Here's a moment most sales teams know. You open an account you closed last year, because the renewal is coming up, and the person you sold to isn't there anymore. Their email bounces. The record still says their old title at their old company. So where did they go? And who else do you know over there? Often the honest answer is nobody.
How common is that? We ran the numbers on 238,628 people records and 113,417 companies from the CRMs of 40 teams that use Ln2CRM, on five different CRMs. No survey and no opinions, just counts. You'll find the method at the end of the page.
The numbers at a glance
Every share below is out of all 238,628 people records, unless a row says otherwise.
| What we measured | Result | Why it matters |
|---|---|---|
| People with no LinkedIn URL on the record | 63.5% | Nothing stable to find them by once they move |
| People with no company at all | 23.1% | Floating contacts nobody can tie to an account |
| No LinkedIn URL and no company | 15.6% | A name, maybe an email, and that's it |
| No LinkedIn URL, no company, no email | 5.0% | Records you can't do anything with |
| People with an email address | 52.0% | Barely half |
| Emails on a company domain (out of all emails) | 71.5% | Work addresses stop working when the job ends |
| Reachable only through a work email, no LinkedIn URL | 27.3% | Lost the day they leave |
| Companies with exactly one contact | 79.4% of 113,417 | One departure and the account is empty |
| Companies whose only contact has no LinkedIn URL | 49.3% | Half of all accounts sit on one hard to trace person |
| People who share their exact full name with another record in the same CRM | 11.9% | Matching by name alone goes wrong about one time in eight |
Why does a LinkedIn URL matter so much?
Think about what changes when somebody switches jobs. Company, title, work email, phone, often the city. What stays? Their name, which isn't unique, and their LinkedIn profile, which is. The profile URL follows the person, not the employer, so it's the one field on a CRM record that still points at the right human two jobs later.
And yet 63.5% of the records we looked at don't have one. It varies a lot by team. Among accounts with at least 200 people in their CRM, the share without a LinkedIn URL goes from 2% to 100%, with the middle account at 53%. The 100% teams simply never added a LinkedIn field to their CRM, which most CRMs don't have out of the box. That's worth knowing, because it isn't laziness. It's a missing field nobody set up.
One contact per account: the quiet risk
This one surprised us most. Out of 113,417 companies in these CRMs, 79.4% have exactly one person attached. Another 11.2% have two. So for most accounts, the whole relationship lives in one record.
Is that a problem? Only when that person leaves, which, given how long people stay in jobs, is a question of when. And here's the part that hurts. In 49.3% of all companies, that single contact has no LinkedIn URL either. When they go, you lose the person and the account in one move, and you won't even know where they went.
Sales people call the fix multithreading, and it's mostly about deals in progress. We'd argue it matters just as much after the deal closes. Two or three saved contacts per account, each with a LinkedIn URL, means a job change costs you one relationship, not the whole customer.
Contacts with no company at all
Nearly one in four people (23.1%) isn't tied to any company in the CRM. Some came in from an event list, some from an old import, some were typed in a hurry. They're findable today because you remember them. In a year? Probably not.
The worst slice is 15.6% of all records with neither a LinkedIn URL nor a company, and 5.0% that also have no email. That's a name in a database. You can't email it, you can't look it up, and you can't even be sure which of several people with that name it was.
Work emails die with the job
How many people in these CRMs even have an email? Only 52.0%. Of those addresses, 71.5% sit on a company domain and 28.5% are personal ones like Gmail or Outlook. A personal address usually survives a job change. A work one usually doesn't, because the company closes the mailbox when the person leaves.
Put the two together and you get the number we'd underline: 27.3% of all contacts can be reached only through a work email, with no LinkedIn URL on the record. For more than a quarter of the CRM, the only line to the person runs through an employer they might already have left.
We didn't want to just assume that, so we tested it. We took several hundred addresses from a sales CRM we run ourselves and from the CRMs of two customer teams that agreed to take part, grouped them by when the record was created, and ran every one through an email check.
| How old the record is | Work emails that fail the check | Personal emails that fail |
|---|---|---|
| Under 3 months | 21% | 0% |
| 6 to 9 months | 21% | 1% |
| 1 to 2 years | 33% | 3% |
Two things jumped out. Even brand new records fail about one time in five. So a good share of work addresses are bad from the day they're saved. We'd guess most of them were built from a name pattern or typed in a hurry, and some sit on domains that won't confirm anything. Then time does the rest. In records one to two years old, about one in three work emails doesn't work anymore. Personal addresses barely move at all.
Namesakes: why matching by name breaks
Two records with the same full name in the same CRM? It happens to 11.9% of people, roughly one in eight. Some of those are duplicates of one person. Plenty are different people who just share a name. Either way, a tool that matches by name alone will glue the wrong records together or miss the right one.
LinkedIn makes this worse, not better, because a common name there can return dozens of profiles. That's why we match on the LinkedIn URL first and only fall back to name plus company when there's no URL. Without the URL, a name is a guess.
How often do people actually change jobs?
