Checking B2B contact data freshness means confirming, field by field, that each email, phone number and job title has a dated check behind it, because a record's updated date only shows that something on it changed. A contact can look current in the CRM while the email bounces, the mobile reaches someone else and the title belongs to a previous job.
The method applies to any CRM contact base or vendor export where records carry dates or verification flags, and to lists bought before outreach. A sample tells you the list's freshness rate, not which unsampled record is stale, so every record you send to still needs its own check at send time.
What Data Freshness Means For A B2B Contact Record
US wage and salary workers had a median of 4.1 years with their current employer in January 2026, according to the Bureau of Labor Statistics, so every contact record ages toward a job change whether or not anyone edits it. Freshness therefore belongs to each field, and it holds only while a dated check stands behind it.
Records in B2B contact databases and CRMs carry several dates, and each one describes a different event. Four of them matter here:
- Created: The record entered a named system. It says nothing about whether the data is right today.
- Updated: A specified record or field changed. It does not mean every field was reviewed.
- Refreshed: A retrieval step ran, such as a sync from a vendor. The values may be unchanged and unchecked.
- Verified: A stated check ran on a stated value by a stated method. It covers that value and nothing else.
A recent updated date can sit on a record whose email has never been checked. One edited phone number moves the date for the whole record. A sync can also make old information look new, so keep the date a source observed a value separate from the date your system received it.
Job changes, email validity and phone ownership each need their own evidence. A job-change signal says nothing about a mailbox, and a deliverable email says nothing about a title. That is why the rest of this playbook works field by field.
Decay rates by field and segment are covered in the research on how fast B2B contact data decays. This guide is the operational check that follows from them.
How To Check B2B Contact Data Freshness, Step By Step
To check B2B contact data freshness, export the list with every date field, draw a random sample, re-verify each sampled email, phone and title against a named source, record the check date per field, and score the share of fields that still pass. The same six steps work on a CRM contact base or a vendor export.
- Export the list with every date and verification field, and note which event sets each date
- Draw a random sample in Google Sheets
- Re-verify each sampled email with a verifier and each mobile with a call, recording the status and reason
- Check title and employer against a dated job-change signal or the person's current public profile
- Record the source and date of each check, field by field
- Score the sample and decide whether to re-verify, suppress or keep each segment
1. Export every date field. Pull the created date, the last-modified date, any vendor refresh or enrichment date and every verification status into one sheet. Next to each date column, write which event sets it. A CRM's last-modified date records that some field changed, so it is an updated date, not a verified one.
2. Draw a random sample. Give every row a random number, sort by it and take the first rows. Fix the sample as values so it does not reshuffle. The formulas are in the sampling section below.
3. Re-verify contact channels. Run every sampled email through a verifier and keep the status and its reason in separate columns. Only a deliverable status counts as current. Bouncer, for example, returns deliverable, risky, undeliverable or unknown, each with a reason. For mobiles, the evidence is a call that confirms the right person answers. A number that simply appears in a vendor record is not evidence.
4. Check title and employer. Look for a dated job-change signal from a tool that tracks moves, or check the person's current public profile and note the date you looked. A title with no dated check behind it is unverified, however recent the record looks.
5. Log the source and date per field. For every field, write the value checked, the method, the source and the date. The UK Information Commissioner's Office advises recording the source of information you cannot check yourself, so a field with no source is flagged rather than trusted.
6. Score and decide. Count the share of sampled fields that passed, by field and by segment. Suppress records whose email came back undeliverable. Re-verify a whole segment before sending when its sample fails on a field you depend on. Keep records whose fields passed, with the new check date written to them. Set the line for each decision before you look at the results, based on what a bad send or a wasted call costs your team.
What Counts As Verification Evidence
Verification evidence for a contact field is a named check on that exact value with a stated method and date; Bouncer's documentation, for example, classes accept-all, full-mailbox and disposable addresses as risky rather than deliverable. Anything less, such as a recent sync or a filled-in field, shows the data exists, not that it is right.
