A second opinion · No client, no brief · Note 02
Averaging is how loss-making cohorts disappear.
Enter your channels one level down — six numbers each — and this shows how much of the contribution your good channels create is being cancelled by the ones below the line.
It is already filled in with an example, so you can see the output before you type anything.
Everything runs in your browser. No server, no account, no analytics, no network request after the page loads. Close the tab and the numbers are gone.
Edit any cell — everything below updates as you type.
| Channel / partner / seller | Gross addsper month | CACall-in, per add | ARPUper month | Direct cost% of revenue | Churn% per month |
|---|
Gross adds — new subscribers this channel produces in a typical month.
CAC — everything you pay to get that subscriber through that route: commission, bonus, handset or SIM subsidy, the channel's share of marketing spend. If you only count commission, you will understate the damage.
ARPU — average monthly revenue from the customers this channel recruits, not the company average. Using the company average defeats the exercise.
Direct cost — as a percentage of that revenue: wholesale or interconnect, network, SIM, billing, bad debt, care. Variable only. Leave out anything that carries on if the channel stops.
Churn — monthly, for this channel's cohorts. If you have it annually, divide by twelve.
Sorted by total contribution.
| Channel | Paybackmonths to break even | Per subcontribution | Cohort totalper month acquired |
|---|
Churn is assumed constant month to month and future cash is not discounted. Real churn is front-loaded, so an early-churn channel is worse than this shows, not better.
It ignores cross-sell and price changes, and says nothing about your existing base — only about the cohort you acquire from here on. A channel that recruits customers who later buy a second product can be negative here and correct in reality.
It tells you where to look, not what to conclude. A channel can be below the line because of the terms you set, the customers it attracts, or the way it sells — three different problems with three different fixes. And channel averages conceal exactly the way company averages do: if something comes out negative here, run the same test one level further down, on the individual partners, stores or sellers inside it. That is usually where the answer stops being ambiguous.
Send me the shape of what came out and I'll tell you whether I think it's real, and what I'd look at next.
No calendar. No discovery call. If it isn't something I can help with I'll say so, and where I can, who else to ask.