Analyze · Cohort & Channel ROI

Diagnose a Declining Channel's Performance

Decompose a single channel's ROI drop into the exact metric that broke (spend, leads, conversion, deal size) and prescribe the fix.

foundermanagerAdvanced⏱ 3-4 hours
When to use
Use when one channel suddenly underperforms quarter-over-quarter and you need to know why before you cut budget. Isolates the broken link in the funnel.
The prompt
You are a growth analyst for a digital marketing agency analyzing the agency's own GTM ROI. You diagnose channel underperformance by decomposing the funnel.
Agency: [AGENCY_NAME] — [SERVICES] | Channel: [CHANNEL_NAME] | Periods compared: [PERIOD_A] vs [PERIOD_B] | Data:
[CHANNEL_DATA] (per period: spend, impressions/sends, leads, SQLs, closed-won, avg MRR, CPL, CVR, CAC)
Decompose the ROI decline between the two periods. Quantify how much of the CAC change came from spend efficiency, lead volume, lead→SQL rate, SQL→CW rate, and deal size. Recommend a single highest-leverage fix.

- Show formulas: CAC = spend / CW; CPL = spend / leads; lead→SQL %, SQL→CW %; AOV change
- Express each driver's contribution to CAC change in % points
- Flag any metric where sample size dropped >40% as "may be statistical noise"
- Prescribe ONE fix, not five — name the metric to move and by how much

1) Period A vs Period B table. 2) Driver decomposition (waterfall: spend efficiency / volume / lead quality / sales conversion / deal size). 3) Diagnosis sentence. 4) Single recommended fix with target metric.
Variables
  • [AGENCY_NAME] — Your agency name
  • [SERVICES] — Services sold
  • [CHANNEL_NAME] — The channel under review
  • [PERIOD_A] — Healthy/baseline period
  • [PERIOD_B] — Underperforming period
  • [CHANNEL_DATA] — Funnel metrics for each period
Example input
Agency: Sable Search — SEO retainers | Channel: Google Search Ads | Q4 2025 vs Q1 2026 | Data:
Metric | Q4 2025 | Q1 2026
Spend | $12,000 | $13,000
Impressions | 86k | 92k
Leads | 64 | 58
SQLs | 41 | 22
Closed-won | 7 | 3
Avg MRR | $4,400 | $3,800
CPL | $187.50 | $224.13
Lead→SQL | 64% | 38%
SQL→CW | 17% | 14%
CAC | $1,714 | $4,333
Example output
Period A vs B: CAC inflated from $1,714 to $4,333 (+153%).

Driver decomposition (contribution to CAC change):
- Spend efficiency (CPL up 20%): contributes ~$340 of inflation
- Lead volume (-9%): minor, contributes ~$160
- Lead quality (Lead→SQL collapsed 64%→38%): contributes ~$1,650 — DOMINANT driver
- Sales conversion (SQL→CW 17%→14%): contributes ~$310
- Deal size ($4.4k→$3.8k MRR): not a CAC driver but reduces LTV:CAC by 14%

Diagnosis: This is a lead-quality problem, not a spend or sales problem. Lead→SQL fell 26pp — the channel is delivering more bottom-of-form, less-qualified leads despite similar impressions.

Fix: Audit Q1 keyword + audience changes — most likely cause is broad-match expansion or a new landing page that lowered intake friction. Target: restore Lead→SQL to ≥55% within 30 days. Expected CAC recovery: from $4,333 back to ~$2,200.
Pro tips
  • Always compare two periods of equal length — a 90-day vs 30-day comparison hides volume effects.
  • If lead quality is the driver (most common), the fix is usually targeting, not bidding.
  • Don't cut spend on a declining channel until you've isolated the broken metric — you may be cutting the wrong lever.
Works with
ClaudeChatGPTGemini
Done with prompts? Time to install the system
Book a STAOS call
Related prompts