> Enrich.chat gives AI agents verified work emails and mobiles. To use it from an agent, read https://enrich.chat/agents.md . This is the Markdown version of https://enrich.chat/blog/b2b-data-decay/

# How fast does B2B contact data go bad? Decay rates, job changes and how often to refresh

Most estimates say about a quarter of B2B contacts go stale each year. Here is what each published number measures, how fast jobs, emails and phones change, and how to set a refresh schedule.

By Enrich.chat Editorial. Updated October 15, 2026.

**Short answer:** The most quoted B2B data decay rate is 22.5% a year, or about 2.1% a month, from [HubSpot's database decay page](https://www.hubspot.com/database-decay). Published figures run from about 12% to 70% depending on what counts as a change. [BLS](https://www.bls.gov/news.release/tenure.nr0.htm) found that 20.6% of US wage and salary workers had been with their employer a year or less in January 2026.

## Key takeaways

- The most quoted rate is 22.5% a year (2.1% a month) from HubSpot's database decay page, which credits MarketingSherpa research and shows no method.
- Published yearly figures run from 12.25% (Lusha, job changes of US sales leaders) to about 67% (Cleanlist, any change to four fields, checked weekly) and 70.8% (an informal card survey), because each study counts different changes.
- BLS: 20.6% of US wage and salary workers had 12 months or less with their employer in January 2026. The share is 25.5% at ages 25 to 34 and 9.2% at ages 55 to 64.
- Fields age at different speeds. Job titles change fastest in both field-level studies. Work emails follow the job (ZeroBounce found at least 23% of 2025 email data invalid or risky), and the FCC counts about 35 million disconnected numbers made available for reassignment each year.
- Plan on 3 to 5 in every 100 contacts changing employer each quarter. Re-check emails before each send, and measure your own rate with a random sample of 200 records.

## How fast does B2B contact data go bad?

The most quoted answer is 22.5% a year, or about 2.1% a month. [HubSpot's database decay page](https://www.hubspot.com/database-decay) gives that rate and credits it to MarketingSherpa research. The page shows no method, and it does not link the study.

Other sources give other numbers, and they do not count the same thing. The table lists each published figure with what it measures. We read every source on October 15, 2026.

| Figure and source | What it measures | Read it as |
| --- | --- | --- |
| **22.5% a year** (2.1% a month) [HubSpot](https://www.hubspot.com/database-decay), citing MarketingSherpa | Decay of a B2B database, method not shown | A rule of thumb |
| **25% to 30% a year** [ZoomInfo blog](https://pipeline.zoominfo.com/operations/poor-data-quality-impact), updated July 7, 2026 | Share of CRM records that go stale, no study named | A vendor benchmark |
| **At least 23% a year** [ZeroBounce](https://www.zerobounce.net/email-list-decay), 2026 report | Share of verified emails graded invalid or risky, more than 11 billion addresses in 2025 | Email only, a low-end figure |
| **20.6%** [BLS](https://www.bls.gov/news.release/tenure.nr0.htm), January 2026 | US wage and salary workers with 12 months or less at their employer | Counts workers, not contacts |
| **12.25%** [Lusha](https://www.lusha.com/blog/b2b-data-decay-rate-measured/), September 2026 | US sales VPs and C-suite leaders with a recorded job change in 12 months | Job moves only, a floor |
| **About 67% a year** [Cleanlist](https://www.cleanlist.ai/blog/2026-04-07-state-of-b2b-data-quality-2026), April 2026 | Any change to title, company, email or phone, re-checked weekly for 13 weeks on 5,000 contacts | One CRM, yearly figure is a projection |
| **70.8%** [IndustrySelect](https://industryselect.com/blog/b2b-data-decay-and-its-impact-on-marketing), undated | Business cards with at least one change in 12 months, from 1,025 seminar attendees | Informal, self-selected sample |

Read the table as a range, not a verdict. Most of these numbers come from companies that sell contact data, email checking or CRM software. That does not make them wrong. It does mean you should treat them as benchmarks, not as facts about your own list.

## Why do published decay rates range from 12% to 70%?

The rates differ because each study counts a different kind of change, over a different window, for a different group. Four differences explain most of the gap.

