How fast does B2B contact data really go stale? The decay stats everyone quotes, and what the independent evidence shows

By Uassist.ME Editorial Team · Published by Uassist.ME · Published 2026-10-27 · Last updated 2026-10-27

Disclosure: ThePipelineBots is published by Uassist.ME (uassistme.com), which sells outsourced sales-support, SDR, CRM and research teams, including CRM data cleanup and list research. Scary decay numbers help sell that kind of work, so we have a direct stake in this topic. Weigh our take with that in mind. This article cites several data and software vendors, including ZoomInfo, Lusha, HubSpot and Apollo.io. No vendor named here paid for, reviewed or saw this article. Three sources come from lead-generation companies that compete with Uassist.ME: Belkins, ViB (Sean Shea's Forbes post) and Smirnov Consulting Group (Artem Smirnov's trace). Every figure below is listed with its source at the end, and vendor figures are labeled as vendor-reported.

Short answer: Nobody has published an independent, method-disclosed annual decay rate for B2B contact data. The figures everyone quotes, "22.5% a year", "70% a year" and "43% a year for email", trace back to a HubSpot footnote that credits MarketingSherpa research without a link, date or method, a consultant's business-card surveys and one month of one vendor's data. The best neutral evidence is narrower. US government data shows that 20.6% of wage and salary workers had been with their employer 12 months or less in January 2026 (BLS, Sept 24, 2026). A peer-reviewed field study found that bought B2B lists had the right name and company only 67.5–70.9% of the time, and the right job title only 16.3–17.0% of the time (Neumann et al., 2023). Our verdict on the popular decay stats is "bad": don't repeat them as fact. Measure your own list instead.

What does "contact data decay" actually mean?

"Decay" is used for at least four different things, and most published numbers don't say which one they measure:

• The person left the company. The record now points to someone who isn't there.
• Something on the record changed. That could be a new title, phone number or office, even if the person stayed.
• The email address stopped working. Mail to it bounces.
• The record was wrong from the start. That is an accuracy problem, not decay, but it looks the same in a CRM.

These produce very different numbers. A title change inside the same company makes a record out of date, but the email may still work. A rate built from "any change" will always be much higher than a bounce rate. When you see a decay figure, first ask which of these it counts.

Where do the famous decay numbers come from?

We traced each widely quoted figure to the earliest source we could open on Oct 23, 2026.

The 22.5% figure

HubSpot's Database Decay Simulation page (undated) says "Email marketing databases naturally degrade by about 22.5% every year." Its only source is a footnote: "Marketing Sherpas research shows that B2B data decays at a rate of 2.1% per month. This is an annualized rate of 22.5%." The footnote has no link, date or method. The only other material the page cites is HubSpot's own 2013 and 2014 State of Inbound reports. We searched the 2009–10 MarketingSherpa B2B marketing benchmark chapter that HubSpot hosts, and the 2.1% figure is not in it. That is one chapter, not the full report. A separate trace by Artem Smirnov, who runs a B2B lead-generation agency, reached the same dead end (Smirnov, Sept 28, 2026).

The math, at least, is consistent: 2.1% a month compounded over 12 months is about 22.5% a year. But that means 2.1% and 22.5% are one data point quoted twice, not two findings that confirm each other.

The 70% figure

The top of the popular range comes from John Coe, a B2B marketing consultant. In a 2015 Biznology post (Feb 13, 2015), Coe described a business-card survey: more than 1,200 people who took part "were simply asked what has changed on their business card in the last 12 months." Coe summarized the result as B2B data decaying "at a rate of just over 5% per month, or 70.3% per year." The biggest change was title or job function (65.8%, including a new job at the same firm). Email address changes were much lower, at 37.3%. Coe also wrote that when list brokers were asked, they "most often responded that it was about 25-30%."

In a May 27, 2020 post on IndustrySelect, a company that sells B2B data, Coe described "an independent survey of 1,025 businesspeople" done the same way at seminars. 70.8% of the cards had one or more changes, and a show of hands at a UK seminar came to "exactly 70%".