What counts as freshness evidence, by field
| Field | Evidence that counts | Evidence that does not count |
|---|---|---|
| Deliverable verifier status, dated | Risky, unknown or undeliverable status | |
| Mobile | Call confirming the right person | Number present in a vendor record |
| Direct dial or company phone | Dated call reaching the person | Number carried from an old export |
| Job title | Dated job-change signal or re-verification | The record's updated date |
| Employer | Dated job-change signal or re-verification | An email domain on file |
The evidence rule: Each field passes only when you can name the check, the value it checked, the method and the date.
Email statuses show why the rule matters. In Bouncer's documentation, deliverable means the provider confirmed the address exists. Risky covers addresses that may bounce or show low engagement, with the reason low_deliverability (usually catch-all or a full mailbox) or low_quality (usually disposable). Undeliverable carries invalid_email, invalid_domain or rejected_email. Unknown covers DNS errors, unavailable SMTP, unsupported servers and timeouts. Only the first of those four is evidence that the address works.
For mobiles, Cognism describes a routine process that calls the numbers and confirms the right person picks up, and it marks those contacts as Diamonds. That is the kind of method the table asks for.
For titles and employers, a dated job-change signal is direct evidence that a contact moved. Human re-verification on a stated cycle also counts, such as SalesIntel's Human Verified tier, where researchers re-verify every detail every 90 days. The cycle is the vendor's statement, so log the date your own sample confirmed the title.
What Job-Change Dates Tell You And What They Miss
A job-change date tells you when a contact left the company on the record, and UserGems documents that it creates a new CRM record for the new company, links it to the old one and flags the old record as no longer at the company. It answers the title and employer question directly, which few other signals do.
In UserGems the sequence runs like this:
- A new record is created at the person's new company and connected to the old one, carrying over past information.
- The old record is flagged "No longer at company".
- The account owner is notified, by Slack or email.
- Automations can add the new contact to an Outreach or Salesloft sequence by type of job change.
- An optional automation writes the playbook name to a Salesforce field called UserGems Type.
- Records can be assigned by round robin or reassigned.
That makes the job-change date a good trigger for your own checks. When it fires, the old record's title and employer are known to be out of date, and the new record tells you where the person went.
The move event answers one question, who left and when. It says nothing about the person's new email or phone. The new record carries over past information, so treat every contact detail on it as unverified until its own check passes.
The old record needs a decision too. Flag it rather than delete it by default, so its history stays linked to the new record, and stop sending to its email. Erase what your purpose no longer needs.
How To Sample A Contact List For Recency
To sample a contact list for recency in Google Sheets, add a RANDARRAY column, use SORTN to take the first rows by that random value, and paste the sample as values so it stops changing when the sheet recalculates. Every record gets the same chance of selection, and the team gets a fixed set to re-check.
- Put the export on one tab, one contact per row, with a blank helper column on the right.
- In the helper column's first data cell, enter RANDARRAY. RANDARRAY(rows, columns) generates an array of random numbers between 0 and 1.
- On a new tab, use SORTN to return the first rows after sorting by the helper column. SORTN returns the first n items in a data set after performing a sort.
- Copy the result and paste it as values only. Google's documentation notes that, like RAND, RANDARRAY updates its results when you press enter, so a live sample keeps reshuffling.
Helper column, first data cell
=RANDARRAY({{rows}}, 1)
Sample tab, first cell
=SORTN({{data-range}}, {{sample-size}}, 0, {{helper-column-number}}, TRUE)
Then copy the sample and paste it as values only.Replace {{rows}} with the number of contact rows, {{data-range}} with the export range including the helper column, and {{helper-column-number}} with the helper column's position inside that range. The 0 tells SORTN to show at most the first {{sample-size}} rows.
There is no universal sample size. Choose {{sample-size}} to match how many records your team can re-check by hand, field by field, in the time it has. A sample nobody finishes tells you nothing.