- **What counts as a change.** [Lusha](https://www.lusha.com/blog/b2b-data-decay-rate-measured/) counts only job changes its system detected. [Cleanlist](https://www.cleanlist.ai/blog/2026-04-07-state-of-b2b-data-quality-2026) counts a record as changed if its title, company, email validity or phone connectivity changed. More fields mean a higher rate.
- **How often you look.** Cleanlist re-checked its contacts every week for 13 weeks and measured 2.1% decay a week, which it projects to about 67% a year. Its report contrasts that with the 30% figure often cited, which it says came from annual snapshots, not weekly tracking.
- **Who is measured.** BLS counts workers, not contacts. ZeroBounce counts addresses that its customers submitted, and the lists differ each year. The [IndustrySelect survey](https://industryselect.com/blog/b2b-data-decay-and-its-impact-on-marketing) collected business cards from seminar audiences.
- **Whether anyone can trace it.** ZoomInfo's range names no study. The often-quoted figure of up to 70% traces to a [2011 MarketingSherpa article](https://www.marketingsherpa.com/article/how-to/data-hygiene-behavioral-analysis-more), which said a marketing database could lose as much as 70 percent in a year and attributed the point to a Salesforce product marketer. That is an upper-bound opinion, not a measurement.

[Scalelist's review](https://scalelist.com/research/data-decay/) of the published numbers found no study that follows one fixed group of contacts for 12 months across employer, title, email and phone. Scalelist also sells contact data, so check its reading against the sources above. The closest match is Cleanlist's 13-week re-check of one CRM.

### How do you turn a yearly decay rate into a monthly rate?

Compound it. A yearly decay rate of 22.5% leaves 77.5% of contacts valid. The monthly rate is 1 minus the twelfth root of 0.775, which is about 2.1%. The same arithmetic runs in reverse: 12 months at 2.1% leaves 0.979 to the power of 12, or 77.5% of contacts valid.

The period matters as much as the rate. A decay rate of 2.1% a month gives 22.5% a year. A decay rate of 2.1% a week gives about 67%. That gap is why the HubSpot and Cleanlist numbers look so far apart even though both say 2.1%.

**Figure 1: What a yearly decay rate does to 1,000 contacts.** Line chart of valid contacts left from a list of 1,000 that is never refreshed, at three yearly decay rates. At 12.25% a year, 878 are left after 12 months, 770 after 24 and 676 after 36. At 22.5% a year, 775, 601 and 465 are left. At 30% a year, 700, 490 and 343 are left. At 22.5% a year, half the list is stale after about 2.7 years.

*Valid contacts left from a list of 1,000 that is never refreshed. This is illustrative arithmetic, not a measurement. The rates are the 12.25% job-change rate [Lusha](https://www.lusha.com/blog/b2b-data-decay-rate-measured/) recorded for US sales leaders, the 22.5% rate on [HubSpot's page](https://www.hubspot.com/database-decay), and the 30% top of [ZoomInfo's range](https://pipeline.zoominfo.com/operations/poor-data-quality-impact).*

Figure 1 shows the effect for a list you never refresh. At 22.5% a year, a list of 1,000 contacts has about 775 valid after a year, 601 after two years and 465 after three. Use your own measured rate instead of these three, as the later section on sampling explains.

## How often do people change jobs?

About one in five US workers started with their current employer in the past year. The [BLS Employee Tenure release](https://www.bls.gov/news.release/tenure.nr0.htm) of September 24, 2026 found that 20.6% of wage and salary workers had 12 months or less of tenure in January 2026. Median tenure was 4.1 years, up from 3.9 years in January 2024.

That 20.6% is wider than voluntary job moves. BLS says the group includes new hires, workers who lost a job and found another, and workers who changed employers by choice. Treat it as an upper bound on job-to-job changes.

**Figure 2: US workers with one year or less at their employer, by age.** Bar chart from the BLS Employee Tenure release for January 2026. Of all wage and salary workers, 20.6 percent had 12 months or less with their current employer. By age, the share is 25.5 percent at 25 to 34, 15.4 percent at 35 to 44, 12.2 percent at 45 to 54 and 9.2 percent at 55 to 64.

