Publishers: ‘failing fast’ in ad ops only works if someone’s writing it down
BY ANONYMOUS PUBLISHER + LIZ MOOREHEAD, BEELER.TECH
The following editorial is a free Below the Fold feature preview from The Stack subscription.
Move quickly.
Be agile.
Don’t get too attached to a particular initiative.
Fail fast.
We’re going to “learn by doing.”
We’ve all heard this pitch before.
It’s a seductive premise, because it speaks to the parts of us that know all too well the crippling consequences of too many processes, too many meetings, too many spreadsheets, too many post-mortems. Everyone is miserable, and somehow, while tasks are getting checked off diligently, nothing is actually being accomplished.
Of course, this whole “fail fast” concept is only effective when the learning part actually happens: the part where you figure out why something didn’t work, and you take time to purposefully adapt before you move forward.
Unfortunately, as someone who’s worked in ad ops at different publishers that love to talk a big game about “failing fast,” that critical learn-and-adapt phase rarely (if ever) occurs.
The usual cycle looks something like this:
- We “move quickly”
- We “fail fast” (i.e., something breaks)
- We patch it enough to move forward
- Someone asks for a post-mortem
- They’re told “sure, let’s just get through this bit first”
- The conversation never happens, so we “move on”
Then, three months later, the same, now avoidable breakdown happens again.
And everyone acts surprised. “How did this happen? We didn’t account for this.”
We’re not learning these lessons for the first time
In many cases, we’ve already learned them, but we neglected to create the systems or the documentation required to carry what we learned forward.
Instead, these lessons live inside the heads of two or three people at a publisher. Those people tend to be the most experienced: they know which handoffs always break, which assumptions are deeply flawed (at best), and which workflows no one has ever bothered to document, because everyone is too busy “executing” and “focusing on what’s working.”
Of course, when one of those people leaves, the company loses something it didn’t know it had. And the new person who replaces them gets to rediscover an old problem at full price. It happens constantly. I’ve watched it happen and I’ve been the new person doing the rediscovering, and both sides of it are exhausting in different ways.
I’m all for “failing fast,” but if we don’t also take the time to learn from our mistakes and document those learnings, we’re just trapping ourselves in a cycle of… well, failing.
Every time we “relearn” that a particular partner doesn’t pay on time, every time we “relearn” that a specific ad product doesn’t perform the way sales pitched it, every time we “relearn” that a certain agency loves to show up randomly once quarter with a brand-spankin’ new 7,000-keyword block list: somebody is eating that cost.
Somebody downstream… often in ad ops.
But I won’t pretend we’re the only ones dealing with this problem at publishers. It’s hurting all of us.
What’s insane is that this cycle can continue for years
A friend of mine, who’s been in our industry much longer than I have, says he’s watched the same doomed-to-fail initiative come back around five or six times over the past seven years.
Every time, it’s pitched by a new leader as an innovation, or a fresh way to tackle an age-old costly inefficiency. But they genuinely believe it is new, and that they are bringing a fresh perspective to a lingering problem others have failed to address.
Then, a few weeks later, they’re all huddled on a Zoom call, listening to people make excuses or pass the buck on why that initiative inevitably failed. Of course, it did. Because after the previous rounds, no one bothered to document what they learned or research what they already knew.
I asked him, “What about you? You were there before. You knew it wouldn’t work. Why didn’t you say anything?”
He said, “Oh, I tried. For years, I tried. But I learned a long time ago that the more I raised my hand, the more I was looked at as a problem. Sue me, I got tired of being called a ‘bad team player’ every time I tried to help.”
His experience is a familiar one. When I’ve tried to do the same, the response I get is rarely “great, tell us more.”
Usually, it’s some version of:
- “You’re being negative”
- “This is not the culture we’re trying to instill here”
- “You need to be more solutions-oriented”
It’s frustrating when I’m treated as if I’m going against a company core value by speaking up. As a result, the substance of what I’m sharing doesn’t get addressed, because leadership is too busy clocking me for having an attitude problem.
In fact, at one publisher I worked at, I was given a bad mark on a performance review for not “leading with the solution” enough (one of our core values). My crime? Suggesting we institutionalize some sort of ongoing documentation strategy for processes, because high turnover on the sales team was creating issues during hand-off, which (of course) impacted initial campaign performance. I even offered to take on most of that work, to help the team.
But it didn’t matter.
I think a lot of us have been on the receiving end of something like that.
You bring up a structural issue and get told you’re not bringing the right energy.
A lot of the language we’ve absorbed in the last decade does this
Fail fast. Be agile. Live in the solution. Move with urgency.
None of those phrases are stupid on their own, because we don’t have a language problem.
The problem is what happens when an org adopts the language without doing any of the slower work it depends on. Building trust between teams. Writing things down so they survive turnover. Setting expectations before the work starts instead of after. Leaving room to ask whether the last thing actually worked.
Failing fast only works if you slow down afterward to make sure you’re actually learning the lessons you need for the future.
Otherwise, we’re just experiencing our failures and then immediately forgetting them until the next time it happens.
I’m hesitant to say this is being done on purpose, but you can’t deny there are some people who benefit from the coverage. When everything is moving fast, there isn’t really time to ask who agreed to the 80% viewability target nobody can hit, or who set the timeline that didn’t account for trafficking, or who promised the client something operations was never going to deliver.
The pace itself does the work of avoiding those questions.
By the time the campaign is underperforming, the salesperson is three pitches deep into the next thing, and whoever’s trying to make the numbers work after the fact ends up holding it. I’ve held that bag. Most of the people I know in this industry have held that bag. It’s exhausting and it doesn’t ever get acknowledged because acknowledging it would mean slowing down enough to look at it.
I want to be careful here, because I’m not arguing for slowness
Ad ops is not the career for anyone who wants to live life in the slow lane, and pretending otherwise would be insulting to anyone who’s actually done this work.
But we also need to be honest about a few things:
- If your team has shipped the same kind of fix four times this year, that’s something. I don’t know what to call it, because it’s not speed. Maybe just running in place. Whatever it is, it’s worth saying out loud at your company, because the whole reason “fail fast” was supposed to work was that you’d come out the other side smarter. But if you’re not coming out smarter then we’re all just lying to ourselves about what we’re doing.
- If the institutional memory at your company lives in two or three people, that’s a problem worth taking seriously. Those people deserve to be treated like they’re carrying something valuable. And your team or your organization as a whole should be writing things down before the day comes when one of them leaves, because that day is going to come, and a year of expensive relearning is going to follow it.
- If your post-mortems don’t lead to a change someone can point to three months later, they aren’t post-mortems. They’re meetings where we agreed something was bad, and then we did our best to pretend it never happened, because we all got uncomfortable. I’ve sat through more of those meetings than I can count. They feel productive in the moment… and they don’t change anything. Then next time the same problem shows up we have another one of those meetings about it.
Are we learning or are we just recovering?
Recovery is what gets you through the week.
Learning is what would change the next one.
Recovery depends on the same handful of people stepping in over and over to keep things from falling apart.
Learning would eventually mean they don’t have to.
I don’t think most of the publishers I’ve worked at can tell the difference anymore. I think a lot of us in ad ops have been recovering for so long that we’ve accepted it as the cost of our profession, and the people above us calling it “failing fast” have mostly never had to do the work that the recovery depends on.
But I’m tired. A lot of my friends in ad ops are tired. We keep being asked to absorb the same preventable problems on behalf of organizations that have decided learning isn’t worth the time it takes. I don’t have a clean ending for this. I don’t think the situation has one. I just wanted to say it out loud, because the alternative is to keep running fast and quiet, and I’ve done enough of that.