These are informal, self-reported samples; Coe's 2020 post says the cards were collected at his seminars. Coe wrote in 2015 that the study was first run in 2002 and updated in 2009. They measure "anything changed", not "the record stopped working". Coe's own 2015 numbers show that only 37.3% of cards had an email change. Quoting the 70% figure as an email decay rate overstates what the survey found. Landbase (Jan 8, 2026) credits its 70.3% to a 2024 Forbes Business Council post by Sean Shea (Forbes, Apr 18, 2024), which attributes it to Gartner ("download required"). The Gartner page that post links to, which now covers Gartner's 2027 technology trends, doesn't contain the figure, and we couldn't find a Gartner source for it. 70.3% is the same number Coe published in 2015.

The 3.6%-a-month figure, and where 43% comes from

RevenueBase, a B2B data vendor, wrote in a Jan 5, 2025 post by its CEO: "B2B email decay reached 3.6% in November 2024, significantly higher than the traditional 1.5-2% monthly rate." The post says RevenueBase tracks "millions of B2B contact records" but gives no exact sample, data set definition or method. Landbase repeated the figure in its Jan 8, 2026 stats roundup, crediting RevenueBase. ZoomInfo's decay page lists email decay at "3.6%/month" and "43%/year", crediting Landbase and SMARTe. ZoomInfo's page says the monthly rate is "compounding to more than 43% annually". We didn't find 43% on the Landbase or SMARTe pages it credits.

Check that math yourself. A monthly rate compounds, because each month's losses come out of a list that is already smaller:

• Compounded: 1 − (1 − 0.036)^12 ≈ 0.356, so about 35.6% a year.
• Simple multiplication: 3.6% × 12 = 43.2%. That matches the "43%" figure, but simple multiplication overstates a compounding rate.

So even if you accept one month of one vendor's data as typical, it implies about 35.6% a year, not 43%. And one unusual month is a weak basis for any annual figure. RevenueBase itself said November was "significantly higher than the traditional 1.5-2% monthly rate."

The 30% figure

Lusha, a contact-data vendor, published a measurement on Sept 12, 2026. On the 30% figure, it wrote: "We could not find a primary source for it. It appears across vendor blogs, sales decks and pitch material throughout this category, each citing another secondary source." Vendor-reported: among 140,284 US sales VP and C-suite contacts in its own database, Lusha says 12.25% changed roles over 12 months and 25.67% over 24 months. Lusha's conclusion is that the 30% figure "is roughly right. It is being quoted over the wrong period," since it is close to its two-year rate. For the UK, the 12-month figure was 21.40%. Lusha notes that this "counts employment changes only, so the rate at which records become wrong in some way is higher." This is Lusha's own data on a senior, narrow group, and it hasn't been independently checked. But it is a rare vendor figure that comes with a cohort size and a date.

Other vendor figures go the other way: Cleanlist, a data vendor, reports re-checking 5,000 contacts weekly for 90 days and finding 2.1% a week, which it annualizes to about 67% (Victor Paraschiv, Cleanlist, Apr 7, 2026 (updated Jul 17, 2026)). Vendor-reported.

What does government data say about job changes?

The most reliable neutral number is from the US Bureau of Labor Statistics. Its Employee Tenure in 2026 release (USDL-26-1532, Sept 24, 2026) is based on a supplement to the Current Population Survey, a monthly survey of about 60,000 households:

• Median tenure with the current employer was 4.1 years in January 2026, up from 3.9 years in January 2024.
• 20.6% of wage and salary workers had been with their employer 12 months or less, down from 22.2% in January 2024.
• Tenure rises with age. Median tenure was 3.0 years for ages 25–34 and 9.6 years for ages 55–64. Only 9.2% of workers aged 55–64 had 12 months or less.

