The only thing that worked was hand-writing twenty emails a week — do you keep doing that or is it a trap?

Do the arithmetic before deciding anything, because "does not scale" is usually a feeling rather than a number.

Twenty hours a week producing — what, one customer a fortnight? Work out your actual cost per customer in hours, then compare against what a working scalable channel would cost you. Content takes months to produce anything and often produces nothing. Ads cost money you may not have and need a message you have already validated.

When people run this they usually find the manual channel is not merely acceptable, it is the best-performing thing available to them by a distance, and the discomfort is social rather than economic — it feels like it should not count.

The number that would actually change my answer is how many good prospects are left. Twenty a week is sustainable if there are thousands of them. If your total addressable set is four hundred people and you have contacted a hundred and eighty, you have about nine weeks left and you should be building the next thing now. That is the calculation to run, and it is the one nobody does.

So: count the pond before you worry about the fishing method.

Carry on. You are much earlier than the advice you are reading is aimed at, and the advice has been flattened in the retelling.

The original point was never "stop as soon as it works". It was that unscalable work is how you learn what to say, and you graduate when you have learned it — not on a schedule. At eleven customers you have not learned it yet. You have eleven data points, several of which are probably not the same kind of customer.

The honest signal to move is not a customer count, it is one of these:

• You physically cannot keep up, and demand is being lost because of it
• The emails stop working, meaning your reply rate falls as you exhaust the obvious people
• You notice you are writing the same email with the names changed

That third one is the real graduation, and it is worth waiting for, because it hands you the scalable version for free. When twenty personal emails have converged into one argument, that argument is your landing page, your posts and your ads. Trying to write those before convergence is exactly why your scalable attempts returned zero — you were guessing at a message you had not found yet.

So the answer to "is it a trap": no, but stopping too early is. The trap is building a channel for a message you cannot yet state.

The middle path that let me keep going much longer: automate the parts that do not do the work, and keep the part that does.

An hour per email is mostly not the writing. It is finding the person, working out what they do, and checking they are a fit. That research is compressible:

• Build a list in batches rather than one at a time. Sourcing fifty prospects in one sitting is far faster per head than sourcing one.
• Keep notes in one place with the same fields each time, so the research is reusable and you stop re-doing it.
• Have a skeleton — the parts that are always the same, with obvious gaps for the two or three sentences that are actually about them.

Done properly this takes an hour down to twenty minutes without changing what the recipient reads, and the specific sentences are the entire reason it works, so they are the last thing you touch.

The other thing worth doing now, while you are still writing them by hand: keep every email you send and mark which ones converted. You are sitting on the raw material for whatever you build next, and it is much easier to find the pattern in forty saved emails than to reconstruct it from memory in six months.

When you do graduate, that archive is the graduation.