Keep calm and learn fast: prioritizing flow in product development
TL;DR: Prioritizing product development isn’t just about picking the most valuable option—it’s about embracing uncertainty, making clear choices based on reasonable assumptions, and optimizing for faster learning through user feedback.
The life of leading products should be fairly straightforward, right?…Well, is that so? Imagine the following exchange in your mind, as a kind of thought experiment…
– Our job is to figure the most valuable thing from our pool of options, and that’s what we do now!
– Right!? But, wait a minute – how can we be 100% sure we made that assessment of value right? Let’s assume we got it at least directionally right for now… Fair enough?
– Fair. But, wait a(nother) minute – if we have an option that is expected to make me $100k (per whatever period of time) and another of $10K, but the latter is supposed to take about 1 week to develop and put at hands of users, while the former 10-15 weeks, does that change our perspective on things.
– Uh, well, likely… Maybe we are better off going with the $10K for now!
– Happy now?!
– Hmm, somewhat… But it just occurred to me this – what if we are awfully wrong with our estimates on the development time too!? What if we were better off going for the $100k after all – “go big or go home”, some would say, right?!
– Ok, now you are really confusing the “heck” out of me… So we are now back to prioritizing by value alone?
– Not quite my point. I’m just wondering. And pretty convinced value and cost (what development time boils down to) might be not the only factors at play. What else could be there?
– Ah, I think I’m starting to get what you mean. Well, surely chances are that both options will face different levels of risk of success. For all kinds of reasons – from challenges along our way (but OK, that’s already sort of covered by the uncertainty on development time) to users not quite using the solution in the way we thought and thus might not really getting the value out of it… And so much more in between and from other angles.
– Exactly. There! And that’s kind of the overarching point – (big) chances are that a lot around what we do is subject to probabilities. For example, it could turn out that the expected $100K of one of the options would be more like $1K – but also it could rather be $100M! For the sake of keeping our feet down on the grass, let’s assume more often not that extraordinarly positive scenario.
– So, if I understand you correctly, essentially we don’t know – or rather cannot know for sure, right!? In other words, our job is impossible! Or maybe even sort of pointless!?
– Hey, don’t be too harsh on yourself (or ourselves, for that matter). I’m just telling you all of this because we are better off accepting reality, and trying to deal with it… It’s a matter of foundation and grounding.
– But I am still awfully confused where do I go from here now!? You kind of took out the carpet from below my feet! And I get that I now, according to you, have them on the ground, barefoot… But I need more than that. I need practical guidance, I need something else I can walk on (I sort of liked the carpet that used to be there, despite knowing now it was a kind of a false sense of security)… Please, HELP ME!
– I surely will. But let’s take baby steps, and you gotta trust the process. Will ya?!
– Yeah, I think I can do that.
– So let’s start with a new image of what your job should look like. Your job is not to prioritize the most valuable thing anymore. Don’t be “cocky” thinking you can know that upfront. Your job is to make clear choices, based on reasonable assumptions you can ground yourself on now, and then from that point on manage the risks and uncertainties along the way. Doing the utmost best to learn more along the way.
– Sounds a bit complicated and not very concrete. Can you dumb it down for me (at least for now)!?
– Let me put this way:
– Your job is to optimize for faster learning! And there’s a lot I could tell you and even demonstrate to ground that statement, but in the spirit of trusting the process, let’s not get into any rabbithole today. Just optimize for faster flow – putting things in the hand of users sooner and invest on learning faster whether you are wrong through feedback loops.
– Hmm, somewhat better and I think I have a better sense of what you mean…
– Good. Let’s test this out in the wild now, shall we?! Let’s see if you really got the gist of what we just covered until now. From what you just learned, how would you go about prioritizing work from now on, then!?
– Hmm, I may be wrong. Maybe I am oversimplifying this… But I got a sense that one thing I could do is just to focus on the things that should take less time to develop and put on the hands of users, somehow!?
– Voilá. That’s it. Or at least that is a solid foundation to begin with…
– Oh, I wasn’t quite expecting that to be right!
– Again, I said it’s a solid foundation – not quite the same as being right.
– I see your point. But then I’m a bit puzzled… What else is there, then!?
– Remember at some point I talked about making clear choices based on reasonable assumptions!?
– I think I do. Not entirely sure what was truly meant with that though – must confess… Remember I said it sounded a bit complicated, didn’t I?
— You did. Let me answer then by stating an assumption I am making when I imply you are onto something when you say to prioritize by shorter work first. I am assuming you didn’t just blindly listed stuff, that you made some kind of assessment on sizing of the work and put that into ascending order, and off you go…
– I think that would indeed be stupid.
– So you will first exercise some sort of selection or filtering of options to only consider things you have a fair enough understanding that valuable could be there (unless you are deliberated taking a rather bold bet, which could also be a valid scenario). I also assumed you brought that down to a level of the smallest thing you could do accelerate the learning – the technical term of a way to look at that being the idea of “right sizing” the work. See what I am doing there?!
– Wow… I believe I do see what you are doing there! You seem to me alluding that all those other things I used to consider being kind of the core of my work are still largely relevant and things I need to try to put my fingers on.
– But there is a fundamental reframing, isn’t it?!
– Hmm, Yeah… if by that you mean those are not anymore what ground my decision-making but rather are elements of going about the filtering and stuff like that. Then I think I get it.
– Welcome to the world of probabilistic thinking! It’s going to be messy, but you are going to do great. Remember the main part there is the “thinking” part… “Probabilistic” is there just to keep reminding ourselves to be humble, that we will be wrong (and more often than we’d like or think we are…).
– Thanks, I guess… No, I’m kidding! I really like this. I also feel like my job became much more interesting and less mechanistic all of sudden.
– That’s it. That’s the thing that matters most. Enjoy the journey. Keep calm, and optimize for flow – that is, faster learning.
By Rodrigo Sperb, feel free to connect, I'm happy to engage and interact. I’m passionate about leading to achieve better outcomes with better ways of working. How can I help you?