Hey 👋
I was looking at a list of "hot" accounts that were still sitting in a queue three days after the signal.
What struck me wasn't the tools. The tools are fine.
It was that we were treating a depreciating asset like inventory that ages in reverse.
A pricing-page visit, a profile view from the right title, a hiring post that screams GTM pain: those aren't leads. They're timers. By the time someone finally wrote the note, the reason to write it had already cooled.
Last week was booked isn't held. This is one step upstream: the signal that expired while you were still "getting to it."
Here’s what that actually looks like.
TL;DR
Pricing / demo-page intent: ~24-hour half-life; stop new plays past ~72 hours (Unify signal decay table, 2026)
PQL ~5 days; new hire ~14 days; funding (founder/CFO) ~60 days. One cadence for all of them wastes half the funnel
Cold email reply → booked meeting often ~15–30% once someone answers (LeadAngel outbound index, 2026)
Instant scheduling at the form: qualified → booked ~66.7% vs ~30% baseline without it (~4M submissions, Chili Piper 2025)
Speed lore is real but mis-cited: HBR 2011 = 42h avg among responders, 23% never reply, ~7× qualify in the first hour. The 5-min / 21× figure is InsideSales/MIT 2007, not HBR (Elev8 audit, 2026)
This week's move: stamp signal type + age-at-send on every outreach. Kill past 2× half-life. Cap hot approve/send queues at same business day.
Weekly finds
Every buying signal has a half-life. Most teams ignore it. Unify’s table puts pricing/demo intent at ~24 hours and funding at ~60–90 days, so one cadence for both spends most of its life talking to a ghost reason.
The speed-to-lead numbers people quote are usually the wrong study. Elev8’s 2026 audit: HBR 2011 is the 42h / ~7× first-hour paper; the 5-min / 21× figure is InsideSales/MIT 2007, not HBR.
Reply rate is not meeting rate. LeadAngel’s outbound slice: sequence → meeting often ~1–4%, while reply → meeting often ~15–30%, so healthy replies with empty calendars point at reply handling, not the opener.
Kill the gap between "I want to talk" and a slot. Chili Piper on ~4M submissions: qualified → booked ~66.7% with instant scheduling vs ~30% baseline without it.
Salesforce in Claude shipped as beta with 37 seller skills. Anthropic + Salesforce put approve-before-write CRM drafts under the seller’s permissions, which speeds admin without rescuing a three-day-old signal.
THE ISSUE
The Signal Had a Half-Life
You'll like this one if your CRM is full of "warm" rows that haven't earned a reply in weeks.
Signal-led outbound sells a story that feels modern: act when they're already moving.
The leak is quieter. You detect the move. Then the note waits. Approval waits. "I'll personalize it tonight" waits. By send time, you're referencing a moment they already finished.
You're not late to the market. You're late to the timer.
Intent is a depreciating asset
A signal is not a lead. A lead is a person. A signal is a claim that right now they care about something you can talk about.
Unify's table is blunt on commercial pages: pricing/demo intent burns half its predictive value in about a day. Funding and stack changes give you weeks. New-hire openers die faster than the honeymoon window for a champion who just changed jobs.
One cadence for all of that means a 30-day sequence on a 24-hour asset. The first touch might still land. Touches two through seven talk about a reason that already left the building.
Practical: Label every play with signal type and a hard stop at 2× half-life. Past the stop, archive for re-eligibility. Do not delete. Do not re-enroll on Monday because Monday happened.
Age-at-send is the KPI next to reply rate
Reply rate without age is vanity with better lighting.
A note six hours after a hiring post is a different product from the same note six days later. Same words. Different goods. Same CRM cell.
Teams report "personalized outbound" and hide the lag. The lag is the personalization. Without signal → send timestamps, you cannot tell whether the copy worked or the timing did.
Practical: check when your signal date landed. Review the median weekly. If hot signals sit past one business day, the queue is the bug, not the template.
Soft queues manufacture stale outreach
Founder-led teams are good at this. You care about the message. You leave the draft overnight. The signal does not wait for your taste.
Approval is healthy. An unbounded approval queue is a half-life machine. Same for "batch Friday." Hot signals do not batch.
Chili Piper's form data is the inbound twin: remove the waiting room and qualified → booked roughly doubles versus the baseline that waits for someone to "get back to them." Outbound has the same waiting room. It lives in your head and your tools.
Practical: Cap hot-signal drafts at same business day. Unsent drafts don't get a prettier rewrite next week. Kill them, or rewrite only if a new signal fires.
Reply handling is the other half of the timer
Getting the reply is not the meeting.
LeadAngel puts reply → meeting in the 15–30% range once someone answers.
Sequence → meeting looks tiny because most of the list never becomes a real conversation.
When replies look fine and slots don't, stop rewriting openers. Fix the next message: one clear ask, two times, short window.
Elev8's audit is useful because it stops you from mis-citing HBR for the five-minute rule. Direction unchanged: hours beat days. Days beat "when I have a free afternoon."
Practical: For every positive reply this week, measure hours to first calendar ask. Target under one business day. No multi-email negotiation before a slot exists.
What we change
No new stack required.
Stamp the signal. Type + detected-at on every row you touch.
Match the cadence. 24h signals get same-day plays. Funding gets weeks. Don't mix.
Cap the queue. Hot drafts leave the same day or die.
Ask for the slot early. Reply → two times → done. Chat is not a meeting.
Report age-at-send. Next to reply rate. Or stop celebrating reply rate.
We've watched teams celebrate "signal coverage" while sending notes about last Tuesday's intent on Friday afternoon. The instinct is more signals. The fix is treating freshness like inventory that spoils.
The match wasn't the finish line. Neither was the invite. The timer started when they moved.
Practical: This week, stamp signal type and age on every send. Kill anything past 2× half-life. Empty the hot approve queue before end of day. Compare reply rate on same-day sends vs 3+ day sends. The spreadsheet usually shuts up.