*Share of US wage and salary workers with 12 months or less of tenure with their current employer, January 2026. Younger workers change employers most often. Source: [BLS, Employee Tenure in 2026, Table 3](https://www.bls.gov/news.release/pdf/tenure.pdf).*

Age matters most. In January 2026, 25.5% of workers ages 25 to 34 had 12 months or less with their employer, against 9.2% of workers ages 55 to 64. Role matters too. BLS puts median tenure at 6.1 years in management occupations and 4.9 years for management, professional and related occupations as a whole.

A second BLS survey tells the same story from the employer side. The [JOLTS release of March 13, 2026](https://www.bls.gov/news.release/archives/jolts_03132026.htm) reports 38.0 million quits in 2025, an average monthly quits rate of 2.0%. Twelve months at that rate add up to about 24 quits per 100 jobs. Some people quit more than once, and the count is of jobs, not people. Total separations, which add layoffs and discharges, ran at 3.3% a month, or 62.8 million for the year.

Job changes also slow with age. The [BLS National Longitudinal Survey of Youth 1979](https://www.bls.gov/news.release/nlsoy.nr0.htm) followed people born from 1957 to 1964. They held 12.9 jobs on average from age 18 to 58. They held 5.6 jobs at ages 18 to 24 and 1.3 at ages 55 to 58, and a job that spans two age groups counts in both.

Contact data vendors see a narrower group. [Lusha](https://www.lusha.com/blog/b2b-data-decay-rate-measured/) counted job changes recorded in its own database for US sales VPs and C-suite leaders. Its September 12 run found 12.25% over 12 months (12.6% in August) and 25.67% over 24 months. UK sales leaders changed roles at 21.40% over 12 months. Lusha says its numbers are a floor, because it counts only the changes it detected. Read together, these sources suggest a list of senior leaders ages more slowly than a list of junior staff, but no list is safe.

## Which contact fields go bad fastest?

Job titles go stale fastest in both published field-level studies. Employers and work emails follow, because both are tied to the job. Mobile numbers look the most stable, but only one study measured them. After the title, the order depends on what each study counts.

| Field | Cleanlist, per year | IndustrySelect, cards changed |
| --- | --- | --- |
| Job title | 43.6% | 65.8% (title or function) |
| Company | 30.6% | 34.2% (company name) |
| Work email | 18.8% (email validity) | 37.3% (email address) |
| Office phone | 14.5% (direct dial) | 42.9% (phone number) |
| Mobile phone | 5.0% | Not measured |

The two studies disagree because they ask different questions. [Cleanlist](https://www.cleanlist.ai/blog/2026-04-07-state-of-b2b-data-quality-2026) re-checked one CRM weekly with its own tools, counted a work email as changed when it stopped validating, and annualized the weekly rates. [IndustrySelect](https://industryselect.com/blog/b2b-data-decay-and-its-impact-on-marketing) asked 1,025 seminar attendees whether any part of their business card had changed. Treat both as a guide to order, not to exact rates.

### Job titles and employers

Titles change inside a company as well as between companies, so a title can go stale while the person stays put. A move to a new employer breaks more than the title. The work email and the office phone are tied to the old job.

### Work emails

A work email lives and dies with the job. [ZeroBounce's 2026 report](https://www.zerobounce.net/email-list-decay) says at least 23% of the email data it verified in 2025 was invalid or risky, down from 28% in 2024. It reviewed more than 11 billion addresses. The report names career moves and company restructurings as causes, and it gives no B2B-only figure.

ZeroBounce calls 23% a low-end estimate. Catch-all domains accept every address, so those emails cannot be confirmed without sending to them. More than 9% of the emails it checked were catch-all. You can test a domain with the free [catch-all checker](https://enrich.chat/tools/catch-all-checker/).

### Phone numbers

A phone number can go bad in a worse way: it can end up with a stranger. The FCC's [2018 order creating the Reassigned Numbers Database](https://docs.fcc.gov/public/attachments/FCC-18-177A1.pdf) says approximately 35 million numbers are disconnected and made available for reassignment each year. The FCC drew that count from numbering administrator reports for 2013 to 2016 and noted that one party questioned it. The order also set a minimum 45-day wait before a disconnected number can be reassigned.

In Cleanlist's data, mobile numbers were the most stable field, at 5.0% annualized against 14.5% for office direct dials. If you call or text, a reassigned number puts you in front of someone who never agreed to hear from you. Our guides to [TCPA and B2B cold calling](https://enrich.chat/blog/tcpa-b2b-cold-calling/) and [direct dials](https://enrich.chat/blog/direct-dials/) cover the rules, and the free [phone number validator](https://enrich.chat/tools/phone-number-validator/) shows a number's type.

