How many of your CRM addresses are wrong? The address-decay problem

Open your CRM and look at any contact's mailing address. Now ask a simple question: how do you know it's still true? For most B2B databases the honest answer is that nobody knows. The address was captured once — from a data vendor, a badge scan, a signature block — and it has sat there ever since, while the person changed jobs, the company changed offices, and the way people use offices changed entirely. Email data decay is at least visible: send to a dead address and you get a bounce within seconds. A wrong physical address reports nothing. The envelope just disappears.

This post walks through the mechanics of how business addresses go stale, why nobody can honestly tell you what percentage of your CRM is wrong, why the silence of the failure makes it uniquely expensive for direct mail, and why address verification before printing is the single most important criterion when you evaluate a B2B direct mail vendor.

The four ways a business address goes stale

1. People change jobs

The most obvious decay mechanism is also the most complete: when a contact leaves the company, every piece of company-bound data attached to them dies at once — email, phone extension, and mailing address together. B2B sales has always priced in employee turnover for email purposes, but mail teams often forget that the same churn applies to the envelope. A letter addressed to someone who left last quarter doesn't get forwarded to their new company. At best it lands on a desk that no longer exists; more likely it's discarded by whoever now handles that department's mail.

2. Companies move

Offices are not fixed points. Leases end and companies relocate; growing companies open new offices and consolidate old ones; cost-cutting closes floors and buildings; mergers and acquisitions fold whole companies into someone else's real estate, often under a new name. In recent years many companies have deliberately downsized their footprint — the headquarters address a data vendor recorded a few years ago may now be a smaller office across town, a different city, or no office at all. None of these events sends an update to your CRM. Postal-forwarding systems like NCOA processing catch some corporate moves when companies file them, but coverage of business moves is partial at best — and it can't tell you anything about the individual person.

3. The HQ-vs-actual-desk problem

Even a perfectly current company address can be the wrong address for your contact. Data vendors overwhelmingly record the corporate headquarters, because that's what's publicly listed. But the VP you're targeting may sit in a regional office two time zones away, a satellite site, or a different campus building. Mail sent to a multi-thousand-person headquarters for someone who works elsewhere depends on an internal re-routing chain that may or may not exist — and that's before it survives the mail room. The address isn't wrong, exactly. It's just not where your buyer's desk is, which for an envelope amounts to the same thing.

4. Hybrid work rewired the office

The quiet revolution underneath all of this is that 'their office' is no longer a stable concept. Hybrid schedules mean your contact may be in the building two or three days a week. Hot-desking means there may be no desk with their name on it at all. Some roles went fully remote and their 'office' is a home address no B2B data vendor has — or should have. Remote and hybrid work changed the calculus for direct mail: mail to offices still works, but only when someone has verified that this person actually works from that building. An assumption that was safe in 2019 is a coin flip now, and the CRM field doesn't tell you which kind of contact you're looking at.

Why nobody can tell you your decay rate — including us

Here is where an honest post has to part ways with most marketing content on this topic. Email data decay has published research behind it, because email failure is observable at scale: every send returns a bounce or it doesn't, so anyone with volume can measure decay directly. Physical business addresses have no equivalent feedback loop. USPS mail to a wrong business address mostly just vanishes — no bounce, no notification, usually no return. The failure produces no data, so the decay produces no statistics.

So when you see a precise figure — 'X% of B2B addresses go bad every year' — treat it with suspicion, and ask where the number came from. As far as we can tell, there is no credible published measurement of B2B physical-address decay the way there is for email. We're not going to invent one. What we can say from the mechanics is directional and confident: every input to the address field decays continuously — the person's tenure, the company's real estate, the mapping between person and building — and none of those changes ever writes back to your CRM. The number is not zero, it grows every quarter the data sits untouched, and — the genuinely uncomfortable part — you cannot compute it for your own list by looking at the list. Staleness is invisible from inside the database. The only way to know if an address is good is to actively verify it, now, against the present.

Silent failure is expensive failure

For email, decay is cheap. A bounced send costs fractions of a cent and instantly cleans your list. For direct mail, decay is expensive twice over — and the second cost is worse than the first.

• The direct cost: every piece printed and shipped to a bad address is real money — printing, materials, postage or courier fees — spent on an envelope that could never have worked. Direct mail is a per-piece-priced channel; waste scales linearly with your decay rate, whatever it is.
• The attribution cost: because wrong addresses fail silently, dead envelopes are indistinguishable from ignored envelopes in your results. If a meaningful slice of a campaign never physically reached anyone, your response rate understates what the channel actually does — and you may kill a play that was working on the accounts it actually reached.
• The follow-up cost: reps making the send real — 'I mailed you something last week' — burn credibility when the prospect never got anything. The rep sounds like every other cold caller inventing a pretext, and worse, they've now spent a call on it.

That second bullet deserves emphasis, because it corrupts decision-making, not just budget. Teams evaluate direct mail on cost per meeting, and every undelivered piece silently inflates that number: same spend, fewer envelopes that ever had a chance. If you model your campaign economics, the delivered-rate assumption is doing more work than almost any other input — which is exactly why it shouldn't be an assumption at all.

Verification-before-print: the buying criterion

All of this points to one conclusion for anyone evaluating direct mail vendors or building the motion in-house: the decisive question is not print quality, formats, or design tooling. It's what happens between your list and the printer. There's an important distinction hiding here — finding an address and verifying an address are different jobs. Finding answers 'what address might work for this contact?' Verifying answers 'is this specific person deliverable at this specific address, today?' A verified business address is one that's been actively confirmed as current and deliverable for that named person — not one that merely parses as a valid location. Plenty of perfectly formatted, postally valid addresses are wrong for the human you're trying to reach.

This is the problem B2BMail's address verification exists to solve, and the sequencing is the point: verification happens before anything prints. Every contact on an uploaded list gets a deliverable business address found and verified first. If an address can't be verified as deliverable, that envelope is never printed and never shipped — the budget simply doesn't get spent on mail that can't land. Decay still exists; it just stops being your cost. And because every piece that does ship travels FedEx with per-piece tracking, delivery stops being a hope and becomes a logged event with a date — which repairs the attribution problem at the same time. You know exactly how many envelopes landed, so your response math finally describes reality.

When you evaluate any vendor — including us — the questions to ask are concrete: Is verification per-contact or company-level? Does it confirm the person still works there, or only that the building exists? What happens to unverifiable contacts — are they skipped, or printed anyway? Is delivery confirmed per piece? A vendor doing real address hygiene will have crisp answers. A vendor whose process is 'we mail what you upload' is quoting you a price that includes an unknown percentage of envelopes to nowhere.

What to do with your list this quarter

You don't need to solve your whole database to act on this. The practical move is small and diagnostic: take a slice of your highest-value accounts and run a pilot through a verification-first process — B2BMail runs pilot programs with no minimums for exactly this purpose. The verification pass itself is informative before a single envelope ships: how many of your contacts' addresses verified as deliverable, and how many didn't, is the closest thing to a measured decay rate for your data that you can get. Some teams find their data is in decent shape. Others discover the silent problem was bigger than any spreadsheet suggested. Either way, you'll finally know — and every envelope you do send will be one that could actually land.

The address-decay problem, in one sentence: business addresses rot continuously through job changes, office moves, HQ-vs-desk mismatches, and hybrid work; the rot is invisible because physical mail fails silently; and the only real defense is refusing to print anything that hasn't been verified as deliverable, to that person, today.