Stratify when segments age at different speeds. Split the export by seniority, region or source, and run the same formulas on each segment's tab. Tenure is one reason segments differ. The same BLS release that puts the median for all wage and salary workers at 4.1 years gives 4.9 years for management, professional and related occupations. A list weighted toward one group will age differently from a list weighted toward another, and a single blended sample hides that.
What Each Tool Shows As Freshness Evidence
SalesIntel documents that its Human Verified contacts are re-verified by researchers every 90 days and its Email Verified contacts every six months, while Cognism marks phone-verified mobile numbers with a Diamond icon in its web app and Chrome extension. Each tool below exposes a different kind of evidence, in a different place, under different plan conditions.
This is a map of evidence, not a ranking. Teams comparing vendors can start with the guides to the best B2B contact databases and the email verification tools worth shortlisting.
SalesIntel
SalesIntel labels each contact by how it was verified, and by default users see the Human Verified and Email Verified contacts.
Freshness evidence: Human Verified contacts have every detail re-verified by researchers. Email Verified contacts have their email verified by the research team. Machine Verified contacts are verified by a machine-learning engine.
Re-check cadence: Every 90 days for Human Verified contacts and every six months for Email Verified contacts, according to SalesIntel's help centre.
Permissions or plan: Each contact returned from a research request is charged one research credit. Re-verifying Human Verified contacts is not charged.
Real cost driver: Research credits drive the spend. Reporting incorrect information and asking for an update does not cost a research credit, so sending your failed checks back to SalesIntel adds nothing to the bill.
Limitation: The tier is a cycle promise, not a dated check on the field you are about to use. Log your own check date when a sampled field passes, and check Machine Verified contacts before use, since no researcher cycle applies to them.
Cognism
Cognism marks contacts whose mobile number has passed a phone check as Diamonds.
Freshness evidence: A Diamond is a contact whose mobile number has been phone-verified. Cognism's routine process calls the number and confirms the right person picks up.
Re-check cadence: Cognism states that 95% of director-level and above contacts are refreshed every 30 days. The claim covers that seniority band only.
Where evidence appears: A Diamond icon shown before the number in contact info, in the web app and the Chrome extension.
Permissions or plan: Diamonds-on-Demand lets users request phone verification for specific contacts, with a stated turnaround under 48 hours. The Diamond package is required to access these features.
Real cost driver: The Diamond package is the cost to plan for when your sample shows mobiles failing, since on-demand phone verification sits inside it.
Limitation: The Diamond icon is a pass flag on one field. It does not speak for the email or the title, and the 30-day refresh claim does not cover contacts below director level.
UserGems
UserGems tracks job changes rather than verifying contact channels.
Freshness evidence: A new CRM record at the person's new company, linked to the old one, and the old record flagged "No longer at company".
Where evidence appears: In the CRM, with owner notifications by Slack or email, and an optional Salesforce field, UserGems Type, that holds the playbook name.
Permissions or plan: Playbooks, filters and automations are configured with your UserGems customer success manager.
Real cost driver: Plans are metered in credits, spent as data actions or AI actions. Beyond the subscription, UserGems lists a one-time implementation fee of $3k, $5k or $10k depending on plan, and an optional one-time fee of $3k for Retroactive Job Change Leads.
Limitation: It dates the move, not the new contact details. The new email and phone still need their own checks.
Bouncer
Bouncer verifies email addresses and returns a status with a reason for each one.
Freshness evidence: Four statuses (deliverable, risky, undeliverable and unknown), each with a reason, plus flags such as domain.acceptAll, account.fullMailbox, domain.disposable and account.role.
Re-check cadence: Batch Sync results are kept for up to 24 hours. After that, the email is verified again and credits are charged.
Where evidence appears: In each result's status, reason and flags, which you copy into the sheet or CRM beside the address.
Permissions or plan: API calls authenticate with an API key, sent in an x-api-key header or by basic auth, and the key is generated in the dashboard. Real-time verification defaults to 1,000 requests per minute and Batch Sync to 100 per minute.
Real cost driver: Credits per verification, including the re-verification that follows once a Batch Sync result passes 24 hours. Store results as they arrive rather than querying again.
