Think of a small piece of feedback you've been sitting on. You've been rehearsing it in your head for days, and perhaps talking to yourself: "I wish I didn't have to give this, or at least be better at giving this kind of feedback by now."
I was on a call the other day with a senior engineer in that spot. We were talking about how to say a hard thing to a colleague we care about, how to give it without hurting the relationship. Towards the end of that, I heard myself ask: can we shift our communication a bit to the left?
Is there something we could do better way earlier, back at the beginning, when we still feel excited about the work and good with each other, so that we wouldn't end up here, trying to give feedback neither of us wants to give?
I realised we spend so much energy getting better at feedback, as we should, but nowhere near enough on what comes so much before it: the agreements we could co-create with each other to set our collaboration up for success.
The way it usually goes
This one is a hypothetical but quite a common scenario. Let's say a peer or a stakeholder hands you some usual work. In your head, you've got two weeks. In theirs, one week felt doable. But no one really said it out loud, because well, you know each other.
You carry on with everything else on your plate, and then they check in after three days, and you go: yeah, I haven't started yet, but I'll get to it.
You get just a little irritated that the check-in came so fast. They, on the other hand, are losing a little trust in the timeline. But still, it's not big enough to mention. Another three days pass, and they're back again, this time saying they actually need it done by tomorrow.
Now you're stressed, a bit panicked maybe, but you care about your work and this person, so you make it happen. All is well. Except now both of you are sitting on feedback you want to give each other.
You want to tell them it really wasn't ideal to check in three times for this simple thing. They want to raise that you don't share progress proactively, that you need chasing to get things done.
None of it feels great. But somehow you both believe it has to be said, the moment it happens, because that's what all the feedback articles tell you to do. And so figuring out how to say it constructively becomes the work itself.
The way it could go
Let’s go back to the beginning. The moment you were asked to take on something.
What if you'd had the chance to slow down that first conversation to make things clear for both of you?
In those few minutes, you might align on a deadline, double check your own capacity, or you can get curious about their needs to find better ways to solve the problem.
Here are a few examples:
"I'd love to take this, but I'm a little full right now with X. When do you need it?”
"Currently, I'm finishing another project. The earliest I can have a look would be next week. Shall we find someone else from the team if it's urgent?"
"Oh nice, I can do this, but I know Julie is the expert on it. I'll loop her in and see what we can do. Can I get back to you tomorrow so we can agree on how and when?"
"Sure thing. I think this can be done two different ways: A or B. What do you think of these options?"
And the person might come back with:
“It is kind of urgent, thanks for letting me know that your capacity is low. Who would be the best person to go to?”
“2 weeks is fine, but I will double check with my own manager and get back to you today.”
“Uh, I think actually B sounds so much better. Can we test it together in a week?”
It might not exactly look like this, these are just examples I can think of from my own experiences.
Here is how it helps you:
It gets you more data, and with that you can scope the request better, understand what it takes.
It brings your peer or stakeholder into the thinking, which often helps them get clearer on what they actually need and how and when.
You can co-create an explicit agreement, a steppingstone together that serves both of you and the work.
That agreement is priceless. Not only does it boost confidence in your ability to set yourself up for success, it can also serve as a third thing between you and the person you collaborate with. It makes progress possible, effective and clear.
And even after this when things don’t work out the way you both intended to do so, talking about it and making it better next time feels so much lighter and easier than sitting in frustration.
I refer to this as co-created agreements > expectations. In the context of this call, we called it shifting left in your communication.
We already know how to do this
Is that not like everything we've done for the last twenty years to write better software?
Doing important work earlier in the development lifecycle instead of waiting until later.
Like we try to catch things in a unit test. We design for availability from the start. We build short feedback loops into the systems and processes. We also still make sure things exist all the way to the right with end-to-end testing, black-box monitoring, retrospectives on incidents.
Can we just copy-paste that to our communication and collaboration?
Sounds easy enough. What makes us skip it?
Being human is a little more complex
We all work in environments that collectively want us to operate in a rush, especially with the speed at which things move today.
In that sense of urgency, staying in uncertainty with another person feels almost "inefficient”.
Plus, it can also feel a little vulnerable and exposing too:
Like if we have to ask a few more questions, maybe we missed something we should already know. We’d rather look capable in the moment.
Sometimes we also don’t want to be the difficult one, the person making a small thing bigger than it needs to be.
And sometimes, we just care about our connection to this person. We like them, and spelling things out can feel almost cold. As if agreeing on the details somehow means we don’t trust each other.
None of these reasons are silly, really. They’re raw, human responses to the sociotechnical environments we work in, and I feel them too. I still remember the time when I cried over a pull request review.
But I also know that great engineering requires a solid foundation, and part of building that foundation is our ability to stay in the uncertainty long enough to lay things down properly. Human to Human.
Making that part of the “work” is in our own interest in succeeding.
It might matter more now
I hope this doesn't land as me being against feedback. On the contrary, I want to make it easier, a lighter thing.
I also wholeheartedly believe our communication with humans matters even more today. The more we hand our work to agents and get swept up in the speed of it, the less we actually talk to each other. And that's widening the alignment gaps we already have, often leaving us working on the wrong problems. (and tell me if it doesn't start to feel a little lonely)
Technology keeps advancing, but we're not going anywhere. And I hope we still want to get better at working with each other as much as we do at working with the agents.
If that's you, experiment with shifting left on your conversations. Stay a minute more curious, and lay the foundation to succeed on.
