By Lester Laine · October 2026
Every intent platform is built around recency. Bombora's Company Surge, the data behind many of them, compares an account's research on a topic over the past three weeks against its typical level over the past twelve, and refreshes weekly (Bombora/6sense FAQ). The practitioner advice that follows is always the same: act within days, because the window closes.
As a rule for timing a first touch, that's sound. As a rule for deciding which accounts deserve attention, it quietly throws away some of the most valuable information you have.
A signal older than 90 days isn't dead. Most of the time, it's a timestamp.
The buying cycle is longer than the signal window
The latest 6sense Buyer Experience Report (November 2025) puts the average B2B buying cycle at 10.1 months globally and 11.1 months in North America. Buyers make first contact with vendors about 61% of the way through that journey (CustomerThink summary), and the winning vendor was on the Day One shortlist 95% of the time (Corporate Visions summary).
Run the arithmetic on a real case. An account surges on your category in January. Your tools flag it, someone sends a sequence, nothing happens, and by April the account has aged out of every "hot" list. But if that surge marked the start of a ten-month cycle, April is month three or four. The buying group has defined the problem, is building requirements and is drafting a shortlist. On the 6sense numbers, it will start contacting vendors around month six.
The signal didn't expire. The account moved into the phase you can't see, which is exactly the phase where preference forms.
Not every signal decays at the same speed
Applying the same 30- or 90-day decay to every signal is the core mistake. Different signals carry different information and lose value at different rates. I work with three classes.
|
Decay class |
Example signals |
What it still tells you after 90 days |
|---|---|---|
|
Fast (timing) |
Topic surges, pricing-page visits, review-site comparisons |
The account had a live problem in your category. The timing is stale; the context isn't. |
|
Medium (phase) |
New executive hires, functional hiring sprees, funding rounds, a champion changing jobs |
A window of change opened. After the first 90 days, the new leader has made first moves and is now living with the gaps. |
|
Slow (structural) |
Contract and renewal dates, technology installed or removed, regulatory deadlines, fiscal-year planning |
A dated event you can count backward from. These signals gain value as the date approaches. |
Two examples. Job-change data vendors build their pitch on the first 90 days in a role, when new executives make their initial vendor decisions; treat their figures with the care you'd give any vendor's own research. But the same logic implies something they rarely sell: on day 120, that executive has inherited problems their first decisions didn't solve. That's a second, quieter window.
A funding round works the same way. The announcement is the loud signal. Deploying the money takes quarters, and the hiring it funds produces a new wave of signals six to twelve months later.
Signal archaeology: a method for old signals
This is the process I run with clients before they buy another data source.
- Pull every signal older than 90 days, at account level. Intent surges, past visits from identified companies, webinar attendance, old form fills, stalled proposals, closed-lost opportunities marked "timing" or "no decision."
- Date and classify each one by decay class.
- Project the timeline. Signal date plus the typical cycle length for that deal size gives an estimated phase today: problem definition, requirements, shortlist, decision or post-purchase.
- Look for a fresh corroborating signal. A return visit after silence, a new hire in the relevant function, a job post that disappeared, a budget-cycle date. An old signal plus a fresh one is worth more than either alone.
- Match the message to the phase, not the signal. Never open with "I saw you were researching X four months ago." It's stale and it's creepy. If the account is probably shortlisting, send what shortlisters need: evaluation criteria, a procurement pack, a reference in their industry. If they likely bought from someone else, note the probable renewal date and plan around it.
The RevOps problem: most CRMs erase history
Signal archaeology only works if the history exists. Most CRMs store intent as fields: last intent date, current score, top topic. Every update overwrites the previous value. Six months later you know the account is "cold," but not that it surged twice, attended a webinar in between, and hired a new CFO in the meantime.
The fix is structural. Store signals as timestamped events on the account timeline, not as overwritable properties. Keep the raw date, the source and the type, then build scoring on top of the history instead of the latest value. It's an unglamorous change. It's also the difference between a CRM that remembers and one that only knows today.
Why this matters even more for small teams
Enterprise teams can pay 6sense or Demandbase to surface fresh signals every week. A 15-person B2B company usually can't, and doesn't need to. Its most valuable signal archive is already in its own systems: last year's webinar attendees, proposals that went quiet, deals lost to "timing," the inbound lead that wasn't ready.
That archive is first-party, it costs nothing and nobody else has it. A competitor can buy the same intent feed you do. They can't buy your history.
In SCALING™ terms, this is Nurture done properly: a prospect who wasn't ready in month one shouldn't be discarded, they should be tracked until they are. It's also Integration: marketing, sales and the data layer agreeing that an account's past is part of its present.
One caution. In Europe, behavioral data about identifiable people is personal data. Keep signal history at account level wherever you can, respect retention limits and never reference tracked behavior in a message. Use it to decide what to say, not to show what you know.
Want the full report?
The Signal Archaeology Kit includes decay classifications for 14 signal types, a timeline-projection worksheet by deal size and cycle length, the CRM event schema that stops signal history from being overwritten, and message templates matched to each buying phase.
Send me a message on LinkedIn with the word STALE and it arrives right away.
Sources
- 6sense FAQ: Using Bombora Surge Data — Bombora / 6sense
- 2025 B2B Buyer Experience Report — 6sense, November 2025
- Research Round-Up: 6sense 2025 study — CustomerThink
- B2B Buying Behavior in 2026: 57 Stats and Five Hard Truths — Corporate Visions