B2B Data Decay: Why Prospect Lists Bounce and the Fix

B2B contact data goes stale at about 22.5% a year on average, and vendor reports put tech at 30% to 40%. Stale does not mean bounced. A verification waterfall checks each address through several independent sources before any send, removes the dead ones and holds bounces under 2%. Blesso runs this step inside its outbound service.

A 30% decay rate does not mean a 30% bounce rate

The two numbers measure different things. Decay counts records in your database where something changed: a job title, a phone number, an email address. Bounce counts messages the receiving server refused. A contact who changed title but kept the same mailbox adds to decay and never bounces.

Published figures show the gap. Cleanlist, a verification vendor, reports that unverified or purchased B2B lists bounce at 10% to 20%. Pintel, a data vendor, reports 8% to 15% for a list not cleaned in 12 months. Both sell fixes, and neither is an independent study. Read them as a range, not a promise.

The email-only decay figure is closer to the bounce driver. ZeroBounce analyzed more than 11 billion verified emails from January to December 2025 and found that at least 23% of an email list degrades each year, down from 28% in 2024, as Prospeo summarizes. That is the number the rest of this guide uses.

Tech contacts go stale faster than the average list

The widely quoted baseline is 2.1% a month, or 22.5% a year, from MarketingSherpa through HubSpot's database decay tool. The Landbase roundup of decay studies gives a wider range of 22.5% to 70.3% a year, depending on how many fields count as wrong. The top of that range is an any-field-wrong figure, so do not use it for email.

For tech, Cleanlist cites 30% to 40% a year and Pintel cites 35% to 40%. Pintel gives 50% to 70% for recruitment and staffing. All of these are vendor numbers with no public method, so treat them as planning inputs. The cause is easy to follow: people in fast-hiring companies change jobs, teams reorganize, and a departed employee's mailbox is shut down. The next send to it is a hard bounce.

Bounces cost you the whole sending domain, not one message

Mailbox providers score the domain you send from. A high bounce rate tells them you do not know who you are mailing, and the penalty applies to every message from that domain, including mail to valid contacts.

The published rules are looser than the 2% guardrail people quote. Google and Yahoo name a spam complaint rate, not a bounce rate: keep spam under 0.3%, and Google says to aim below 0.10% in its email sender guidelines. Yahoo's sender best practices tell senders to monitor hard and soft bounces and remove invalid recipients promptly. Spamhaus, which runs the best-known blocklists, says in its sender guidelines to keep the overall bounce rate below 5%.

So 2% is a working target, not a published line. It leaves room for the addresses no verifier can judge and keeps you well away from the 5% mark. Spamhaus also warns that purchased and harvested lists are the most common source of spam trap hits, and that verification does not clean spam traps.

Build the verification waterfall in five stages

A waterfall passes each address through checks in order, cheapest first. Each stage runs only on what the previous stage could not settle. The word comes from waterfall enrichment, where several data providers are queried in sequence until one returns an answer. Verification uses the same idea with the aim of catching what a single provider misses.

1. Dedupe and syntax check. Merge duplicates, drop malformed addresses and remove role accounts such as info@ and sales@. This costs almost nothing and shrinks every later stage.
2. Domain and MX check. Confirm the domain exists, has mail servers and is not a disposable-address service. Companies acquired or shut down show up here.
3. Primary mailbox verifier. Ask the receiving server whether the mailbox exists, without sending a message. This stage removes most dead addresses. It cannot judge catch-all domains, which accept every address.
4. Second source on everything left unsure. Run risky, unknown and catch-all results through a different verifier or a different data provider. Two tools that disagree are the signal. Two that use the same upstream data add little.
5. Freshness check. Drop or re-source any contact whose record is older than 90 days, or whose job has changed on LinkedIn or the company site. A mailbox can still exist while the person is gone.

A worked model: 10,000 contacts, 12 months old

The numbers below are an estimate we built to show the mechanism. They are not measured results, and they are not from any Blesso campaign. Change the assumptions to match your list.

Assumptions:

• The list has 10,000 emails, none verified for 12 months.
• 23% are invalid (the ZeroBounce email figure), so 2,300 would hard-bounce.
• Each stage catches a stated share of the invalid addresses still left: stage 1 catches 15%, stage 2 catches 90%, stage 3 catches 60% and stage 4 catches 50%. These catch rates are our assumptions, not vendor claims.
• Caught addresses are removed before sending. The bounce rate is remaining invalid addresses divided by addresses sent.

The math for stage 2: 1,955 invalid addresses left times 90% is 1,760 removed, which leaves 195. Divide 195 by the 7,895 addresses still on the list and you get 2.47%.

Two results matter. First, one verifier at a 90% catch rate does not get you under 2%. Second, the cost is real: 2,261 of 10,000 contacts, about 23%, never get mailed. That is the decay showing up as lost reach instead of damaged reputation.

Catch-all addresses are the part a waterfall cannot settle

A catch-all domain accepts mail for any address, so a verifier cannot tell a real mailbox from a made-up one. ZeroBounce data cited by Prospeo puts catch-alls at 9% or more of checks. In the model above they sit inside the 7,739 sendable addresses.

Here is the sensitivity. Say 900 of the 10,000 are catch-all, and 5% of those bounce. That adds 45 bounces to the 39, for 84 in total, or 1.09% of 7,739 sends. The rate stays under 2%, but the risk now sits in one segment. Send those addresses last, in small batches, and watch the bounce rate by segment before you open the rest.

Re-verify every 30 days, not once a year

Verification is a snapshot. At 23% a year, about 1.9% of a list's addresses go bad each month (23 divided by 12). That is nearly the whole 2% budget. A list verified once and mailed over the next three months drifts by about 5.8% before the last send, using the same straight-line math.

For active sequences, three habits keep the rate down:

• Verify again within 30 days before the first send of any batch, and again before each follow-up wave that is more than 30 days later.
• Suppress every hard bounce at once and never mail it again. Spamhaus recommends this, and recycled bounced addresses are a known source of spam traps.
• Track bounce rate by list source and record age. A single bad source shows up there first.

How Blesso handles list decay

Blesso runs outbound for B2B teams as a done-for-you service. Its Outbound Engine builds lists past the standard databases with custom scrapers written per campaign, then runs waterfall enrichment after dedupe, with deliverability included. Replies trace back to the signal that triggered the send.

Blesso has published no bounce-rate figures, so this article claims none. Its published customers, such as HappierHire and Airbills, sell into hiring and staffing, a segment where Pintel reports some of the fastest decay. That makes fresh signals matter more than a big list. See the three motions or published results. Engagements start at $2k a month, with a three-month minimum.

Want list decay handled with the rest of outbound? Book a call.