The 48‑Hour Heat Score: Rebalancing LinkedIn Budgets From CRM Signals (No New Creative Required)
Most teams set LinkedIn budgets quarterly. Buyers heat up and cool down daily.
That disconnect is killing ROAS.
We’ve been running a simple system that moves LinkedIn budget every 48 hours based on live CRM intent. No new creative. No new offers. Just better allocation.
It’s boring. It works.
The problem we kept seeing
- Static budget splits (60% prospecting, 40% retargeting) regardless of what’s actually heating up in CRM.
- Persona targeting with no stage context—treating a pricing‑page visitor the same as a new cold profile who matches a title.
- Creative debates masking a budget routing failure. Teams keep swapping ads while 30%+ of spend sits on people who aren’t ready.
LinkedIn is expensive. We can’t afford set‑and‑forget.
The play: the 48‑hour heat score loop
We connect LinkedIn delivery to a lightweight CRM “Heat Score” that reflects real buying motion. Then we route budget to the hottest pockets of demand—fast.
Step 1: Build a simple CRM heat score
Score at the account or contact level. Keep it simple and decaying.
Example model (customize weights to your funnel):
- +6: Last 7‑day session on pricing, ROI, or comparison pages
- +5: Product doc or case study view
- +4: Responded to SDR or AE email/LinkedIn
- +8: Lead form submitted (high‑intent offer)
- +12: Meeting booked or rescheduled within 14 days
- +15: Opportunity moved forward a stage
- −50: Closed‑lost in last 30 days (cooldown)
Apply a 7‑day half‑life so old actions fade quickly. Recompute daily.
Segment into tiers:
- Hot: score ≥ 12
- Warm: 6–11
- Cold: ≤ 5
Step 2: Build audiences that mirror the score
Create three matched audiences and keep them fresh:
- Hot List: emails/accounts with score ≥ 12 (upload or sync)
- Warm List: 6–11
- Cold List: 0–5
Layer in website retargeting and video viewers for each tier to expand reach where it makes sense. Exclude current customers, open opps, and disqualified in last 90 days across all campaigns.
Step 3: Map creative to intent (don’t overthink it)
- Hot: direct “Book a demo,” ROI one‑pagers, competitor comparisons, proof snapshots.
- Warm: product explainer, customer stories, ungated document ads that tease value, webinar replays.
- Cold: problem framing, POV content, category primers, founder narrative.
You likely have this content already. Reuse it. The win here is allocation, not new assets.
Step 4: Build the three‑lane campaign setup
Three lanes. Clean measurement. No audience overlap.
- Lane A — Hot: target Hot List only (narrow), optimize for conversions or lead forms.
- Lane B — Warm: Warm List + website visits last 30–60 days, optimize for conversions or leads.
- Lane C — Cold: persona/ABM lists minus Hot/Warm, optimize for reach or landing page views.
Bidding (guideline):
- Hot: cost cap (guard CPA), small daily budgets that can scale on good days.
- Warm: max delivery or a moderate cost cap to prioritize volume.
- Cold: max delivery with tighter frequency monitoring.
Step 5: Budget ladders that react every 48 hours
Every two days, pull three metrics by lane:
- Meetings per 1,000 impressions (M/PM)
- Cost per quality meeting (CPQM) or cost per SQO if you prefer
- 7‑day frequency
Apply simple rules:
- If Hot M/PM is ≥ 2x Warm and frequency < 7, shift +15–25% budget into Hot.
- If Warm CPQM is within 20% of Hot and frequency < 5, move +10–15% into Warm from Cold.
- If a lane’s frequency > 8 and M/PM is flat for 4 checks (one week), cut its budget by 20% and rotate creative.
- If Cold CPMs are unusually cheap and click quality is rising (LP view rate up, bounce down), drip +5–10% into Cold for future Warm fill.
No need to be heroic. Small, consistent rebalances compound.
Step 6: Control frequency without mythical knobs
LinkedIn doesn’t hand us perfect frequency caps. We manage it indirectly:
- Split creatives into multiple ad sets to spread delivery.
- Rotate new variants every 10–12 impressions per member.
- Trim low‑CTR ads fast in Hot/Warm to protect CPMs.
- Reduce audience overlap (strict exclusions) so lanes don’t fight each other.
Step 7: Attribute like an adult
Measure cohorts by first heat tier seen, then follow to opportunity:
- Meetings per 1,000 impressions by lane
- Cost per meeting and cost per SQO by lane
- Pipeline per 1,000 impressions (Pipeline/PM)
- Win rate and sales cycle by lane
Tie every campaign to UTM’d CRM campaigns. Treat “follows,” “likes,” and “view‑throughs” as supporting signals—valuable, but not the scoreboard.
A short case breakdown
Context: PLG + sales‑assist SaaS, mid‑market focus, average contract value meaningful enough that paid social can work. Prior setup was static: 70% persona prospecting, 30% retargeting. Creative library was solid but performance plateaued.
What changed:
- Implemented the heat score with 7‑day decay
- Built Hot/Warm/Cold lanes and exclusions
- Rebalanced every 48 hours with the rules above
90‑day snapshot after stabilization:
- Meetings per 1,000 impressions: Hot 3.4, Warm 1.8, Cold 0.4
- CPQM: Hot down ~37%, Warm down ~22%, Cold flat (as expected)
- Pipeline/PM up ~2.1x overall versus the static‑budget baseline
- Spend mix moved from 70/30 (cold/ret) to roughly 20/50/30 (cold/warm/hot) during peak demand weeks
No new creative. Same offers. Faster money.
Your mileage will vary; the principle holds. Allocation beats artistry when intent is moving quickly.
Why this works (and keeps working)
- Recency matters. A pricing‑page visit from yesterday is not the same as a persona lookalike from last quarter.
- LinkedIn delivery likes clarity. Clean lanes with tight exclusions reduce internal auction fights.
- Decay avoids chasing ghosts. Last month’s webinar watcher isn’t today’s buyer.
- Small, frequent moves compound. You don’t need perfect signals—just consistent ones.
Common failure points to avoid
- Over‑engineering the score. Start with 5–7 signals. Ship. Tune later.
- Bloated Warm lists. If everyone is Warm, no one is. Tighten thresholds.
- Creative mismatch. Don’t show a vision piece to Hot and a hard CTA to Cold.
- Ignoring frequency. Rising frequency with flat meetings is budget begging for a new home.
- Treating “followers” as success. Followers are a means to cheaper reach later, not a goal today.
Advanced add‑ons (when the core loop is stable)
- Document Ad filter: run ungated document ads in Warm to qualify interest cheaply, then promote only those who click or save into Hot.
- Job‑change micro‑lane: spin up a tiny lane for champions who changed companies in the last 90 days (fresh budgets, new tools). Keep it separate from core three lanes.
- Dayparting by SDR coverage: during hours when humans can respond fast, nudge +10–15% into Hot/Warm. Pull back when your team sleeps. Manual works.
The 48‑hour checklist
- Recompute heat scores (decay applied)
- Refresh Hot/Warm/Cold audiences (sync or upload)
- Pull M/PM, CPQM, frequency by lane
- Shift budgets per rules (+/− 10–25%)
- Pause underperforming creatives, rotate 1–2 new variants if needed
- Log changes and next check’s hypothesis
15 minutes. Every other day. No new assets.
Final takeaway
If we accept that intent is fluid, then budgets must be fluid.
LinkedIn won’t get cheaper. But it can get smarter when CRM tells it where to look.
Want the heat score template and the rebalance rules as a one‑pager? Drop “HEAT” and we’ll share the worksheet we use to run this loop across accounts.