Limitation: A verifier result is a check on one address at one moment. It cannot prove that a mailbox behind a catch-all domain exists.
Google Sheets
Google Sheets holds the sample and the evidence log for every other tool.
Freshness evidence: None of its own. It holds the evidence the other tools produce, with columns per field for the value, status, source and check date.
Where evidence appears: In the evidence log you build, next to each sampled row.
Permissions or plan: Connecting a sheet to Bouncer through Zapier needs a Zapier account, access to the spreadsheet and a connected Bouncer account.
Limitation: RANDARRAY updates its results when you press enter, so a sample left as formulas changes under you.
Freshness evidence each tool exposes
| Tool | Evidence it exposes | Fields it covers |
|---|---|---|
| SalesIntel | Verification tier per contact | Every detail or email |
| Cognism | Diamond icon | Mobile number |
| UserGems | Dated job-change record | Title and employer |
| Bouncer | Status, reason and flags | Email address |
| Google Sheets | Your dated evidence log | Whatever you log |
How Often Should You Re-Check B2B Contact Data
Re-check each B2B contact field on its own clock, verifying email at send time, title and employer whenever a job-change signal fires, and any field older than the vendor's stated re-verification cycle for its tier before it goes into a sequence. A vendor's refresh cycle sets an upper limit on age; it does not replace your own check.
Re-check triggers and limits by field
| Field | Trigger | Maximum interval | Evidence |
|---|---|---|---|
| Every send | Verifier's result lifetime | Dated deliverable status | |
| Mobile | Every calling campaign | Campaign start | Call reaching the person |
| Job title | Job-change signal | Vendor's stated tier cycle | Dated signal or re-verification |
| Employer | Job-change signal | Vendor's stated tier cycle | Dated signal or re-verification |
Stored email results expire. Bouncer keeps Batch Sync results for up to 24 hours and verifies the address again, for new credits, after that. Treat any stored deliverable status older than your verifier's window as a prompt to check again before the send.
For titles and employers, use the vendor's stated cycle as a ceiling, not a guarantee. SalesIntel states 90 days for Human Verified contacts and six months for Email Verified ones. Cognism's 30-day refresh claim covers director-level and above contacts only. A contact outside a stated scope has no vendor cycle to lean on, so your own sample sets its interval.
Tie the interval to the cost of an error. A stale mobile wastes a rep's call. A stale email list puts sending reputation at risk, and Google publishes spam-rate thresholds that senders to Gmail are asked to stay under, covered in the measures below. The higher the cost, the shorter the interval, and the closer to send time the check should run.
How To Build A B2B Contact Data Workflow That Keeps Records Fresh
A B2B contact data workflow keeps records fresh when every new or changed row triggers a check, as in Zapier's template that sends each new Google Sheets row to Bouncer's Verify Email action, and writes the result, source and date back to the record. Each step below names the access and setup it needs.
- Verify new rows on arrival. The template's trigger, New Spreadsheet Row, fires when a row is added to the bottom of a sheet, with the spreadsheet and worksheet as inputs. Its action, Bouncer Verify Email, checks the row's email address. It needs a Zapier account, access to the spreadsheet and a connected Bouncer account. Add new and changed records as new rows so they reach the trigger.
- Call Bouncer directly at volume. Bouncer's API authenticates with an API key in an x-api-key header or by basic auth, generated in the Bouncer dashboard. Real-time verification defaults to 1,000 requests per minute and Batch Sync to 100 per minute, so pace large lists to those limits. Batch Sync results are kept for up to 24 hours, so write them to the record before they expire.
- Route job changes into the CRM. UserGems creates the new record, flags the old one, notifies the owner and can enrol the new contact in an Outreach or Salesloft sequence. The playbooks, filters and automations behind this are configured with your UserGems customer success manager.
