Skip to content

Implementation: Lead Dedupe Rules

Catch a duplicate lead before it hits your reporting.

The same person submits your form twice, calls after already filling it out, or gets caught by a chat widget and a form on the same visit. Left alone, that becomes three leads in your CRM instead of one. This workflow catches it before it does.

  • $0 setup, month to month
  • 30-day guarantee, pay only if satisfied
  • Free proposal within 1 business day

What this workflow covers

The matching key, the merge decision, and the edge cases.

The field map

Dedupe runs on a normalized key: the phone number stripped of dashes, parentheses, and country code, and the email address lowercased. Raw string matching on either field, without normalizing first, misses obvious duplicates that differ only in formatting, not identity.

Fuzzy identity match, not exact

A strict exact-match rule misses John Smith entered as J. Smith with the same phone number attached. The check has to weight phone and email far more heavily than name, since a name is the least reliable field for confirming the same person submitted twice.

Where the data enters

The check runs either natively inside the CRM, since many CRMs offer a built-in duplicate rule keyed on one or two fields, or through a middleware filter step that looks up the incoming email or phone before deciding whether to create a new record or update the existing one.

How failures surface

A rule that's too aggressive can merge two different people at the same address, a couple sharing a landline for example, overwriting one person's real inquiry with the other's. A rule that's too loose lets an obvious duplicate, like a typo'd email, slip through and inflate your lead count and cost per lead math.

Where a dedupe rule quietly gets it wrong.

What gets kept on merge matters as much as catching it

When two submissions merge, a decision has to be made about which one's source data wins, the original inquiry or the newer one. Defaulting to the newer submission's source quietly erases a legitimate first-touch attribution that was captured correctly weeks earlier.

A resubmission isn't always a mistake

Someone submitting the same form twice, a month apart, asking about a different service, looks like a duplicate by phone number but is actually a second, genuine inquiry. The rule needs a time window built in, not just a matching key, so old and new inquiries aren't collapsed into one record.

Chat and form submissions are the most common near-duplicate

A chat widget and a contact form on the same page frequently capture the same visitor within minutes, under a slightly different name spelling. Catching that specific pattern usually needs its own short time window rule, separate from the general dedupe logic applied to everything else.

We test against real messy data before shipping

Before this goes live, we run the rule against a sample of your actual historical leads, not a clean synthetic test case, because real names, phone formats, and typos behave in ways a tidy test never surfaces.

How we set it up.

01

Pull a sample of recent leads

We look at what real duplicates actually look like in your specific CRM before writing any matching logic.

02

Build the matching key and merge rule

We set the normalized key, the time window, and what happens to source data when two records merge into one.

03

Run it against history, then go live

We test the rule against past leads to check its behaviour before turning it on for new submissions coming in.

Duplicate lead dedupe rules FAQ.

Stop paying to generate the same lead twice.

We'll pull a sample of your real leads, build a matching rule around your actual data, and test it against history before it ever touches a live submission.

Keep exploring:

All implementation workflowsLead source capture, hidden fieldsLead routing by service areaGet a free proposal