### Company data

Companies close, merge and rename. BLS tracks private-sector establishments by the year they opened. Of those that opened in the year to March 2015, [79.6% were still open after one year, 50.2% after five and 34.7% after ten](https://www.bls.gov/bdm/us_age_naics_00_table7.txt). BLS counts work locations, not whole companies, so read the table as a guide to small and new businesses. Our guide to [firmographic data](https://enrich.chat/blog/firmographic-data/) covers company records in depth.

## What does outdated contact data cost you?

Outdated data costs you in bounces, calls to the wrong person and wasted rep time, and in some places it conflicts with privacy law. Email bounces are the easiest cost to measure.

A hard bounce is an email sent to an address that does not exist, as [Amazon SES's documentation](https://docs.aws.amazon.com/ses/latest/dg/reputationdashboardmessages.html) defines it. Amazon SES tells senders to keep the bounce rate below 5%. It places an account under review at 5% or more and may pause sending at 10% or more. Those thresholds belong to one provider, and others set their own.

The margin is thin. [Hunter's 2025 report](https://hunter.io/the-state-of-cold-email), which covers 31 million emails sent by its users, puts the average bounce rate at 3.6%. That average sits only 1.4 points below Amazon SES's review line. Our guide to the [email bounce rate](https://enrich.chat/blog/email-bounce-rate/) explains what to aim for, and the [cold email guide](https://enrich.chat/blog/cold-email/) covers list prep.

There is a legal side too. The UK Information Commissioner's Office explains that [Article 5(1)(d) of the UK GDPR](https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/data-protection-principles/a-guide-to-the-data-protection-principles/accuracy/) requires personal data to be accurate and, where necessary, kept up to date, and inaccurate data to be erased or rectified without delay. The ICO says whether you must keep data current depends on what you use it for. If you rely on a record's current details, as with payroll or delivery addresses, you should update it. The ICO also says to record the source of data you cannot verify. This is not legal advice, and keeping data fresh does not by itself make any use of it compliant.

## How often should you refresh B2B contact data?

Re-check work emails before every send, re-check title and employer every quarter for accounts you actively work, and refresh the whole record once a year. This is a schedule we suggest, built from the rates above. It is not an industry standard.

The quarterly step comes from the job-change rates. If 12.25% of US sales leaders (Lusha) and 20.6% of all US workers (BLS) change employers in a year, about 3 to 5 in every 100 contacts move each quarter. A quarterly check catches most moves before they pile up.

**Figure 3: A suggested refresh schedule for B2B contact data.** A 12-month calendar with four rows. Work email status is checked before every send. Title and employer are checked every quarter, in months 3, 6, 9 and 12. The whole record is refreshed once a year, in month 12. Any bounce or reply saying the person has left triggers an immediate re-check of that record.

*A suggested refresh schedule, built from the job-change rates above. It is guidance, not an industry standard. Change the gaps to match the decay you measure in your own CRM.*

### Refresh by value, not by date

Check the records that can earn money first: open deals, named accounts and contacts about to enter a sequence. Low-value records can wait for the yearly pass.

### React to signals

A hard bounce, an auto-reply that says the person has left, or a different title in an email signature all mean the record is stale now. Suppress the old address, then look the person up again. Our guide to [data enrichment](https://enrich.chat/blog/what-is-data-enrichment/) explains how those lookups fit into a CRM workflow.

### Date-stamp every field

Store a last-checked date and a source for each field, not just for the record. A title checked last week and a phone number checked two years ago need different treatment. The ICO's accuracy guidance also tells you to record the source of data you cannot verify.

## How do you measure decay in your own CRM?

Measure it with a sample. Published rates vary too much to use for budgeting, so check a random slice of your own database and count what changed. Five steps do it:

1. Pull 200 contacts at random, across old and new records.
2. Re-check each one field by field: title, employer, work email status and phone. Use a fresh lookup, not your own copy.
3. Count the contacts where anything changed, and count each field on its own.
4. Divide the changed contacts by 200. If 45 changed, your rate is 22.5% for the age of those records.
5. Repeat each quarter to see whether your schedule is tight enough.