- Request vendor re-verification where the plan allows. In SalesIntel, reporting incorrect information and asking for an update costs no research credit, and re-verifying Human Verified contacts is not charged. In Cognism, Diamonds-on-Demand requests phone verification for specific contacts, with a stated turnaround under 48 hours, and needs the Diamond package.
- Write the check back to the record. For each field, store the value checked, the method, the source and the date. Recording the source of anything you cannot check yourself is also what the ICO's accuracy guidance advises.
What To Measure After A Freshness Check
After a freshness check, measure the share of sampled emails that came back deliverable, the share of contacts who changed jobs, and the share of fields with a dated source, then watch sending health against Google's guidance to keep spam rates below 0.1% and never reach 0.3%. Together, those four numbers show whether the list is safe to work.
Measures to track after each check
| Measure | How to calculate | What it tells you |
|---|---|---|
| Email status mix | Each status divided by emails sampled | Share safe to send |
| Job-change share | Contacts who moved divided by sample | How far titles drifted |
| Sourced-field share | Fields with dated source divided by fields sampled | How much is evidenced |
| Spam rate | Daily rate in Postmaster Tools | Whether list quality hurts sending |
Split the email status mix into deliverable, risky, unknown and undeliverable, because risky and unknown are not passes. Count catch-all addresses inside risky, not inside deliverable.
The job-change share tells you how much of the sample has a stale title and employer. Read it by segment, since a single blended figure hides the segments that age fastest.
The sourced-field share measures evidence, not accuracy. A low share means the list has fields nobody can vouch for, which is a reason to re-verify before the next send.
Google calculates and updates spam rates daily in Postmaster Tools, so the downstream measure moves soon after a send. Track bounce rate beside it in your sending tool. List quality matters here because email warmup alone cannot fix deliverability.
Set no universal target for the first three measures. Compare each check with the last one on the same segment. A falling deliverable share or a rising job-change share is the signal to shorten that segment's re-check interval.
Does GDPR Require You To Keep Contact Data Up To Date
Under Article 5(1)(d) of the GDPR, personal data must be accurate and, where necessary, kept up to date, according to the UK Information Commissioner's Office, which also advises recording the source of data you cannot check yourself. Whether a record must be kept current depends on what you use it for, and outreach depends on reaching the right person.
The ICO's guidance adds practical points. Every reasonable step must be taken to erase or rectify inaccurate data without delay. If mail comes back undeliverable, you should update the record. Where you cannot check a value, record its source accurately. The ICO notes that this guidance is under review.
This section is general information, not legal advice.
Frequently Asked Questions
Does an updated date mean the contact was verified?
No. An updated date shows that some field on a contact record changed, not that every field was checked. A sync or a single edit can move it. Read freshness from the verification date of each field instead, and treat a field with no dated check behind it as unverified, however recent the record looks.
Is a catch-all email address verified?
No. A catch-all domain accepts mail for any address, so a verifier cannot confirm that the specific mailbox exists. Bouncer, for example, returns these addresses as risky with the reason low_deliverability and flags the domain as accept-all. Count only a deliverable status as verification evidence, and keep catch-all addresses out of sequences until another signal supports them.
Can a vendor's refresh cycle replace your own freshness checks?
No. A refresh cycle holds only at its stated scope. Cognism's 30-day refresh claim, for example, covers director-level and above contacts, and it is the vendor's own figure. Use a vendor cycle as an upper limit on age, then run your own sample and send-time checks, because a cycle says nothing about the record in front of you.
Should you delete a contact who changed jobs?
Not by default. Flag the old record as no longer at the company and stop sending to its email. Job-change tools such as UserGems link the old record to a new one at the new company, so the history carries over. Verify the new record's email and phone before outreach, and erase old data your purpose no longer needs.
Sources
- Employee Tenure in 2026 · bls.gov
- unifygtm.com
- Email sender guidelines FAQ · support.google.com
- support.salesintel.io
- cognism.com
- Principle (d): Accuracy · ico.org.uk
- help.cognism.com
- Add new Google Sheets rows and verify email addresses with Bouncer · zapier.com
- salesintel.io
- salesintel.io



