The form promised a copy by email. Nobody was sending it.
My own funnel promised prospects a copy of their results and never sent one. The rebuild: a report email that has to convert alone — and actually reach the inbox.
My own lead-gen site has a marketing self-assessment on it — a short quiz that scores where a company stands and tells them what to fix first. Under the submit button, a friendly line: a copy of your results will be emailed to you.
This week, while chasing a junk submission, I read the actual code path. The prospect got nothing. No copy, no results, no email. The only message the system sent was a plain-text notification to me. The promise on the screen had never once been kept.
Found while looking for something else
That's the embarrassing, useful part: I wasn't auditing the funnel. A bot-ish lead had landed in my inbox and I went in to tighten the spam guards. Reading the submission handler end to end — the way I'd read a client's — is how the missing email surfaced. On client systems I've learned that the system you can see is rarely the system that matters; it turns out my own stack had earned the same skepticism.
An automated promise you don't keep is worse than no promise. The prospect just watched you not deliver — on the first interaction.
The constraint that shaped the whole build
Before writing a line, I fixed the design constraint: I don't chase these leads. No sequence, no sales calls out of the blue. Which means this one email is the last proactive touch the funnel gets — it has to carry the entire conversion on its own back.
So it became a real report, not a receipt: the overall score front and centre, per-area bars, a heatmap of the answers, and the three highest-leverage fixes ranked — genuinely useful even if we never speak. Booking CTAs sit under the score and again beside the levers, and the closing line explicitly hands the tempo to the reader. The same logic as selling the risk away when you have no numbers to show: if you only get one shot, load it with value and remove every reason to hesitate.
Speed is the other half of why the email must be automatic. The classic Harvard Business Review audit of 2,241 companies — The Short Life of Online Sales Leads — found firms responding within an hour were roughly seven times likelier to qualify a lead than those an hour slower, and sixty times likelier than those who waited a day. A report that arrives while the person is still sitting with their answers is a different product from the same report tomorrow.
An email that lands in spam doesn't exist
The unglamorous half of the work was deliverability. A transactional report from a small domain, sent by a shared web server, is exactly the mail that dies in a spam folder. Since Google's sender guidelines tightened, authentication isn't optional hygiene — SPF, DKIM and DMARC alignment are the price of entry to a Gmail inbox. So the site now sends through a proper email service over SMTP, DKIM-aligned on the sending domain, with delivery, bounces, opens and clicks visible in a log instead of vanishing into hope.
Then the loop got closed: the booking buttons route through a tracked redirect, so each click is logged against the specific lead. Not analytics theatre — when someone books, I want to know which report and which button did the work. It's the same discipline as letting six search impressions tell me what to write: small data, but mine, and causally connected to revenue.
The mechanics — the templating, the SMTP wiring, the redirect — an AI assistant built in an afternoon. What it couldn't decide is that the email must convert alone because I won't follow up, or that keeping a UI promise ranks above adding a feature. That's still the operator's half of the deal.
Check what your forms promise. Then check what they actually send. Mine were two different documents.
Sources & further reading
External
- Harvard Business Review: The Short Life of Online Sales Leads
- Google: Email sender guidelines (SPF, DKIM, DMARC)
Related posts