The Funnel Intelligence Playbook: How Small GTM Teams Diagnose Pipeline Bottlenecks Without a Data Team
The Funnel Intelligence Playbook: How Small GTM Teams Diagnose Pipeline Bottlenecks Without a Data Team
You run a lean GTM team — maybe it’s just you, a co-founder, and a part-time contractor. You have 25 tools, a CRM half-updated, and a nagging feeling that leads are slipping through cracks you can’t even see. A full-time data analyst isn’t in the budget. But you still need to know: Where is my pipeline breaking?
Enter Funnel Intelligence — a systematic, no-fluff approach to diagnosing bottlenecks using metrics you already have, a simple composite score, and a prioritization framework that doesn’t require a statistics degree. This playbook is built for small teams that need to move fast, without drowning in spreadsheets.
---
1. The Funnel Health Composite Score: Your Single Source of Truth
Stop tracking 40 vanity metrics. Instead, measure one number that tells you where to focus.
The Funnel Health Composite Score is a weighted combination of three funnel stages against a rolling 30-day baseline. Here’s the formula:
Funnel Health = (TOFU Score × 0.25) + (MOFU Score × 0.40) + (BOFU Score × 0.35)
Each stage score is calculated by comparing current performance to your 30-day average. A score of 1.0 means “on baseline.” Below 1.0 means underperforming.
- TOFU (Top of Funnel) – Measured by Search Console impressions & clicks. This tells you if enough people are discovering you.
- MOFU (Middle of Funnel) – Measured by GA4 engaged sessions. Are the right visitors sticking around and exploring?
- BOFU (Bottom of Funnel) – Measured by GA4 conversions (signups, demos, trial starts). Are you actually closing the deal?
Why this works for small teams: No data pipeline required. These metrics are free in Google Search Console and Google Analytics. You can pull them weekly in 10 minutes.
Example: If your TOFU score is 0.8 (20% below baseline), but MOFU is 1.1 and BOFU is 0.9, the composite is: (0.8×0.25) + (1.1×0.40) + (0.9×0.35) = 0.2 + 0.44 + 0.315 = 0.955 — below 1.0. The bottleneck is TOFU (lowest weighted score). Action: invest in awareness.
But here’s the nuance — if a stage is lower than its weight in the composite, that’s your bottleneck, regardless of the absolute number. In the example, TOFU is 0.8 with a 0.25 weight — it’s dragging the whole score down.
---
2. Diagnosing Bottlenecks in Three Clicks
Once you compute your composite, use this decision tree:
| If the lowest stage is… | Then the bottleneck is… | Typical root cause | |--------------------------|--------------------------|-------------------| | TOFU | Awareness | Low content reach, poor SEO, wrong targeting | | MOFU | Engagement | Weak messaging, slow site, no lead magnet | | BOFU | Conversion | Friction in signup, unclear value prop, no urgency |
Run this every two weeks. Trends matter more than single data points.
Pro tip: Use a simple dashboard tool (or even a Google Sheet) to track the composite over time. If the score drops below 0.9 for two consecutive periods, that’s a yellow flag. Below 0.8? Red flag — pause everything else and fix.
---
3. Prioritize Fixes with ICE – No Spreadsheet Required
You’ve identified the bottleneck. Now what? You have a dozen possible experiments. Which one to run first?
Enter the ICE Framework – a prioritization method that democratizes decision-making:
ICE = (Impact + Confidence + Ease) / 3 Each scored 1–10.
- Impact – How much will this move the needle on the bottlenecked stage?
- Confidence – How sure are we that this will work? Based on past data, benchmarks, or gut?
- Ease – How fast/cheap is it to implement? (1 = requires engineering for two weeks, 10 = change a headline in five minutes)
Rule of thumb:
- ICE ≥ 7 → Run immediately.
- ICE between 4 and 6 → Consider after high-ICE items.
- ICE < 4 → Skip unless ease is extremely high (8+).
Example bottleneck: TOFU (awareness).
- Experiment A: Write 5 SEO blog posts targeting long-tail keywords. Impact=7, Confidence=6, Ease=6 → ICE = 6.3 → Do it, but not urgent.
- Experiment B: Repurpose top 3 organic social posts into LinkedIn carousels. Impact=6, Confidence=8, Ease=9 → ICE = 7.7 → Run this week.
This framework prevents analysis paralysis. You don’t need a data scientist to score these – just your team’s best estimate.
---
4. Close the Loop with Multi-Touch Attribution (The Missing Piece)
Diagnosing bottlenecks is useless if you can’t connect fixes to pipeline movement. Most small teams use last-touch attribution – the worst possible model for understanding why something worked.
Multi-touch attribution (MTTA) gives credit to every touchpoint in a buyer’s journey. With dolv.ai, you get five built-in models (linear, time-decay, U-shaped, etc.) and a cross-metric correlation engine that ties campaigns directly to pipeline velocity – not just volume.
For small teams, even a lightweight U-shaped model (40% first touch, 20% middle, 40% last touch) is a game-changer. It reveals that your TOFU blog post didn’t just generate “awareness” – it contributed to a closed-won deal three months later.
How to use it in your playbook: After running a bottleneck fix (e.g., a LinkedIn ad campaign targeting awareness), check the attribution report after 30 days. Did the campaign correlate with improved TOFU metrics and an eventual lift in MOFU or BOFU conversions? If yes, double down. If not, re-ICE and try something else.
---
Conclusion: You Already Have the Data to Win – You Just Need a System
You don’t need a data team. You need a playbook – a repeatable way to measure, diagnose, and act. The Funnel Health Composite Score gives you a single number to watch. The weighted bottleneck rule tells you exactly where to dig. ICE prioritizes experiments without endless debate. And multi-touch attribution closes the loop so every action informs the next.
Tools like dolv.ai automate this entire cycle: pulling the funnel health score daily, flagging bottlenecks, suggesting ICE-scored experiments, and even executing approved actions with AI agents that work in your CRM, email, and analytics platforms. But even without the automation, you can start today with a Google Sheet and 20 minutes a week.
Your pipeline is leaking. Now you know exactly where. Go fix it.
See dolv run the work
Grounded AI that executes and measures — with you in the loop.