How to debrief a launch that flopped without taking it personally

Four sales. You spent six weeks on it, you told people it was coming, you rewrote the sales page twice, and the launch produced four sales. It is now the evening of day three and you have drafted a post apologising for the whole thing, which you have not sent, and you are wondering whether the sensible move is to take the product down before anyone notices it is still there. Do not do the review tonight. Everything you conclude in this state will be about you rather than about the launch, and it will be wrong in a way that costs you the next twelve months.

Why shame gives you the wrong answer

A flop is not a neutral event. You put your name on it, you asked people to look, and the silence that follows feels like a verdict delivered by everyone you know at once. In that state the brain does something specific and unhelpful: it looks for a single, total explanation. Nobody wants what I make. I am not good enough at this. The market has moved on. Those are stories about identity, and identity is the one variable you cannot test.

The genuinely difficult part is that some launches fail for reasons you will never establish. A big news week. A price that was fine but arrived in a bad fortnight. An audience that had already spent this month. You can do a careful review and end up with two plausible explanations and no way to choose between them, which is unsatisfying and also the honest outcome much of the time.

What you can do is narrow it. There are only a handful of places a launch breaks, and the numbers on your product page will tell you which one it was in about twenty minutes, provided you look at them rather than at your feelings about them.

The midnight version and the daylight version

Same launch, two readings. The left column is what most people conclude the night it happens. The right is what the same evidence supports once you have the per-product figures in front of you. Full analytics with traffic source start on Starter, and the pricing page sets out what each plan shows.

The debrief, in order

Book an hour, at least two days after the launch closed, ideally in the morning. Open your analytics and work through these in sequence rather than jumping to the conclusion you already suspect.

1. 1Write the launch down before you look at anything: what you sold, at what price, to whom, over how many days, and what you expected. The expectation matters, because a flop is a gap between two numbers and you need both.
2. 2Open the product in analytics and note three figures: views, buyers, revenue. Just the figures, no interpretation yet.
3. 3Divide buyers by views. That is your conversion, and its only useful comparison is your own other products, not a number you read somewhere.
4. 4Look at traffic source. Note where the views came from and, separately, where the buyers came from. These two lists are rarely the same, and the gap between them is often the most useful thing in the whole review.
5. 5Decide which failure you had. Low views with normal conversion means not enough people saw it. Normal views with near-zero conversion means the page did not convince the people who did.
6. 6If it was conversion, reread the product page as a stranger who has never heard your voice. What does it promise in the first two lines, and to whom.
7. 7If it was traffic, count the actual places you announced it and how many times. Most launches described as heavily promoted turn out to have been mentioned twice.
8. 8Check what has happened since the launch window closed. Some products are simply slow, and a week of quiet followed by a steady trickle is a different animal from a flat line.
9. 9Write three sentences: what happened, the most likely cause, and the one thing you will change next time. Date it and save it.

The same four sales, read properly

Take those four sales and put two different sets of numbers behind them. In the first, the page had 620 views over five days and four people bought. That is a page problem: hundreds of people arrived, read what you wrote, and decided not to. The offer, the promise or the price is not landing, and no amount of extra promotion will fix it. Sending another thousand people to a page that convinces nobody just produces a larger silence.

In the second, the page had 55 views and four people bought. That is not a flop, that is a product converting at a rate you would be delighted with, shown to almost nobody. The fix is entirely on the distribution side, and the traffic source breakdown will tell you which channel actually produced those four, which is where the next month of effort should go. Same four sales, opposite conclusions, and the only way to tell them apart is by looking at the per-product figures. If the numbers are genuinely close to zero on both counts, the zero-sales troubleshooting guide works through the mechanical causes first, which is the right order.

The mistake most people make

They change everything at once. New price, new name, new sales page, new audience, relaunched a fortnight later with a different angle. If it works, you have learned nothing, because you cannot tell which change did it. If it fails again, you have learned less than nothing, because now you believe the whole category is dead. Change one thing. Give it enough traffic to mean something. Then change the next one. This is slower and it is the only version that accumulates knowledge, which is the actual asset a flopped launch can leave you with.

A flop is data wearing an unpleasant costume. Take the costume off by waiting two days, look at views, buyers and source in that order, and write down the one thing you will change. You will still be disappointed. You will just be disappointed about something specific, which is the only kind of disappointment you can do anything with.