In a conversation on LinkedIn, a “shocking” statement is delivered: “If you feel like you haven’t done your best, don’t deliver.”
As it stands, this sentence means nothing to me. Everything will depend on the context and the implications. To highlight this, I respond: “If you don’t feel uncomfortable with what you’re delivering, it means you’re delivering too late” said the CEO of LinkedIn, I don’t remember when or where. I add “it would be a shame not to deliver and miss out on a learning opportunity. Perfect is the enemy of good, etc. Doing your best sounds too much like perfection and perfection is a bad advisor these days.” This famous phrase cited approximately won’t mean anything either if it doesn’t have more context.
What follows is therefore a long conversation that’s not necessarily useful, because precisely everyone is floundering in their own context.
But that same evening around 9pm I’m challenged on our internal communication group (slack) by a colleague about this dialogue.
Her:
Do you really think that (we’re talking about: “If you don’t feel uncomfortable with what you’re delivering, it means you’re delivering too late”)? Isn’t there a risk on the contrary of ending up getting used to mediocrity?
Me:
Nothing to do with mediocrity. The guy (from LinkedIn) he tests his hypotheses. 85% of what he delivered was removed. (85% of LinkedIn’s features were reportedly removed1) And he’s especially not talking about quality. He’s talking about product scope. Quality is not in question.
Her:
Hmmm I’m tired, maybe I misunderstood… but when I read that I get the impression that it justifies three-quarters of crappy products “sorry we did our best, we didn’t have time to focus on what creates value at the risk of losing the user”.
Me:
It’s hard for you, I understand. Because you’re reading it in terms of “perfection”. Too bad (for the LinkedIn conversation) I’ll let people believe what they want.
Her:
Yes I don’t know, but I have the daily impression of hearing ALL my teams tell me “uh sorry, but we didn’t have time to do it properly, the user experience isn’t the target, but we’ll make up for it in the next sprint”, 6 months later we’re not there and we’re accumulating debt.
Me:
The conversation you’re describing doesn’t talk about the substance. What do we want? What value? Especially not quality or user experience or code. What real product value? After the questions and understanding of these things read much better. In my eyes. Just saying “6 months later” means we’re not answering the real question.
Her:
And honestly I’m exhausted because I work like crazy and not even 5% of my work is finally properly put into production, I’m at a stage where I deliver targeted paths AND degraded versions all the time “that we can’t do” when if we really wanted to do it well we could.
Me:
You need to deliver much more into production and measure. And learn. But often politics takes precedence over value.
Her:
Well I don’t know I’m a bit dark right now I see everything through a spectrum where teams are without passion and deliver crappy stuff! There’s zero value.
Me:
You don’t know. You’re not delivering. I’m provoking, but that’s the idea. If you deliver crappy stuff to me in production And you measure against the real value sought. Well believe me you’ll learn quickly.
Her:
No I’m not delivering, but I don’t understand how, why we’re not delivering value to the user honestly? I feel like I’m a worse “designer” than a year ago and it’s costing me.
Me:
A “designer” is like a coder. It’s the soft sciences that will often – not always – make the difference.
Her:
I’d prefer we deliver later, but well, even if it means delivering things in pre-production, testing learning, but that in front we’re perfect when we deliver except that for everyone delivering means in front2. And that costs me a lot then they call us back “to put out the fire”.
Me:
I’d prefer you deliver much less. But well. And that you learn and measure. So I’d like you to deliver as early as possible. So the smallest thing of value. Nothing to do with good or bad quality. It’s never the debate.
And you write: “in front we’re perfect”, wrong. It’s never perfect before being put into production.
Her:
Yes, but that’s also my problem I deliver the target (that’s way too much) then we’d need to break it down with everyone. Except it’s never followed through! “They forget the target paths in the meantime”
Me:
No not break down. Expand. Expand. Expand.
Her:
It’s never perfect I have trouble with not perfect haha
Me:
Drug yourself :) What you’ll think is perfect won’t be for someone else. So what do we do? There’s no shared perfect.
Her:
Well I don’t know, I know the problem must also come from me, but now I don’t know how to work so that we go towards not always “OK that’ll do”, but towards “guys here we give everything”.
Me:
Giving everything sounds like “full speed”. You don’t get quality by going full speed. The problem isn’t in the pace, but in the scope.
Her:
Why no quality if we take time, but with a more long-term vision? Here we patch, we patch, we patch and we end up with too many patches that make evolution impossible.
Me:
Do things at your own pace for quality. But the more you release complete things the more you risk messing up. Because too many hypotheses. Not enough learning.
Her:
But before I did user tests, I confirmed my intuitions and made them evolve!
Me:
Why not take your time. But work on small pieces. And observe learn in production. It changes and it stays work in a vacuum. And if everything stays OK the longer you wait the less it will be. So, release small pieces quickly. Quickly is different from “going fast”. More small pieces.
Her:
I don’t know, I don’t know anything anymore… I can’t be satisfied with what we deliver! Delivering a small piece of registration flow isn’t possible.
Me:
This battle shouldn’t be limited to work? Battle for perfection.
Her:
Is a life without constant search for elevation worth it? I’m not sure. Well OK for others, everyone deals with themselves, but not for me
Me:
Constant search for elevation has nothing to do with perfection. Someone who constantly seeks to elevate themselves knows by definition that there’s no perfection.
Her:
Hmm I don’t know perfection for me is being able to sublimate everything and be as close as possible to the truth. Elevating yourself is wanting to be as close as possible to the truth so as not to suffer.
Me:
And does it change, does it move, this truth? Does it evolve? And when you do this search for elevation. Do you do one big annual piece? Or is it constant small adjustments that enrich each other?
Her:
In any case I know where I’m going. Or at least where I want to go.
Me:
Yes a sense, a direction, but not the details. A why, a direction. But not the details. For products it’s the same. Or you have the details of your current reflection.
Her:
But if at some point we don’t get our hands into the details it remains “ideation”
Me:
And the next one will be enriched by the previous one. And then you’ll think about those details. You don’t prepare the details of your introspection in 3 months…
Her:
No of course it’s the hindsight that gives meaning to what we’ve experienced and therefore to the details. But there are also the means we give ourselves to get there and that goes through the details otherwise we go towards the lighthouse and we risk tearing our hull. Anyway I’m not being sharp tonight, but just that I’m tired “of OK, we can’t do it, so we settle for that”
Me:
Yes the details, but not of what comes later.
To conclude.
There are several fundamental debates. I note these four:
- Deliver to experiment and learn rather than deliver to finish.
- Quality comes from regularity.
- We don’t go faster, we deliver smaller pieces, for which we’ll try to densify the value following the learnings.
- Introspection in each person is essential.