Two cautions apply. A sample of 200 gives a margin of about 6 points either way at a 22.5% rate. And the result describes the age of the records you sampled. To get a yearly rate, sample records that were last checked about a year ago. For a step-by-step approach to the re-check itself, see [CRM data enrichment](https://enrich.chat/blog/crm-data-enrichment/).

## How can an API or an AI agent keep contact data fresh?

You re-run a lookup on the records you choose, compare what comes back with your copy, and update what changed. The Enrich.chat [person enrichment API](https://enrich.chat/api/person-enrichment/) does this with one request per person. It returns the title, the company and a work email with a status of `verified`, `catch_all` or `unknown`, taken from our records. Treat each status as a strong lead, not a guarantee.

Send a LinkedIn URL when you have one, an email when you have only that, or a name with a company website. Enrich.chat never logs in to LinkedIn, and the URL is only a lookup key, as the [LinkedIn to contact API](https://enrich.chat/api/linkedin-to-contact/) page explains. A lookup that finds nothing returns `found: false` and uses 0 credits. Each matched response carries `credits_used`.

```
curl -G "https://enrich.chat/v1/person" \
  --data-urlencode "linkedin_url=https://www.linkedin.com/in/jane-doe-ops" \
  --data-urlencode "fields=email"

{
  "found": true,
  "title": "VP Operations",
  "company": {"name": "Acme Logistics"},
  "work_email": {"value": "jane.doe@acmelogistics.com", "status": "verified"},
  "credits_used": 10,
  "free_lookups_remaining": 9
}
```

The free tier gives 10 lookups per IP address per day, with no signup and no key, which is enough to test the 200-contact sample from the last section. Check the [pricing page](https://enrich.chat/pricing/) for the monthly plans. As a workload, re-checking the work emails of 5,000 contacts once a quarter uses 50,000 credits in the month you run it if every lookup matches. That fits the Growth plan at $29 a month. Credits renew with each monthly payment, and unused credits do not roll over.

```text
Use www.enrich.chat to re-check the 200 contacts in my sample list. For each LinkedIn URL, look up the current title, company and work email status, flag every contact whose title or company differs from my copy, and tell me the total credits_used
```

An AI agent can run this loop for you. Point it at [the agent setup page](https://enrich.chat/agents/) once, then ask for the re-check in plain English. For a scheduled job outside an agent, [n8n](https://enrich.chat/blog/n8n-lead-enrichment/) and similar automation tools can call the same endpoint, and the [waterfall enrichment](https://enrich.chat/blog/waterfall-enrichment/) guide shows how to combine more than one source.

### What Enrich.chat does not do

- It does not monitor people or send job-change alerts. It answers when you ask, so you choose the schedule. If you need an alert the day a champion changes jobs, use a tool built for monitoring.
- It is not a CRM plug-in. You call it from an agent, a script or an automation tool, and you write the results back yourself.

## Frequently asked questions

### What is B2B data decay?

B2B data decay is the loss of accuracy in contact and company records as people change jobs, companies rename or close, and emails and phone numbers are retired. The record stays the same while the world moves on. Published yearly rates run from about 12% to 70%, depending on what is counted. Our guide to [CRM data enrichment](https://enrich.chat/blog/crm-data-enrichment/) covers how teams repair it.

### What is the monthly B2B data decay rate?

The most quoted monthly rate is 2.1%. [HubSpot's database decay page](https://www.hubspot.com/database-decay) credits it to MarketingSherpa research and annualizes it to 22.5%. [ZoomInfo's blog](https://pipeline.zoominfo.com/operations/poor-data-quality-impact) gives 25% to 30% a year instead. [Cleanlist](https://www.cleanlist.ai/blog/2026-04-07-state-of-b2b-data-quality-2026) measured 2.1% a week on one CRM, which compounds to about 67% a year. The rate depends on what counts as a change.

### How often do people change jobs?

[BLS](https://www.bls.gov/news.release/tenure.nr0.htm) found that 20.6% of US wage and salary workers had been with their employer for 12 months or less in January 2026, and median tenure was 4.1 years. The share was 25.5% at ages 25 to 34 and 9.2% at ages 55 to 64. [Lusha](https://www.lusha.com/blog/b2b-data-decay-rate-measured/) recorded a job change for 12.25% of US sales VPs and C-suite leaders over 12 months.