How to use this, and how not to. BLS measures employer changes across the whole workforce. Treat it as a rough indicator of job churn, not a decay rate: it counts some people who were never on a list, and misses people who left a job and are not working. Senior, older buyers move less often than the workforce average (median tenure in management occupations was 6.1 years), and young, junior contacts move more.

What does independent research say about list accuracy?

Bought B2B lists were often wrong from the start

The strongest independent evidence on B2B contact accuracy is a peer-reviewed study in Quantitative Marketing and Economics by Nico Neumann (Melbourne Business School), Catherine E. Tucker (MIT Sloan), and Kumar Subramanyam and John Marshall (HP Inc.), published Oct 24, 2023. The field work ran from September 2019 to April 2020, with a large IT brand trying to reach "IT decision-makers" in the US.

• Name and company. The researchers sampled 884 people from one list vendor and 470 from another, and looked each person up online. Name and company matched for 67.5–70.9%. For another 20.2–24.8%, the search found the person at a different company.
• Title and location. Only 16.3–17.0% of job titles and 21.5–26.1% of locations matched.
• Targeting results. The two bought "IT decision-maker" segments built from named lists reached 16.1–18.2% actual IT decision-makers, against 16% for random users on the same publisher network. The difference was not statistically significant. Three segments built by statistical inference did worse, at 9.4–11.0%.

The authors write that they "can only speculate about the underlying reason, one of which could be simply outdated data." That is the honest reading. This study shows how often bought lists were wrong. It doesn't measure how fast they went stale.

Caveats: 2019–20 data, unnamed vendors, two co-authors from HP Inc. (which supported the study), a published consulting disclosure from Tucker, and an ad-targeting question, not email outreach; even so, it is the best independent check of B2B list accuracy we found.

What do benchmarks and vendors report?

These newer, more transparent sources measure accuracy at a point in time, not decay over a year.

Two takeaways:

1. "Valid" isn't the same as "deliverable". In B2BDataBenchmarks' run, 20.6–50.0% of the addresses each provider returned were on catch-all domains, which accept any address. Nobody can confirm those without sending.
2. Verifier checks and real sends tell different stories. The two benchmarks that sent real email (Dropcontact and Belkins, both vendor-reported) report low hard-bounce rates on the addresses they kept. Benchmarks that only use verifiers can't measure bounces at all. B2BDataBenchmarks says so itself: "There is no real send, so there is no bounce rate."

What's still missing?

• An independent annual decay study for B2B contacts. We found none that follows a defined set of business contacts over a year and publishes its method. The closest academic measurement is of scientists' email addresses published in MEDLINE up to mid-2004: about 24% had stopped working within a year of publication (Wren et al., EMBO Reports, Feb 2006).
• An independent B2B bounce benchmark with real sends. The real-send figures we found come from vendors and an agency.
• Data by segment. Decay likely differs by seniority, country and industry. Lusha's US vs UK gap and BLS's age figures suggest how much, but no neutral source breaks this down for sales contacts.

Until those exist, any single "X% a year" figure is an estimate, not a measured rate. We are not going to add our own.

How do you spot-check your own list?

These are our suggestions, not findings from the sources above.

1. Pull a random sample. Take 50–100 records at random, not the newest or the top of a sort.
2. Check each field separately against a first-party source. Use the company's website, a recent press release or the person's own profile. Record whether the person is still at the company, whether the title is right, and whether the phone and location match. The Neumann study shows these fail at very different rates.
3. Run emails through a verifier, but count catch-all as "unknown", not "valid". Report valid, invalid, catch-all and unknown separately.
4. If you send, separate hard bounces from soft bounces. Only hard bounces say the address is gone.
5. Stamp each record with the date it was last verified. Re-check the same sample in 3 and 6 months. The change you measure is your list's decay rate. It is the only one that applies to you.
6. Question any vendor's figure, and do the compounding math yourself. Ask for the sample size, date, definition of "decayed", whether it was a real send, and who ran it. For a monthly rate r, the annual rate is 1 − (1 − r)^12, not 12 × r.

