Glossary
The vocabulary of forms that are supposed to make money
25 concepts, each with a definition, why it matters, how it is actually measured, and the mistake people make. Nothing here is only a definition — if we had nothing to add, we left the term out.
Corey Haines · San Diego · Aug 2026
5 of these are words we coined. They are marked Ours and carry a disclaimer on the page.
Most glossaries in this category are keyword pages with a paragraph attached. The rule for this one is that a page does not ship if all it does is define the term.
So every entry has to carry something the first search result does not: a number, a quote from somebody who ran into it, or a judgment we are willing to be wrong about in public. Several are built on the research corpus behind this site — around forty threads and a hundred and fifty reviews — which means the page on form abandonment can quote people describing it in their own words rather than citing a statistic nobody can source.
Where a term is ours, the page says so at the top. Presenting our own vocabulary as though the industry already used it would be a small version of exactly the dishonesty this site is about — see the dishonest dashboard.
Measurement
The numbers a form reports, and what each one can and cannot see.
Completion rate
Submissions divided by the people who could have submitted — and the denominator is the half nobody agrees on.
Form abandonment
Someone started your form and left — the only funnel metric most teams infer rather than observe.
Form drop-off analysis
Seeing exactly which question people stop at — the feature even competing vendors name as the one worth paying for.
Time to disposition
How long from submission until you know what the lead was worth — the number that decides whether outcome-based testing is possible for you at all.
Per-response pricing
Being billed by the submission — which makes spam a line item and turns a bot wave into a closed door for real buyers.
Form mechanics
How the form itself behaves, and what that behaviour costs.
Partial submission
What someone typed before they left. Both a recoverable lead and the raw material for every drop-off report you will ever run.
Conditional logic
Showing, hiding, skipping or routing based on answers — the category’s number one functional complaint, and the missing piece is a debugger.
Progressive profiling
Ask three fields now and three more next time — a trade of completion for enrichment that almost nobody measures.
Multi-step form
Splitting one form across several screens — the category’s most confident best practice, and one of its least evidenced.
Lead quality
What happens to a submission after it stops being a submission.
Cost per lead
Spend divided by leads — a metric that improves when your lead quality collapses.
Lead scoring
Ranking leads by attributes and behaviour — usually by scoring what the submitter told you about themselves.
MQL vs SQL
The handoff between marketing and sales, and the cheapest lead-quality instrument most companies already own and never use.
Data plumbing
How a lead gets from a browser to a CRM to an ad platform without losing the parts that matter.
Hidden field
An input the visitor never sees, carrying your attribution into the submission — and the first thing an automated filler reads.
Duplicate leads
The same person, twice — inflating your conversion count and, through it, the signal you send back to the ad platform.
GCLID
The click id Google appends to your landing page URL — the join key between your CRM and your ad account, and it is discarded by default.
Offline conversion import
Uploading what happened downstream back to the ad platform against the click id — a standard practice, and one that teaches the platform while teaching the form nothing.
Server-side tracking
Sending conversion events from your server rather than the visitor’s browser — more reliable measurement of a dataset that may be worse.
Conversions API
The server-to-server endpoint each ad platform provides for conversion events. A transport, not a truth.
Outcome webhook
A callback that tells the form what happened to a submission — the direction webhooks in this category almost never run.
Agents
What changes when the thing filling in your form is not a person.
WebMCP
A proposed browser standard letting a page offer tools an agent can call, instead of an agent pretending to be a person with a mouse.
Our vocabulary
Five words we made up. They are not industry standards and we are not going to pretend otherwise.
OriginOurs
Our name for the provenance stamp on every submission: Human, Agent, or Unverified.
VerdictOurs
Our name for the outcome written back onto a submission: Won, Lost, Disqualified, or Awaiting verdict — plus a value.
YieldOurs
Our name for the quality-adjusted metric: Yield rate is the share of submissions that reached a good verdict; Yield value is revenue per hundred submissions.
Hindsight split testsOurs
Our name for split tests ranked on Yield rather than completion rate — where the winner cannot be known at submit time, so the test waits.
ManifestOurs
Our name for the machine-callable declaration a form publishes alongside its human interface — one definition, two surfaces.
Where to start
If you are here because your leads are junk, the practical route is the anti-spam teardowns — twelve defenses taken apart, including the one that works and is not ours.
If you are here because the numbers look fine and the pipeline does not, start with completion rate, then cost per lead — a metric that improves when your lead quality collapses — and then the argument those two pages are building toward.
And if you would rather put your own numbers into it than read about it, the calculators do the arithmetic these pages describe.