you hope it frees them up to do more of? We were having a laugh about this at Mind the Product in San Francisco. A group of us were sitting around about, you know, where are we seeing the future go? GPT stuff was like fresh and new. We were all super excited about what it was going to do.
And somebody likened it to the dishwasher. Right. What the dishwasher was going to do for the housewife in the fifties is going to free up the time. We're supposed to be sitting there with our feet up now, right? That's not true. We all have jobs and we're all working them because we can't afford anything.
And so the joke was made and I love this was that product managers are going to have all these AI tools and we're going to have more time. To go do stuff like spending time with customers, and I pointed out, I'm like, is it that we're going to spend more time with customers? Or is it that we have been carefully avoiding hanging out with customers this whole time?
We're going to find out the stuff to do? Is it that we do want to see the customers and get out of the thing? Or are we just going to find other stuff to go to tweak the AI model or something like that? I hope that it's finally going to give us time to get out there and spend time with the customers and ask those [00:22:00] questions and learn more.
I think that's where Transcription Real product management needs to go. But of course, I think, you know, people have been avoiding customers because they're saying, Oh, I'm spending all my time writing specs. I've got no time. I'm nurturing my backlog. I've got to write that report. That's why I'm not on with the customers.
Um, those excuses are going to fall away and soon it'll be your only job is to get out there and find out what the customers want because everything else is taken. And this is where I actually think product management is really going to get to. I mean, ultimately the job of a product manager. is to find out what is technically feasible and what is valuable to the market and to the business and what is desirable to the customers.
When it, um, comes down to it, AI can solve a lot of that. I mean, AI tends toward the norm, the mediocre, the middle, right? Which is great for a lot of things, right? When you're building something technical, you don't want it to be outside the norm. You want it to be just like the last thing, because it mostly [00:23:00] worked.
So it's going to be able to build things. and build it without any surprises. Bug free. You don't want surprises. An app that has a login system. I don't want the login system to have surprises. I don't want it to be a new login system. I want it to be the exact same as everybody else's so they can just log in.
The thing that makes that app stand out isn't the fact it's got a slick login. It has a functional login that has no surprises. Everyone can use it. It's something that that app does. That adds value. So I've got to define what that thing is. Now, I don't know what that thing is, right? Like for prod pad, it has its own unique value points for, you know, your company out there, whoever's listing, you have your own unique value points and you've got to figure out what those things are in the future.
All the hard parts of building all the apps, how many times have you built a login page or a settings page or filters or all this other stuff that's going to be built for you? No code tools and AI will build that all in a jiffy and then it's going to fall down to you talking to customers, [00:24:00] talking to the market, understanding trends to figure out what that 1 percent that 2 percent whatever that percent is that makes your app.
Magical and special that people will pay for and designing that and by designing I don't necessarily mean Using the same tools and code that we do today It might be designing with your voice and having the machine do it but it will be coming up with and crafting that based on what you learn from Real people, someone's going to have that job.
So in the future, product managers, aren't going to be the same project delivery people that we have been in the past. And we're already moving away from that. We know that trend is happening. We're going to be more like curators. I love it. Jenna, thank you so much. I love peering into the future and, uh, I'm excited for all the ops people and the AI tools to let us just sort of kick back.
That sounds great. Yeah. Likewise.
That's Janna Bastow. This is Crafted. I'm Dan Blumberg, and Crafted is produced by Modern Product Minds, where I advise companies on product [00:25:00] discovery, growth, and experimentation. Learn more and sign up for the Crafted newsletter at ModernProductMinds. com. Crafted is sponsored by Artium, a next generation software development consultancy that combines elite human craftsmanship and artificial intelligence.
See how Artium can help you build your future at artium. ai. And please, if you've made it this far, put it on your roadmap. In fact, put it in the now column of your roadmap to share crafted with a friend. Those excuses are going to fall away and soon it'll be your only job.