Related reading

• Do you need to clean your CRM before using an AI sales agent?
• Can AI research agents do account research as well as an SDR?
• Before you let an AI agent touch your CRM: what should you check?

FAQ

What is the real B2B data decay rate? No independent, method-disclosed annual rate exists, as of Oct 23, 2026; the popular 22.5% and 70% figures trace to a HubSpot footnote that credits MarketingSherpa research without a link or date and to informal business-card surveys. For context on job churn, BLS found that 20.6% of US wage and salary workers had been with their employer 12 months or less in January 2026.

Where does the 22.5% per year figure come from? From a HubSpot footnote that credits Marketing Sherpas research with 2.1% a month, which compounds to about 22.5% a year, with no link, date or method. We could not find it in the one MarketingSherpa chapter we could open, which HubSpot hosts.

Is it true that 70% of B2B contact data goes bad every year? Not as a measured decay rate: it comes from John Coe's informal business-card surveys, which counted any change, including new titles. Only 37.3% of cards in his 2015 write-up had an email change.

Does email data decay 3.6% a month, or 43% a year? The 3.6% is one month (November 2024) of RevenueBase's own data. Even if it held every month, it would compound to about 35.6% a year; 43% matches multiplying 3.6% by 12, which overstates a compounding rate.

How accurate are bought B2B contact lists? In a peer-reviewed study with 2019–20 data, bought lists had the right name and company 67.5–70.9% of the time and the right job title only 16.3–17.0% of the time. In one 2026 benchmark (B2BDataBenchmarks: 60 people, verifier checks only), the best valid-email rate was 75.6%.

How often should I re-verify my contact list? No source supports a single schedule. Re-check a random sample of 50 to 100 records every few months, and more often for the segments that change most.

Sources (opened 2026-10-23)