Our data doesn't track job moves, so here we lean on public figures. The US Bureau of Labor Statistics found that the median worker had been with their current employer for 3.9 years in January 2024. In the private sector it was 3.5 years, and for people aged 25 to 34, the age of a lot of buyers and champions, just 2.7 years.
You'll also see vendors quote that B2B contact data decays by about 22.5% a year, which is 2.1% a month compounded. Apollo cites that figure, though we couldn't find the original study behind it, so treat it as an industry estimate rather than a measured fact. The BLS tenure numbers point the same way, though. In a few years most of today's contacts will be somewhere else.
How to keep CRM data from decaying
Do you need a data team for this? Nope. It's mostly habits, and a couple of fields.
- A LinkedIn URL on every person. Add a LinkedIn field if your CRM doesn't have one, then fill it every time you save someone. It's the one identifier that survives a job change.
- Always a company. A person without an account is easy to lose and impossible to report on. If you don't know the company yet, that's a sign to look before you save.
- Two or three contacts per customer. Your champion, their manager, someone in operations. When one leaves, the account stays warm.
- A personal email, when someone gives you one. People share it more often than you'd think, and it outlasts any job.
- Notice job changes early. When a contact's LinkedIn shows a new company and your CRM still has the old one, that's both a lost contact at the old account and a warm lead at the new one.
- Old work emails, before a campaign. Anything a year old has a real chance of bouncing. Why care? Bounces hurt your sender reputation.
Full disclosure, we make Ln2CRM, so this is where we come in. When you open a LinkedIn profile, it shows whether that person is already in your CRM, and the badge turns yellow when their title or company on LinkedIn doesn't match the record. Saving someone stores the LinkedIn URL and links the company in the same click, which is exactly the part most CRMs end up missing.
A quarterly cleanup: what we do with old contacts
Fixing new records is the easy half. What about the thousands already sitting in the CRM? We don't try to clean everything at once. Once a quarter, and always before a campaign, we go through the records nobody has touched in six months, account by account.
For each person the question is simple. Are they still where the CRM says they are? Opening their LinkedIn profile answers it in a few seconds, and with Ln2CRM installed the badge on the profile already tells you, since it turns yellow when the title or company there doesn't match the record. From there it's one of three outcomes.
- Same company, new title. You just fix the title. And a promotion's often a good excuse to get in touch, isn't it?
- New company. That's two things at once. The old account just lost its contact, so it needs someone new, and the person is now a warm lead at a company that may never have heard of you. We save them at the new place and keep the old record's history, rather than overwriting it.
- Out of the picture. Retired, moved into a field you don't sell to, or simply gone. We archive these instead of deleting them, so the history stays but they stop clogging lists, reports and sequences.
Then the emails. A work address that hasn't been used in months is the riskiest field on the record, so we run it through a check before anyone sends to it. Ln2CRM can find and verify a work email right from the profile, which also fills the gap for the half of contacts who never had an address at all.
One more lesson from our own prospecting. Lead lists go stale in the same way, just faster. A Sales Navigator list looks clean at the top, and the deeper you scroll the more you hit people who've changed roles or left the field entirely. So the cleanup isn't only for old customers. It's worth doing for any list older than a couple of months, before it goes into a sequence.
Methodology
Where do the numbers come from? To show whether someone's already in your CRM, Ln2CRM keeps a small index of match keys for each connected CRM, so the name, the company, the LinkedIn URL and the email, which we store encrypted. We counted that index on 20 November 2026. It covered 238,628 people records and 113,417 companies from 40 teams. By CRM, 57.7% of the people were on Zoho CRM, 32.7% on Pipedrive, 3.3% on HubSpot, 3.3% on Capsule and 3.0% on Salesforce, so the mix leans heavily toward Zoho and Pipedrive users.
We only looked at totals. No one opened individual records. To sort emails into work and personal ones, each address was decrypted in memory, its domain checked against a list of free mail providers, and nothing but the count was kept. Companies were matched by name within each CRM after dropping legal endings like Inc, LLC and GmbH, so "Acme" and "Acme Inc" count once. Namesakes mean an exact match on the normalized full name inside the same CRM. Teams that never set up a LinkedIn field show up as 100% without a URL, and we left them in on purpose, because that's what their CRM really looks like.
The email check is separate from the index. It used several hundred addresses from a sales CRM we run ourselves and from the CRMs of two customer teams that agreed to take part, split by when each record was created, and an address counts as failed when the check couldn't confirm it can receive mail.
Want to quote these numbers? Go ahead, and please link back to this page as the source: Ln2CRM, CRM data decay study, November 2026.
FAQ
What is CRM data decay?
How fast does B2B contact data decay?
Why should I save a LinkedIn URL in my CRM?
What happens to a CRM contact when they change jobs?
How do I find a contact who changed jobs?
Can I use these numbers in my article?
Related reading
- Logging LinkedIn messages in a CRM, with what 753 synced conversations show
- How to find someone's email from LinkedIn
- The best CRM for LinkedIn prospecting