FlowsBook a callSign inCreate free accountSign up
Sales and prospecting

Keeping a prospect list free of duplicates, week after week

Duplicate prospects waste calls and embarrass reps. The four places repeats come from in a sales list, and the checks that stop each one before a lead reaches the team.

Part of the guide AI employee for sales prospecting

By the xMap team · · 5 min read

A prospect list stays free of duplicates when four checks run on every new batch before anyone sees it: one within the batch (the same business found twice by overlapping searches), one against earlier batches (the same lead found again next week), one against your customers and CRM (people you already sell to), and one between reps (two people chasing the same account). Miss any one of them and the same business reaches your team twice, usually at the worst moment.

This is for sales managers and sales operations people who feed reps a steady stream of new leads. Cleaning an old list once is a different job; this article is about keeping a flowing list clean, batch after batch. It covers where repeats come from, how to tell two leads are one business, a worked example and the flow.

Why duplicates in a live list cost more than in a file

A duplicate in a spreadsheet nobody uses is harmless. A duplicate in a live prospect feed means two reps call the same owner on the same morning, or a customer of three years gets a cold introduction. Owners talk to each other and to your existing customers; a team that seems not to know who it already knows loses credibility fast. Repeats also quietly inflate the numbers: a pipeline report that counts one bakery three times flatters the month.

The four places repeats come from

Where the repeat comes fromExampleThe check that stops it
Overlapping searches in one runA cafe found by Nearby places and again, spelled a little differently, by a Map business scanRemove duplicates, within the run
Earlier runsThe same new restaurant found again next MondayOnly new, across runs
Your existing customersA shop that already buys from you appears as a new opening after a refitMatch with my table or CRM lookup
Two repsOne branch of a chain found in each rep's areaSplit by territory, after the other checks

Each kind of repeat needs its own check; none of them catches all four.

How to tell two leads are the same business

Pick the comparison that matches how your leads are found:

  • From the map: use the place's own ID. xMap's places come with one, and two items with the same ID are the same place, however they are spelled.
  • From job postings: compare by company name, because one company may post the same role in several cities.
  • From several sources: compare by phone (its last nine digits, so country codes and spacing do not matter) or by website domain.
  • From your own systems: compare by the customer or account ID they share.

Then decide which copy to keep: the first, the best rated, or the one with the highest fit score from your AI scoring. For prospecting, the highest score is usually right, because it is the version the AI understood best.

A worked example: a payments provider in Johannesburg and Cape Town

This example is made up. A company selling card terminals to small shops and restaurants in Johannesburg and Cape Town runs two daily searches for newly opened places in five kinds of business, one in xMap's places database and one on online maps. Its reps complained of three things: the same restaurant appearing twice under two spellings, the same leads coming back the following week, and calls to merchants who already had a terminal from them.

  1. Within each day's batch, places found by both searches are merged by phone, keeping the copy with the better rating.
  2. Anything the flow has sent before is dropped, so only places never seen reach the reps.
  3. Each remaining place is looked up in the merchant table by phone, then by website; any match is dropped.
  4. What is left is split between the reps by area, so each lead belongs to exactly one person.
  5. Each rep gets a morning list; the team lead gets a count of how many repeats each check removed.

The count matters. If the customer check suddenly removes half the batch, the merchant table probably changed shape, such as a renamed phone column; if it removes nothing for weeks, it may not be matching at all.

How to do it with an AI employee

The order of the blocks matters: remove repeats before you spend effort on them, and split between reps last.

  1. Schedule and Area of interest: every weekday morning, your cities.
  2. Nearby places: the kinds of business you sell to, opened in the last 6 months; and beside it, from the same area, Map business scan for the same kinds.
  3. Remove duplicates, where the two searches meet: same phone, keeping the best rated.
  4. Only new: nothing this flow sent before.
  5. Match with my table on your customer table, by phone, keeping only the places not found; or CRM lookup against your CRM.
  6. xMap LLM: a fit score and a one-line reason.
  7. Split by territory: one owner per lead.
  8. Email: each rep's leads, with the rep's name on each line.

What lands in the inbox each morning is a list in which every business appears once, has never been sent before, is not already a customer and belongs to one named rep. The Prospects not yet in my CRM template starts with the customer check built in; Enrich a lead sheet and push it to HubSpot suits teams whose leads arrive in a sheet.

Habits that keep it clean

  • One source of truth for customers. Keep the table the flow checks against up to date; an old export lets last month's new customers through as leads.
  • Do not reset the memory casually. Only new remembers what each flow has found. A copy of the flow starts with an empty memory and will resend everything once.
  • Watch for near misses. A business that moved two streets away may come back with a new ID. A phone or website check against your own table still catches it.
  • Agree who owns a chain. Without a rule, every rep with a branch in their area will claim the head office.

Questions

What is the difference between Remove duplicates and Only new?

Remove duplicates looks within one run: it merges items that arrive twice in the same batch. Only new looks across runs: it drops anything the flow has already found before.

How are phone numbers compared?

By their last nine digits, so the same number written with or without a country code, spaces or dashes counts as one.

Can I skip businesses that are already in my CRM?

Yes. CRM lookup searches your CRM for each item through its own search and can let through only the ones not found; Match with my table does the same against a table or saved list in My data.

What happens to items with no phone or website?

Remove duplicates never merges items that have nothing to compare: they are all kept. With a single source, the place's own ID is the better check: Only new and Remove duplicates both use it by default.

Templates that do this

Read next

More on sales and prospecting · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.