1. U.S. Bureau of Labor Statistics, "Employee Tenure in 2026" (USDL-26-1532), Sept 24, 2026. Independent (government). https://www.bls.gov/news.release/tenure.nr0.htm
2. Nico Neumann, Catherine E. Tucker, Kumar Subramanyam and John Marshall, "Is first- or third-party audience data more effective for reaching the 'right' customers? The case of IT decision-makers," Quantitative Marketing and Economics, published Oct 24, 2023 (field work Sept 2019–Apr 2020). Peer-reviewed. Two co-authors at HP Inc., which supported the study. https://link.springer.com/article/10.1007/s11129-023-09268-7
3. Jonathan D. Wren et al., "E-mail decay rates among corresponding authors in MEDLINE," EMBO Reports 7(2):122–127, Feb 2006. Peer-reviewed. Academic emails, not B2B. https://pmc.ncbi.nlm.nih.gov/articles/PMC1369259/
4. HubSpot, "Database Decay Simulation," undated. Vendor page. Decay figure credited to MarketingSherpa, with no link or date. https://www.hubspot.com/database-decay
5. MarketingSherpa, 2009–10 B2B Marketing Benchmark chapter (PDF hosted by HubSpot, created Dec 2009). One chapter excerpt (Chapter 4). Checked for the 2.1% figure: not present. https://www.hubspot.com/hubfs/docs/B2BSherpaHandbookChapterfromHubSpot.pdf
6. John Coe, "B2B data decay and list rental – buyer beware!," Biznology, Feb 13, 2015. Consultant. Informal business-card survey. https://biznology.com/2015/02/b2b-data-decay-list-rental-buyer-beware/
7. John Coe, "B2B Data Decay and Its Impact on Marketing," IndustrySelect blog, May 27, 2020. Published by a B2B data seller. Informal survey. https://www.industryselect.com/blog/b2b-data-decay-and-its-impact-on-marketing
8. SiriusDecisions, "The Impact of Bad Data on Demand Creation," PDF metadata dated Nov 2008; no date in the text. Analyst claim. No method published. https://www.ecrs.com/wp-content/uploads/assets/TheImpactofBadDataonDemandCreation.pdf
9. ZoomInfo, "B2B Data Decay: Rates, Costs, and How to Stop It," updated 06/30/2026. Vendor-reported. https://pipeline.zoominfo.com/marketing/b2b-data-decay
10. Landbase, "Data Decay Rate Statistics: 20 Critical Facts Every GTM Leader Should Know in 2026," Jan 8, 2026. Vendor-reported. https://www.landbase.com/blog/data-decay-rate-statistics
11. SMARTe, Sanjay Gala, "What is B2B Data Decay and How It Is Impacting Your Business," last updated Oct 22, 2026. Vendor-reported. Contains no 3.6% or 43% figure. https://www.smarte.pro/blog/b2b-data-decay
12. Mark Feldman, "The Alarming Rise in B2B Email Decay: What It Means for Your Data, Sales, Marketing, and Ops Teams," RevenueBase, Jan 5, 2025. Vendor-reported. https://revenuebase.ai/blog/b2b-email-decay-what-it-means-for-your-data-sales-marketing-and-ops-teams
13. Sean Shea (founder and CEO of ViB, a B2B lead-generation company), "The B2B Data Decay Epidemic: How To Protect Your Bottom Line," Forbes Business Council, Apr 18, 2024. Council post by a lead-generation company CEO. Says "B2B contact data decays as fast as 70.3% per year, according to Gartner (download required)." The Gartner page it links to (opened 2026-10-23) does not contain the figure. https://www.forbes.com/councils/forbesbusinesscouncil/2024/04/18/the-b2b-data-decay-epidemic-how-to-protect-your-bottom-line/
14. Gartner, "Top Strategic Technology Trends" page linked from the Shea post, opened 2026-10-23; the page now covers 2027 trends and has changed since 2024; no decay figure. https://www.gartner.com/en/information-technology/insights/top-technology-trends
15. Nirit Braun, "How Fast Does B2B Contact Data Decay? Measured Twice on 148,000 Records," Lusha, Sept 12, 2026 (updated Sept 13, 2026). Vendor-reported. https://www.lusha.com/blog/b2b-data-decay-rate-measured/
16. Victor Paraschiv, Cleanlist, "B2B Data Quality 2026 Report," Apr 7, 2026 (updated Jul 17, 2026). Vendor-reported. https://www.cleanlist.ai/blog/2026-04-07-state-of-b2b-data-quality-2026
17. B2BDataBenchmarks, "Email Data Benchmark" (test set checked 2026-09-25) and Methodology page. Operator not disclosed. https://www.b2bdatabenchmarks.com/benchmarks/email
18. Openbenchmarks, "Company Enrichment API Benchmark 2026" (snapshot 2026-08-03) and About page. Independence self-stated; sells evals to vendors. Company data only. https://openbenchmarks.com/company-enrichment
19. Dropcontact, "Email Finder Benchmark 2025: 15 Tools, 20,000 Contacts Tested," updated Feb 2026. Vendor-reported. Run and funded by Dropcontact. https://www.dropcontact.com/email-finder-benchmark
20. Vladyslav Podoliako (Belkins co-founder and CEO), "What Are B2B Cold Email Deliverability Rates? (2026 Study)," Belkins, updated 2026-06-29. Vendor-reported. Belkins is an outbound agency that competes with Uassist.ME. https://belkins.io/blog/email-deliverability-rates
21. ZeroBounce, "The Email List Decay Report for 2026," undated page; data from Jan–Dec 2025. Vendor-reported. https://www.zerobounce.net/email-list-decay
22. Artem Smirnov, "The B2B Data Decay Rate Everyone Quotes, Traced to Source," last updated Sept 28, 2026. Secondary trace by the head of a lead-generation agency that competes with Uassist.ME. https://smirnovartem.com/blog/b2b-data-decay