### How often should you clean a B2B contact list?

Re-check work emails before every send, titles and employers every quarter for accounts you work, and the full record once a year. Yearly job-change rates of 12.25% (Lusha, US sales leaders) and 20.6% (BLS, all workers) mean about 3 to 5 in every 100 contacts change employer each quarter. Treat any hard bounce as an immediate trigger. See our [email bounce rate](https://enrich.chat/blog/email-bounce-rate/) guide.

### How fast do email addresses go bad?

[ZeroBounce](https://www.zerobounce.net/email-list-decay) reports that at least 23% of the email data it verified in 2025 was invalid or risky, down from 28% in 2024. It calls 23% a low-end figure because catch-all addresses cannot be confirmed without sending. [Amazon SES](https://docs.aws.amazon.com/ses/latest/dg/reputationdashboardmessages.html) puts a sender under review at a 5% hard bounce rate. Test risky domains with a [catch-all checker](https://enrich.chat/tools/catch-all-checker/).

### Can an API refresh outdated contact data?

Yes, if you re-run lookups on the records you choose. With the Enrich.chat [person enrichment API](https://enrich.chat/api/person-enrichment/), one request per contact returns the title, the company and a work email with a status, and a lookup that finds nothing uses 0 credits. Enrich.chat does not monitor people or send job-change alerts, so you set the schedule. The free tier gives 10 lookups a day with no signup.

## Sources

1. [Employee Tenure in 2026](https://www.bls.gov/news.release/tenure.nr0.htm)
2. [Employee Tenure in 2026, full release with Table 3 (PDF)](https://www.bls.gov/news.release/pdf/tenure.pdf)
3. [Job Openings and Labor Turnover, January 2026, with 2025 annual estimates](https://www.bls.gov/news.release/archives/jolts_03132026.htm)
4. [Number of Jobs, Labor Market Experience, Marital Status, and Health for those born 1957 to 1964](https://www.bls.gov/news.release/nlsoy.nr0.htm)
5. [Survival of private sector establishments by opening year](https://www.bls.gov/bdm/us_age_naics_00_table7.txt)
6. [Database Decay Simulation](https://www.hubspot.com/database-decay)
7. [CRM and the Marketing Database: Data hygiene, behavioral analysis and more](https://www.marketingsherpa.com/article/how-to/data-hygiene-behavioral-analysis-more)
8. [The Email List Decay Report for 2026](https://www.zerobounce.net/email-list-decay)
9. [How fast does B2B contact data decay?](https://www.lusha.com/blog/b2b-data-decay-rate-measured/)
10. [State of B2B Data Quality 2026](https://www.cleanlist.ai/blog/2026-04-07-state-of-b2b-data-quality-2026)
11. [B2B Data Decay and Its Impact on Marketing](https://industryselect.com/blog/b2b-data-decay-and-its-impact-on-marketing)
12. [The Real Cost of Poor Data Quality for B2B Teams](https://pipeline.zoominfo.com/operations/poor-data-quality-impact)
13. [B2B Data Decay Study: What the Published Data Measures](https://scalelist.com/research/data-decay/)
14. [Second Report and Order, FCC 18-177 (Reassigned Numbers Database)](https://docs.fcc.gov/public/attachments/FCC-18-177A1.pdf)
15. [Reputation metrics messages](https://docs.aws.amazon.com/ses/latest/dg/reputationdashboardmessages.html)
16. [The State of Email Outreach](https://hunter.io/the-state-of-cold-email)
17. [Principle (d): Accuracy](https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/data-protection-principles/a-guide-to-the-data-protection-principles/accuracy/)

## Try this prompt

Re-check a sample of your contacts

```text
Use www.enrich.chat to re-check the 200 contacts in my sample list. For each LinkedIn URL, look up the current title, company and work email status, flag every contact whose title or company differs from my copy, and tell me the total credits_used
```

## Related

- [CRM data enrichment: what, how often, how](https://enrich.chat/blog/crm-data-enrichment/)
- [Email bounce rate: what is good](https://enrich.chat/blog/email-bounce-rate/)
- [What is data enrichment?](https://enrich.chat/blog/what-is-data-enrichment/)