← The blog
Guide · Prospecting

Qualifying a B2B Prospect Database, Line by Line

Qualifying a database means deciding which records deserve a call and documenting why, based on clear criteria, a set sequence of checks, target fit, and data freshness.

By Nicolas Morillot Co-founder of Scalon
The short answer

To qualify a database is to decide which records to call, verify, or set aside, and to write down why. For each record, check the contact details and your target, then note the decision, its reason, the source, and the date. To check whether a business location is still registered as active, use the Sirene directory, which Insee, France's national statistics office, says is updated daily (September 2026).

Get the list of businesses in your market, with the contact details available: tell us which one.

Get the list for my market

Published on October 23, 2026 · Updated on October 24, 2026

In this article

What does it mean to qualify a database?#

Qualifying a database means evaluating, verifying, cleaning, and enriching its data to make it reliable and usable, especially for prospecting. The term has no regulatory definition; it is used by sales teams and providers that qualify prospect lists.

This work is more than cleaning. It means deciding which records deserve a call, and writing down why.

Three actions often get mixed up.

Updating corrects what has changed, such as an email or a phone number. Enriching adds what is missing. Once the data is current and complete, qualification determines whether the record fits your target.

Which quality criteria should you check in a B2B database?#

Data quality is judged on a few simple criteria. The ISO/IEC 25012 model (a standard that describes data quality) lists several, including these three:

  • Accuracy: the value describes reality. The phone number on file is the one for the business location.
  • Completeness (the information is complete): every record has a value for all the expected fields. A mailing address without a postal code is missing information.
  • Consistency: for the same business location, the information matches from one record to the next. A company can have several business locations at different addresses.

Add one practical criterion: the contact details let you reach someone. Even correctly formatted contact details are useless if emails bounce or the business's main phone line is not taking calls.

Data reliability also depends on age. A B2B database that was accurate two years ago may not be anymore.

In what order should you check the contacts in a prospect list?#

In B2B prospecting, the order of checks matters: enriching a duplicate record means paying twice to research the same business. For a list, follow four steps.

Before any check, write down your target: business sector, area, size, revenue when it is public. Use these segmentation criteria to determine which records fit your sales target.

  1. Remove duplicates. Count business locations, not lines: compare the name, address and phone number to spot records that describe the same business location, and keep only one. Our guide shows how to deduplicate a database step by step. Before importing the list, also check it against your customer records and CRM (the software that tracks customer history).
  2. Verify the contact details. Address, phone, email: each contact must lead to the right place. Compare the phone number with the one on the business location's website, also checking the name and address, or call the number. Note what you did: "number found on the website" or "call reached the right business location." A number that leads elsewhere needs fixing. An unanswered call does not prove that the business location is closed; mark the record "to verify."
  3. Fill the gaps. Data enrichment fills the empty fields: business sector, headcount, revenue when it is public.
  4. Date it. Each record carries the date of its last check.

A record that fails a check has three possible outcomes. Either correct the record, flag it for verification with the source and keep it off the call list, or set it aside. Leaving it on the call list "for now" undermines subsequent marketing campaigns.

Here, "contacts" means the contact details of a business location, not the profile of a person.

Prospecting local businesses? Get the list for your market, your area and your target, ready to call.

Get the list for my market

Line-by-line qualification: does this record fit your target?#

A clean database is not a targeted database. Cleaning tells you whether a record is accurate; qualification tells you whether it fits your target.

Go back to the target you wrote down before the checks. Assign each record a decision and a written reason: call if it fits your target and its contact details are verified, verify if you need more information to decide, or set aside if it does not fit your target. The reason lets a colleague understand and check why the record was kept. A sales rep who reads "all-makes auto repair shop, vehicle lift visible" knows why they are calling.

Fictional example: you are targeting auto repair shops in your area that service all makes and have a vehicle lift.

  • Shop A: services offered and phone number verified, lift visible in workshop photos, source and date recorded. Call; reason: "all-makes auto repair shop, vehicle lift visible."
  • Shop B: services offered verified; no workshop photos available. Verify; reason: "equipment not confirmed."
  • Shop C: town outside your area. Set aside; reason: "out of area."

