Thursday, November 12, 2026
You've known for weeks exactly what needs doing. You still haven't started. Here's why "just do it" doesn't work, and the question that actually does.
Total revenue is the most dangerous number on an SME's dashboard, because it can go up for two completely different reasons: the business is genuinely getting stronger, or it is running faster just to replace the customers quietly leaving out the back door. Bain and Company's research, built on the work of Fred Reichheld, found that a 5 percent increase in customer retention can increase profits by 25 to 95 percent. Most SME owners have heard some version of that stat. Very few have actually measured whether their own growth is retention-driven or replacement-driven, because the one number that would tell them - cohort retention - almost never appears on a standard dashboard.
If you searched "how do I scale my business fast" or "small business growth tips" to land here, this is the uncomfortable prerequisite nobody mentions: growing a leaking bucket faster just means pouring in more water.
Total revenue answers one question: is this month bigger than last month. It cannot answer the question that actually determines whether a business compounds or plateaus - are the customers already won still buying, or is the entire marketing budget being spent replacing the ones who quietly stopped.
The named enemy is not weak marketing or slow sales - it is a dashboard that shows total revenue climbing and lets an owner assume everything underneath it is healthy. A business can grow revenue every month for two years on new-customer acquisition alone while its actual customer base is churning underneath, and the total revenue line will never once tell you. By the time the acquisition well runs dry, or the marketing budget is cut, the truth arrives all at once, and it arrives as a crisis rather than a gradual warning.
The fix is a Cohort Retention Audit: grouping customers by when they first bought, then tracking what each group actually spends over time, rather than looking at revenue as one undifferentiated total.
1. Group customers by month or quarter of first purchase. This is the cohort - everyone who became a customer in the same window.
2. Track what each cohort spends in month one, month six and month twelve. A healthy business sees spending per cohort holding steady or growing as customers stay longer.
3. Compare cohorts against each other. If newer cohorts spend less and leave faster than older ones, growth is being propped up by acquisition, not strengthened by retention.
4. Set one retention target before one acquisition target. A 5 percent improvement in retention is worth chasing before the next round of ad spend, given the profit multiplier attached to it.
"You make success tangible. Help me go from a wish to reality. Now have clarity of purpose and belief. You made tripling the size of my business go from seeming impossible to really possible."
- Lina, Good Gravy
Most SME owners know retention matters in theory. Almost none run a genuine cohort analysis, because it requires pulling data most accounting and point-of-sale systems were not built to surface cleanly, and because celebrating a new-customer number feels better in a board meeting than admitting an older cohort is quietly churning.
This is exactly the kind of blind spot outside eyes catch fastest. A coach who has sat inside the numbers of 200+ businesses coached one-on-one knows to ask for the cohort breakdown before congratulating anyone on a strong month, because a strong month built on replacement rather than retention is a warning wearing a celebration's clothes. The businesses that catch this early fix it with a phone call and a loyalty offer; the ones that catch it late fix it with a much harder conversation about why growth suddenly stopped.
If you have never actually measured whether your growth is compounding or just replacing what walked out the back door, that's where Kevin starts with you.
...and you become the owner whose growth compounds, because the customers you win stay.