Introduction — Twenty-Three of a Hundred
One real customer's brand dropped into a hundred designs. Twenty-three still looked right.
⚠ Kit spent about fifty-five hours making a hundred social media templates and sold them on the number. In month five they dropped one real customer's brand into all hundred and counted what still looked right. Twenty-three did.
The brand that broke it
⚠ A garden-maintenance business. Kit asked for their actual assets to test with.
| ⚠ What arrived | ⚠ Detail |
|---|---|
| ⚠ Primary colour | ⚠ ⚠ A dark, heavy green |
| ⚠ Secondary colour | ⚠ ⚠ None; they had never chosen one |
| ⚠ Font | ⚠ ⚠ A serif, from their signwriter |
| ⚠ Logo | ⚠ ⚠ Wide, with the business name inside it |
| ⚠ Photographs | ⚠ ⚠ Four, all landscape, one blurred |
⚠ Kit's set was designed in soft pink and cream, with a light geometric sans, tight text set over pale photographs, and square crops.
⚠ Nothing about the customer's brand was unusual. It was simply not Kit's.
What happened to the hundred
| ⚠ Outcome after the swap | ⚠ Designs |
|---|---|
| ⚠ Still looked right | ⚠ ⚠ 23 |
| ⚠ Text unreadable against the darker colour | ⚠ ⚠ 31 |
| ⚠ Layout broke when the serif replaced the sans | ⚠ ⚠ 24 |
| ⚠ No room for a wide logo | ⚠ ⚠ 19 |
| ⚠ Required a photograph shape they did not have | ⚠ ⚠ 27 |
| ⚠ Palette fought the green directly | ⚠ ⚠ 41 |
⚠ The categories overlap, because most failures failed in more than one way at once.
⚠ Kit's note: "I had not designed a hundred templates. I had designed one look, a hundred times, and the look was mine."
Why nobody had told Kit
⚠ The set had been on sale for five months. Section 2.7.
| ⚠ Signal available | ⚠ What it said |
|---|---|
| ⚠ Sales | ⚠ ⚠ Steady |
| ⚠ Reviews | ⚠ ⚠ Positive, and vague |
| ⚠ Refunds | ⚠ ⚠ A few, without explanations |
| ⚠ Questions | ⚠ ⚠ "How do I change the colours?" |
| ⚠ Complaints about the design | ⚠ ⚠ None |
⚠ The commonest question was the whole problem, asked politely. People were not asking how to change the colours because they were curious. They were asking because the design stopped working the moment they did.
The rebuild
⚠ Month eleven. Kit rebuilt the idea with a fixed constraint: every design derives from one system, and nothing ships until it survives three real brands.
| ⚠ Measure | ⚠ Original | ⚠ Rebuilt |
|---|---|---|
| ⚠ Designs | ⚠ ⚠ 100 | ⚠ ⚠ 24 |
| ⚠ Hours to build | ⚠ ⚠ ~55 | ⚠ ⚠ ~21 |
| ⚠ Brands tested against | ⚠ 0 | ⚠ 3 |
| ⚠ Brand-swap survival rate | ⚠ ⚠ ~23% | ⚠ ⚠ ~86% |
| ⚠ Price | ⚠ ⚠ Blank | ⚠ ⚠ Blank, similar |
| ⚠ Refund rate | ⚠ ⚠ Higher | ⚠ ⚠ Lower |
| ⚠ Reviews | ⚠ ⚠ Polite | ⚠ ⚠ Specific |
⚠ "Twenty-four designs that work in somebody else's colours are worth more than a hundred that work in mine, and it took me eleven months to be able to write that sentence." Chapter 21.
The other discovery
⚠ Month sixteen. Kit noticed something in the survey answers.
| ⚠ Question | ⚠ Commonest answer |
|---|---|
| ⚠ ⚠ How many of the designs have you used? | ⚠ ⚠ Between four and eight |
| ⚠ ⚠ How often do you post? | ⚠ ⚠ Two or three times a week |
| ⚠ ⚠ What do you need most? | ⚠ ⚠ "Something for next month" |
⚠ A customer posting three times a week needs about a hundred and fifty graphics a year. They had bought one pack, used six designs, and gone back to having nothing to post.
⚠ The hole in their week was recurring, and Kit had sold them a one-off. Chapter 6.
| ⚠ Layer | ⚠ Share of revenue at month thirty-four |
|---|---|
| ⚠ Monthly pack | ⚠ ⚠ ~34% |
| ⚠ Evergreen bundles | ⚠ ⚠ ~26% |
| ⚠ Single sets | ⚠ ⚠ ~17% |
| ⚠ Brand setup service | ⚠ ⚠ ~13% |
| ⚠ Custom commissions | ⚠ ⚠ ~10% |
The four rules
⚠ This trade has four structural facts that do not go away.
⚠ The template must survive their brand, not yours. Every design decision has to hold when the colours, the font, the logo and the photographs are replaced by a stranger's — and yours are the only ones you ever look at. Rule 1.
⚠ They do not need a template; they need next month's. The customer's need repeats weekly, and a set sold once does not fit a hole that reopens. Rule 2.
⚠ You are selling inside your competitor's shop. Your buyer is already in the tool, surrounded by thousands of free templates made by the platform itself. Rule 3.
⚠ A link is not a file. Delivery is a link that can leak, break or be replaced — and once they copy it, their version is theirs and you cannot reach it. Rule 4.
The two numbers
⚠ The economic unit is revenue per brand-tested design — what a set earns divided by the designs that actually survived contact with three real brands. Untested designs count as nothing, because that is what they are worth. Chapter 30.
⚠ The leading indicator is the brand-swap survival rate: the share of your designs that still look right when a stranger's colours, font, logo and photographs are dropped in. It is measured by actually doing it, it takes twenty minutes, and it moves about a season before refunds and reviews. Chapter 24.
Where to start
⚠ Read Chapters 5 to 8 first, in order.
⚠ Then Chapter 21, before you publish anything else.
⚠ Then take one real small business's colours, font and logo, drop them into your own set, and count. That takes twenty minutes and it is the most useful twenty minutes in this book. Resource 31.
⚠ Kit's last note on month five: "The customer asked me how to change the colours. I sent them a helpful reply about the brand kit. The correct answer was that my template could not survive it, and I did not know that yet."
©2026 James Henderson / https://localhandyman.work