To apply this to your list, add four columns: Decision, Reason, Verification source, Checked on. For Shop B, that gives: Verify | Equipment not confirmed | Shop website: no photo of the workshop | [date of the check]. In "Verification source," describe what was checked so that "verified" means the same thing to everyone on the team.

Edge cases go to human review. Without this screening, sales reps waste calls on prospects that do not fit your target.

How do you check the freshness and origin of a piece of data?#

Information with no date and no source cannot be verified. Note on each record where it comes from and when it was collected. If two sources contradict each other on a criterion that matters for the decision, mark the record "to verify" and note both sources. It happens: in October 2026, Scalon compared 617 companies whose websites displayed their activity codes, mostly in the automotive sector, and found that about 1 in 10 displayed a code that differed from the one in the official directory (NAF code, APE code, the French activity classification codes).

For information about a business location, use Insee's Sirene database (the official directory of French businesses) as the authoritative source. It is updated daily and lists only two statuses: active or closed, according to Insee. A location listed as closed in the directory should therefore be removed from your database. An active listing does not prove that a location is still operating: it remains listed as active until Insee receives a closure filing. On October 24, 2026, among the business locations active in the Sirene directory that Scalon links to a Google listing, 1.1% have their listing marked "permanently closed" by Google, about 1 in 92. Before calling a location listed as "active", check that the location is really open.

For personal data, the GDPR (the European data protection regulation) goes further. Article 5 requires data to be "accurate and, where necessary, kept up to date." Keeping the database up to date is therefore not a nice-to-have.

At Scalon, we retain the source of every record: here is where the data we deliver comes from.

Need the list for your market? We send it with the contact details available, for the scope you choose.

Get the list for my market

Build a prospect database, buy a list, or commission a custom list?#

Whatever the origin of your prospecting data, the checks stay the same: quality criteria, order of checks, targeting, freshness.

Building a database yourself gives you control over the target, but it takes time from sales teams. For a purchased list, first check the source and the collection date. For data acquired from a third party, the CNIL (France's data protection authority) asks you to make sure the person was informed that the data might be used for prospecting and can object to it.

The third option is a custom prospect list, prepared for your target market and area and then kept up to date. Business locations are qualified before delivery, area by area, with human review of edge cases and updates included. No database is resold.

The Essentiel plan starts at €2,000 excl. VAT per year; setup fees are added in the first year. Ask for a quote that states the annual price, the setup fees, and the number of business locations to be included for your target market and area.

To assess the data for yourself, you can receive a sample of one hundred qualified business locations, free and with no commitment.

Method#

Scalon published this activity-code finding on October 15, 2026. On October 14, 2026, we reviewed the websites of 617 companies, mostly in the automotive sector, that displayed their APE code along with either a SIREN number (the French company ID) or a SIRET number (the French ID of a business location). We compared the displayed codes with those in the September 1, 2026 Sirene directory; 59 companies displayed a different code. The result holds for sites that display their code. Details in the article NAF code, APE code.

The closed-listing figure comes from Scalon's graph, which links business locations in the Sirene directory to their Google listings. We count locations active in the Sirene directory, in their latest version, that Scalon's graph links to a Google listing through a reliable ("green") link; only these reliable links count. A listing counts as closed when Google marks it "permanently closed"; a location linked to several listings counts only if all of them are. The links are those in the graph on October 24, 2026 at 9:19 a.m., after the new index was added (713,735 new reliable links since October 23). The Sirene data dates from September 1, 2026; the Google listings were collected between October 2025 and October 12, 2026 (median date: September 21, 2026), so a closure that Google already shows may not yet appear in that Sirene data. Only 5.3% of active locations are linked to a Google listing: a location with no linked Google listing is not counted. A listing whose status is unknown counts as not closed: the 43,953 locations linked only to listings with an unknown status are counted as not closed, which pulls the rate down. The measured share, 1.09%, is accurate to within 0.02 percentage points (95% confidence interval: 1.07% to 1.11%).

Sources#

The list for your market, your area and your target: we send it within two business days.

Get the list for my market