~200 cold emails in: 3 humans replied, all 3 wanted the fix — none of them needed to pay for it
Update on the cold-email experiment for AbleScan (a free accessibility scanner for small online shops).
The numbers, ~200 cold emails over 12 niches (coffee → ceramics → art prints): 3 real human replies (~1.5%), 0 sales.
What the three replies actually said:
- A chocolatier told me he'd fix his image alt text "soon now the summer is over."
- A ceramics maker asked, "can you tell me a bit more about this?" — I sent her all three fixes with copy-paste code, for free.
- A pottery studio said she'd pass it to her web support, and pointed out she doesn't sell in the EU and is a micro business, so she probably isn't covered by the European Accessibility Act at all.
What I think I'm learning (n=3, so treat this as a hypothesis):
- People do want the fix itself. Nobody was annoyed; everybody engaged with the actual code.
- My paywall never got tested. When someone replies, I've been giving away the whole answer, so there's no moment where €9.90 is the obvious next step. That's on me.
- My legal hook is weaker than I thought for UK micro shops. The pottery studio caught something I'd glossed over: "EAA compliance" only bites if you sell to EU consumers.
What I'm changing next: target more EU-facing shops, and route replies to the free scan report (all issues listed, one fix previewed) instead of handing over everything.
Also — I promised here I'd A/B test my CTA. It's running now: "reply and I'll send the list" vs. a one-line "worth sending?" yes/no. ~9 emails each, so I'll report it as anecdote, not statistics.
Question for anyone who's sold to small shops: when a stranger replies with interest, what made them actually pay?
Before reading too much into 1.5%, I'd check how many of the 200 actually reached an inbox. I sent 135 cold emails from a domain that was about a week old with no SPF, DKIM, or DMARC and got zero replies, and only set those up afterward, so I still can't tell how much of that zero was the copy. If you haven't already, confirming all three pass on your sending domain would make your CTA test a lot easier to trust.
3/200 is rough, but the fact that every human who replied wanted the fix is a cleaner signal than most cold-email "wins." Before you put a paywall in front of that CTA, I'd try a soft ask on the same thread: "happy to send it — curious what you'd expect to pay for something like this?" You keep the momentum and get pricing data from people who already raised their hand.