A short algorithm for owners: how to tell a channel that needs tuning from one that's just burning budget and distracting the team.
Businesses either kill a channel too early or keep it too long due to sunk cost. The reason is the same: there's no pre-agreed threshold for when a channel is considered working or dead.
One of the hardest decisions in B2B marketing: keep investing in a channel that hasn't delivered expected results yet, or stop and reallocate the budget. Stop too early—you won't see a return on what you've put in. Wait too long—you're burning budget on a channel that doesn't fit your ICP.
A "scale or kill" decision can't rest on gut feel or a time horizon ("it's been a month, it doesn't work"). It has to rest on three concrete signals.
3 signals: scale or kill
CPL (cost per lead) ≤ target value AND CAC (customer acquisition cost) ≤ 1/3 LTV. If these numbers are in check, the channel is working—increase the budget.
Example: CPL $50, lead-to-customer conversion 15%, CAC = $333, LTV = $2,000. CAC/LTV = 16.6% — good, scale it.
After optimization (A/B testing ads, changing audiences, new landing pages), CPL drops or conversion rises. If there's a vector for improvement, it's worth continuing.
Signs: outreach response rates are up, MQL→SQL conversion improved, sales says leads feel "warm."
CPL is 2+ times above target after 3+ optimizations. Or leads come in, but none convert to a deal after 20+ qualifications. Or a channel only produces the right leads when combined with another channel — then it's an amplifier, not a standalone channel.
Unit Economics for a channel decision: how to calculate
Core formulas every B2B marketer should know:
CPL = Channel spend / Number of leads
CAC = Channel spend / Number of new customers
LTV = Average deal size × Number of deals per client (or ARPU × Cooperation period)
CAC/LTV ratio — target: <33%. If >50% — the channel is unprofitable.
AI gathers data faster: how to speed up analysis
Traditionally, channel analysis takes weeks — you have to collect data from different sources, consolidate it into a table, and calculate. AI accelerates this cycle:
- Automatic data aggregation: n8n or Zapier collects metrics from Google Ads, Meta, and CRM daily and consolidates them into one dashboard. No manual copying.
- Anomalies and signals: AI notices when CPL spikes or conversion drops — and notifies the team instead of waiting for a weekly report.
- A/B test analysis: AI determines statistical significance faster and more accurately than manual calculation. Test decisions in 7 days instead of 30.
A framework for the decision: 4 questions before "scale or kill"
- 1Does CPL match the target? If not — how many optimizations have been made and is there a vector for improvement?
- 2What leads does the channel bring? Are sales reps happy with quality? Is MQL→SQL conversion higher or lower than other channels?
- 3Is there enough data? At least 50 leads for a statistically significant conclusion. From 10 leads, you can't conclude the channel doesn't work.
- 4What do non-converting leads say? If the problem is "not our ICP," that's a targeting issue. If it's "too expensive," that's an offer or segment problem.
Quick diagnosis
Check off the items you already have. This doesn't replace an audit, but it quickly shows how much control you have over the topic.
Entity block by topic
For a page to work for both people and search, it's important to explicitly name key entities and concepts. This makes the topic denser and the solution clearer, without marketing fog.