One channel until it works

Ask a founder where they are marketing and you will usually get a list: search, a newsletter, two social accounts, a bit of paid, sometimes a podcast someone suggested. Every item on that list is technically true. There is a profile, there is a page, there was a post in June. And not one of them is running, in the sense of a thing that happens on a schedule and produces a number you can read. The list is not a strategy. It is an answer to a question the founder was slightly embarrassed by.

The spread is not laziness, which is why it is hard to argue with. It comes from a genuinely reasonable instinct: you do not know which channel will work for you, so you hedge by putting a little into each one and waiting to see what responds. That would be sound if a small attempt were a scaled-down version of a real one. It is not. Most channels have a floor below which they produce nothing at all — not a weak signal, nothing — because the mechanics need a minimum before they do anything. One post a month is not a tenth of ten posts a month. It is zero, with effort attached.

Do the arithmetic and the problem stops being philosophical. Say marketing gets [thirty minutes a day](/blog/marketing-on-thirty-minutes-a-day), which is around ten hours a month once you subtract the days that get eaten. Split across five channels, that is two hours per channel per month. Two hours a month on paid means you set up a campaign and never look at it again. Two hours a month on a newsletter is one rushed send. Two hours a month on video is a single clip. Now put all ten into one: ten hours a month is four decent emails, or a campaign you can actually watch and adjust weekly, or enough posts to learn what your audience responds to. Same budget, and only the second version can produce an answer.

five channels at twenty percent is not a portfolio. it is five failures with an alibi.

The other half of running one channel properly is deciding, before you start, what working means. Write it as a sentence with a number and a date in it: by the end of eight weeks, this should produce thirty signups at under twenty dollars each. Those figures are invented; yours come from what a customer is worth to you and what you can afford to spend. The point is that the sentence exists on day one, when you are calm, rather than being constructed in week six out of whatever the dashboard happens to show. Without it you will never kill anything, because every result is ambiguous enough to justify one more month, and you will rotate between channels on mood instead of evidence.

There is an honest exception, and it is the reason "one channel" is not quite the whole rule. Some channels do not answer on an eight-week clock at all. Search is the obvious one: [what actually moves your positions](/blog/what-actually-moves-your-search-positions) works on a scale of months, and judging it in six weeks tells you nothing except that six weeks passed. So the practical version is one active channel — the one that gets the hours, the weekly attention, the pass-or-fail sentence — plus at most one slow bet running quietly in the background, which you deliberately do not judge on the same schedule. Two things, on two different clocks, with only one of them competing for this month's hours.

The second active channel gets added when the first one runs without you, not when it is finished. That is a specific and checkable condition: the weekly motion is written down, someone or something else does the routine part, and your involvement has shrunk to reviewing and deciding. This is the part where drafting and scheduling tools genuinely earn their place — not because a machine writes better emails than you do, but because the repetitive middle of a working channel is what consumes the hours you need for the next one. A draft that is waiting for you when you sit down is an hour you did not spend starting from a blank page. If you are handing that middle over, keep the approval step: a campaign that launches paused and copy that waits for a yes are what make the delegation safe rather than reckless.

Concentration has a real cost and it deserves to be said plainly rather than waved away. One channel means one point of failure. Platforms change their rules, ad accounts get flagged, an algorithm shifts and a reliable source of visitors halves in a week. That risk is genuine, and the answer is not to pretend it is not. The answer is that you diversify from one working channel, not from zero — because a second channel added on top of something that works is funded by the first one's results, while five channels added on top of nothing are all funded by the same ten hours. Spread as insurance only makes sense once there is something worth insuring.

Worth noticing too: the channel that works is often not the one you would have picked. Founders tend to choose the channel they personally enjoy, or the one their competitors are visibly using, and both are weak reasons. The competitor's choice tells you about their team and their budget, not about where your buyers are. The enjoyable one at least has the advantage that you will keep doing it, which is not nothing — consistency beats theoretical fit — but it is worth being honest about which reason you are actually using when you pick.

So the question to answer this week is not which five channels you should be on. It is a much smaller one, and much harder to dodge: which single channel gets the next eight weeks, what number would count as it working, and what you will do on the day that number does not arrive.