The Hoodie Marketer
Blog / Strategy

How I decide what to stop doing

The hardest lean skill is killing things that are working a bit, to free capacity for things that could work a lot.

Dom O'Brien
Dom O'Brien
1 Sept 2026 · 7 min read
How I decide what to stop doing
In this article
Not the cost conversation, the activity oneThe standup that was quietly doing damageWhat it was actually costingHow I actually run the audit

I’ve written before about finding savings in a marketing budget. This is a different, harder conversation. It’s not about what costs too much. It’s about what’s working just well enough that nobody ever questions it, while quietly eating the capacity you need for the thing that could actually work.

Not the cost conversation, the activity one

Cutting a line item that’s obviously failing is easy, because nobody questions it. The hard version is cutting something that’s mildly, defensibly working, because “it’s not doing nothing” feels like a reason to keep it. It rarely is. The real question isn’t whether something’s working. It’s whether it’s the best use of the capacity it’s consuming, and that’s a much less comfortable question to sit with.

The standup that was quietly doing damage

For a while, I ran a daily standup built around yesterday’s sales numbers. It felt disciplined. Every morning, the team looked at what sold, and adjusted. The problem was the sales cycle was far longer than a day, so nothing anyone did today was going to move tomorrow’s number in any meaningful way. We were reading noise and reacting to it like signal.

Worse, it fed a habit further up the chain. The founders would see a number move, decide it meant something, and land on a new priority by 9am, different from yesterday’s new priority. Twenty shiny new objects a day isn’t an exaggeration, it’s roughly what it felt like. The team never got the run of uninterrupted weeks that any real initiative actually needs to prove itself, because we were all reacting daily to a number that couldn’t tell us anything useful on a daily timescale.

We were reading noise every morning and reacting to it like signal, and it cost us the one thing every real initiative needs: an uninterrupted run to actually prove itself.

What it was actually costing

Stopping the daily standup, and reporting for the sake of reporting more broadly, wasn’t really about saving the meeting time, though that helped. It was about removing the daily trigger for changing direction based on a number that moved for reasons that had nothing to do with yesterday’s actions. Once that trigger was gone, priorities could survive longer than a day, which meant they could actually be tested properly instead of abandoned before they’d had a fair run.

How I actually run the audit

Where do you start? List everything - recurring, meetings, reports, weekly rituals, and for each one ask what decision it actually changes, and on what timescale that decision can realistically be made well. If a report or ritual operates on a shorter cycle than the reality it’s measuring, that’s the signal, not whether the number it’s reporting looks good or bad on any given day.

Then the harder part: say out loud what freed-up capacity is now going toward, specifically. “We stopped this” without “and here’s what we’re doing instead with the time” just reads as doing less, and it doesnt give founders the confidence that they are gaining something vs losing more.

The takeaways
The hard version of stopping isn't cutting what's failing. It's cutting what's mildly working, because that's where the real capacity is trapped.
Watch for rituals that operate on a shorter cycle than the reality they're measuring. They don't inform decisions, they manufacture unnecessary ones.
Always pair 'we stopped this' with exactly what the freed capacity is now doing instead, or it just reads as cutting for its own sake.
PrioritisationCMO lessonsFocus
Dom O'Brien
Written by

Dom O'Brien

CMO at MATE and author of The Startup Marketing Playbook. Fifteen years building lean marketing teams that punch above their weight.

Work with Dom →
The newsletter

One email a month. No spam, no "crushing it." Unsubscribe whenever.