A contact list is a snapshot of a population that keeps moving. The bill for that movement usually arrives as a bounce report, a quarter late.

The awkward part is that the movement is measurable in advance. Some of the inputs are public statistics, and they cost nothing to look up.

Two populations go stale at two different speeds

Most list-decay conversations treat staleness as one problem. It is at least two, and they have different remedies.

The first is people. The Bureau of Labor Statistics publishes the Job Openings and Labor Turnover Survey, which tracks hires and separations — including quits, layoffs and discharges, and other separations — as a measure of how much movement occurs in the labor market each month.

The second is establishments. BLS's Business Employment Dynamics program measures gross job gains and losses arising from establishment births, expansions, contractions and deaths. That is a different kind of churn: the company on your list may not have changed its contact, it may have stopped existing as an operating entity, been absorbed into a parent, or opened a second location that changes who buys what.

Neither series is a contact-data decay metric. Neither vendor nor buyer should present them as one. What they do is replace a guess with a shape: person-level records rot continuously, and company-level records rot in discrete events that cluster around funding, acquisition, restructuring and closure.

A "verified" badge is a timestamp, not a duration

This section is analysis.

Data providers compete on verification, and verification is genuinely valuable. But a verification badge answers one question: when was this record last checked? It says nothing about how long that answer stays true.

A record confirmed in March is a March fact. The question that determines your bounce rate is not whether the list was verified but what the refresh interval is for this specific segment, and who is accountable when a record crosses it.

That interval should not be uniform. A segment of long-tenured staff at stable enterprises and a segment of recently funded startups hiring aggressively have different turnover profiles, and pricing both on the same cadence is how budgets get spent on the wrong records.

Signals are evidence with a shelf life

Buying signals are useful precisely because they are dated. A filing has a date. A job posting has a date. A hiring surge has a beginning and, if you are paying attention, an end.

That is also where signal-led prospecting goes wrong. A signal that is not timestamped, sourced and explained becomes a score, and a score without its inputs is an opinion in a numeric font.

SEC EDGAR is the cleanest example of a well-instrumented source and of why instrumentation still needs reading. The commission publishes filings with dates and identifiers, which is what makes them useful as evidence. Its own full-text search documentation is equally clear that the search is a conceptual tool rather than an exact-match index, so a hit is a lead to read, not a finding to cite.

Computing your own decay rate (illustrative)

The calculation below is a worked illustration with hypothetical inputs, not a benchmark or a measured result. Its purpose is to show the arithmetic a data contract should be argued with.

Take a segment of 1,000 contacts. Suppose 20% of them change role or employer within twelve months, and that of those, half leave the buying function entirely rather than moving to a comparable role elsewhere. That leaves 100 records that are no longer reachable in the role they were pulled for, before any name or address error is counted.

Now apply a refresh cadence. If that segment is re-verified quarterly, the average record is roughly six weeks out of date at any moment, and the tail is worse. If it is re-verified annually, the average age is six months.

The useful output is not the number. It is the ratio between the refresh interval you are paying for and the turnover you know the segment experiences. When that ratio is more than a few months, the list is doing less work than the invoice implies.

Qualification is a record, not a score

Most qualification debates collapse into scoring arguments. The more durable question is what a record has to carry before anyone calls it an opportunity.

At minimum: the event that caused the contact to surface, the date of that event, the primary source it came from, what the source actually said, and what remains unknown. A record missing those fields is a name with a confidence level attached.

That standard is uncomfortable because it produces a shorter list. It also produces a list where the first sentence of an outreach note would survive being read back to the buyer.

Four questions before you renew a data contract

  • What is the refresh interval per segment? Not the overall figure. Ask for the cadence applied to the segments you actually buy, and get it in writing.
  • What is the provenance of a signal? For any signal-based feature, ask which source produced it, when, and whether you can see the underlying event rather than a score.
  • What happens to records that fail re-verification? Suppression, enrichment, or silent retention? The third answer is the expensive one.
  • Who owns the corrected record? If enrichment overwrites a field your team edited, you have a data governance problem wearing a product's clothes.

A 30-day check

  • Days 1 to 5. Pull one segment and record the last-verified date for every contact. Count how many are older than your assumed cadence.
  • Days 6 to 10. Take 50 contacts and check current role and employer against a primary source. That ratio is your working decay estimate for the segment.
  • Days 11 to 15. Classify the failures: left the company, changed role, company closed or acquired, or never existed as recorded. The distribution tells you which fix matters.
  • Days 16 to 22. Ask the vendor for the per-segment refresh cadence and the provenance of one signal you rely on.
  • Days 23 to 30. Re-price the contract against the ratio you measured in days 1 to 10. If nothing changes, you now have the number in writing.

What remains unestablished

  • Any specific separations or establishment-churn figure for your industry or segment. Current BLS values were not retrieved for this edition; only the series definitions are used.
  • Whether any particular vendor's refresh cadence matches the churn its customers actually experience.
  • Whether signal-led prospecting produces better conversion than list-based prospecting in any specific market. No such measurement was reviewed here.
  • How much of typical contact decay is role change versus employer change versus data-entry error. That split is segment-specific and unmeasured in this edition.
  • Market data. No stock, index or valuation figures were reviewed.

Evidence notes and limits

The Bureau of Labor Statistics' Job Openings and Labor Turnover Survey page is the source for the described scope of that series: hires and separations, including quits, layoffs and discharges, and other separations, reported as monthly labor-market movement. No specific rate or level from the series is asserted.

BLS's Business Employment Dynamics page is the source for the described scope of gross job gains and losses arising from establishment births, expansions, contractions and deaths. No specific value from that series is asserted, and it is explicitly not presented as a contact-data decay metric.

The U.S. Securities and Exchange Commission's EDGAR full-text search documentation is cited for two propositions only: that filings carry dates and identifiers that make them usable as dated evidence, and that the full-text search is a conceptual rather than exact-match tool, so a returned document must be read rather than relied on as a match. No filing, company or enforcement matter is cited.

No vendor pricing page, contract, bounce report, audit or third-party test was reviewed. The two-population framing, the worksheet and its hypothetical inputs, the record standard, the four questions and the 30-day check are this publication's analysis. The worked figures are illustrative arithmetic, not measurements. Nothing here establishes that any organization improved conversion, reduced bounce rates or recovered staff time.

The advertisement above is separate from this analysis and is not a source for any claim in it.