What to Do When Your Customer List Is Always Out of Date

Everyone who keeps a customer list has had the same experience: you go to email someone, the address bounces, and it turns out they left eight months ago. Nobody did anything wrong. The list just aged.

Key takeaways

• B2B data decays at about 2.1% a month, or 22.5% a year, on MarketingSherpa research that HubSpot uses as its benchmark. A list you cleaned in January is about a fifth wrong by December.
• The cause is people changing jobs, not sloppy data entry. 3.2 million Americans quit in a single month, and median tenure is 3.9 years, the lowest since 2002.
• Scheduled clean-ups lose to the decay rate. A quarterly pass leaves the list wrong for most of the quarter.
• The fix is to stop treating the list as something you update and start treating it as something the record updates itself from.
• Every correction you need is already arriving in your inbox, in signatures, out-of-office replies and bounces.

Why the list goes stale no matter how careful you are

Contact records rot at a fairly predictable rate. MarketingSherpa measured B2B data decaying at 2.1% a month, which HubSpot annualizes to about 22.5%, and it compounds quietly. Nothing looks broken until the day you need one specific person.

It rots because people move, which makes the problem structural rather than a discipline failure on your team’s part. 3.2 million Americans quit a job in a single month according to the BLS, and median employee tenure is now 3.9 years. Your contacts are churning underneath you whether or not anyone touches the spreadsheet.

Which is why the usual remedies underperform. A quarterly clean-up assumes decay pauses between passes. At 2.1% a month, a list cleaned perfectly on January 1 is about 6% wrong by April and 12% wrong by July, and you only notice through failures: a bounce, a wrong name, an email to someone who left.

The cost shows up elsewhere too. Gartner puts the average cost of poor data quality at $12.9 million a year across organizations, and 76% of CRM users report problems traceable to bad data. Smaller teams feel it as wasted time rather than a line item.

The three things people try

Periodic clean-ups. Someone spends a Friday fixing the list. It works for a fortnight and then decays at the same rate as before, because nothing about the process changed. This is the most common answer and the least durable.

Buying an enrichment service. Better, because it operates continuously, and it has a real limitation: enrichment knows what a provider’s database knows. It is good at job titles at recognizable companies and weak at whether the person you deal with at a twelve-person supplier still works there.

Discipline. Asking everyone to update the record after every call. This fails for the reason all manual-logging schemes fail: it is work with no immediate payoff for the person doing it, so it happens when things are calm and stops when they are not. Sales teams already spend 70% of their time on nonselling tasks, and adding to that pile does not go well.

The corrections are already in your inbox

You do not need to go looking for the new information, because it arrives on its own.

A signature block shows a current title, a direct line and sometimes a new company. An out-of-office names a replacement and their address. A bounce is a definitive statement that an address is dead. A reply from a colleague you have not corresponded with tells you the relationship has moved, or that the account was always more than one person.

All of that lands in your email, every week, and almost all of it gets read once and thrown away. The information problem is not scarcity. It is that nobody transcribes.

What actually holds

Stop maintaining the list. Build it from the messages instead.

A contact record that assembles itself from your mail and chat is current for the same reason the manual one is stale: it updates on the schedule that reality does, which is whenever somebody writes to you, rather than on the schedule that a human remembers to, which is roughly never.

Two things make it work in practice, and they are worth checking for in any tool you consider.

Local edits have to survive the sync. If you correct a phone number by hand and the next sync overwrites it, you stop correcting things. Whatever you fix manually should outrank the automatic source from then on.

It needs to reconcile one person across channels. The same human has an email address, a Slack handle and a phone number. If those stay three separate records, you have replaced a stale list with a duplicated one.

This is what Contacts in this+that does. It builds a people list from Google Contacts, Outlook, Slack, Google Chat, and Teams, resolves the same person across those channels into one record, and keeps a per-person timeline assembled from the messages themselves. Edits you make by hand are locked and beat the resync. It is deliberately not a system of record and has no deals or pipelines in it; think of it as a more sophisticated address book that maintains itself.

If you already run a CRM, the same principle applies to keeping that current: a workflow can push a corrected contact into HubSpot or Salesforce when the change appears in a message, rather than waiting for someone to type it.

Where to start this week

Run this once, to size the problem. Take the twenty contacts that matter most and check them against your recent mail instead of against the list. You will find the list is behind on several of them, and that the correction was sitting in a thread you already read.

That tells you the size of the gap. It is not a maintenance plan, because doing it by hand every month is the approach that already failed above, and the decay rate does not care how disciplined the person holding the calendar reminder is.

What the exercise proves is that finding the information was never the work. Every correction you turned up had already arrived in your inbox. The cost is the trip to go and get it, twenty contacts at a time, and that is the part to remove rather than schedule. Connect the accounts you already use and the record rebuilds from them, with each person’s recent correspondence sitting next to their record, so the check you just ran by hand is what you see when you open the person.

Frequently asked questions

How fast does customer contact data actually go out of date?

About 2.1% a month for B2B data, which annualizes to 22.5%, on MarketingSherpa’s measurement. The main driver is people changing jobs rather than data-entry mistakes.

Is a quarterly data clean-up enough?

Rarely. At about 2.1% decay a month, a list is roughly 6% wrong three months after a perfect clean-up, and you discover the errors through bounces and misdirected mail rather than at a convenient moment.

Do enrichment services fix this?

Partly. They run continuously, which beats a manual pass, and they are limited to what the provider’s database covers. Coverage is good for larger companies and thin for small suppliers and independents.

What signals show a contact record is stale?

Bounces, out-of-office replies naming someone else, replies arriving from a colleague you have not dealt with before, and signature blocks that no longer match the title you have stored.

Can a contact list update itself?

Yes, if it is built from your messages rather than maintained separately. The two things to check are that your manual edits survive the next sync and that the same person across email, chat and phone resolves to a single record.