How Fast Does B2B Contact Data Decay Each Year?

TL;DR: A typical B2B contact database loses roughly 2-3% of its accuracy every month, which compounds to somewhere between 20% and 35% per year. Phone numbers decay slower than emails on paper, but they go "functionally dead" faster — the number still rings, it just doesn't reach the person you're trying to sell to anymore.

If you bought or built a list twelve months ago and haven't touched it since, you should assume that somewhere between one in five and one in three records is now wrong in some material way — wrong title, wrong company, wrong number, or a person who's simply gone. That's not a worst-case scenario. That's the average.

What is the average B2B data decay rate per year?

Most industry benchmarks put annual B2B contact data decay at 20-35%, with 22-25% being the figure you'll see cited most consistently across data vendors, CRM providers, and list hygiene tools. On a monthly basis that works out to roughly 2-3% of records going stale, and it compounds — a list you don't touch for two years isn't twice as decayed, it's worse, because errors layer on top of errors (a wrong title gets "corrected" to a different wrong title, a dead number gets reassigned to someone else's desk).

The range is wide because decay isn't uniform. Fast-moving sectors (tech, startups, anything with high turnover) tend toward the top of that range or higher; slower-moving sectors (manufacturing, government, education) sit toward the bottom. If you want a planning number and don't have your own data to measure against, 2% per month is a reasonable default assumption.

Why does B2B contact data decay so fast?

Contact data decays because the humans behind it keep changing jobs, roles, and phone numbers, and no database updates itself in real time. The single biggest driver is job changes: somewhere around 15-20% of professionals change jobs in a given year, and each move invalidates a title, sometimes an email, sometimes a phone extension, and occasionally all three at once. Company-level changes (mergers, rebrands, office closures) add another layer of decay on top of individual turnover.

The practical result: a list is never "done" being clean. It starts decaying the moment it's built, whether you paid for it, scraped it, or collected it from inbound forms.

Do phone numbers decay faster or slower than emails?

Phone numbers generally decay slower than emails on a pure "is this number still assigned to someone" basis, but they become useless just as fast in practice. Estimates for phone-specific decay run 20-35% annually — similar to or somewhat higher than the all-fields average — with a meaningful gap between mobile numbers and office direct dials. Mobile numbers are personal and portable, so they tend to stay valid for years even after someone changes jobs, sometimes decaying at only 5-10% a year. Office direct dials, by contrast, are tied to a desk and a role, and turn over much faster — commonly cited in the 15-20%+ range, accelerated by the post-2020 shift toward remote and hybrid work that left a lot of fixed extensions pointing at empty desks.

Here's the catch that raw decay percentages hide: a phone number can be 100% technically valid — it rings, it connects, caller ID even shows the right name — and still be a dead lead, because the person answering isn't in the role you're selling to anymore, or isn't there at all. Validity and relevance are two different things. A list-cleaning tool checks the first. It takes an actual conversation, or an AI agent doing the equivalent, to check the second.

How do you measure data decay in your own list?

The simplest way to measure decay is to re-verify a random sample of your list on a fixed schedule and track what percentage of records fail. Pull 200-500 records at random every quarter, check whether the phone number connects, the email doesn't bounce, and the person is still in the role your CRM says they're in, then calculate the failure rate. Do this consistently and you'll get a decay curve specific to your list and your market, which is far more useful than any industry-wide average — a list of enterprise IT directors decays differently than a list of startup founders.

If a quarterly audit isn't realistic, a rough proxy works too: look at your bounce rate on email sends and your "wrong person" or "no longer here" rate on calls over the last 90 days. Both are decay showing up in your existing outreach metrics, just unmeasured as such.

How often should you re-verify a B2B contact list?

Re-verify a purchased or aging B2B list every 3-6 months, and more often for any segment you're actively running high-volume outreach against. At a 2-3% monthly decay rate, a list that's six months old is already 12-18% stale on average, and a list that's a full year old is a coin flip on any individual record being fully accurate. For lists feeding cold call or cold email campaigns — where a wrong number or wrong title burns rep time and sender reputation, not just a bounced message — quarterly verification is the more defensible cadence.

The cost of over-verifying is small (some wasted checks on records that hadn't changed yet). The cost of under-verifying compounds: reps dial numbers that don't reach anyone, emails bounce and damage domain reputation, and pipeline reports get built on contacts who don't exist anymore in the role you think they're in.

Is it cheaper to re-verify a list or buy a fresh one?

Re-verifying an existing list is almost always cheaper than replacing it, because most of a decayed list is still correct — you're paying to identify and fix the 20-30% that's wrong, not to rebuild the other 70-80% you already have. Buying a fresh list resets your decay clock to zero, but it doesn't fix the underlying problem: whatever list you buy will start decaying at the same 2-3% monthly rate the day it lands in your CRM, and you'll be back in the same spot in six to twelve months.

This is really a build-vs-verify question, not a buy-vs-build one. A verification pass — confirming the phone still reaches the right person, the title is current, the company hasn't changed — costs a fraction of acquiring net-new contacts and, when done through an actual conversation rather than a database ping, catches the "technically valid but functionally dead" records that pure data-matching services miss. Ringfire runs an AI agent through exactly that kind of call-and-confirm pass on an uploaded list, scoring each contact by whether it's still the right person at the right number, rather than just whether the digits resolve to something.