The Closed Loop Score: A 5-Minute Audit That Finds Where Your Business Is Leaking Revenue

Most mid-market businesses leak 30-50% of revenue in the gap between systems they think are working. The leaks are not in any single function. They are in the loops between functions: form-submit-to-call (inbound), call-to-CRM-to-calendar (qualification), brand-to-asset-to-channel (brand), ops-to-dashboard-to-decision (ops), content-to-traffic-to-pipeline (content). The Closed Loop Score finds the biggest leak in 5 minutes.

This is not a kratt theory. Research from McKinsey on operations and from Forrester on customer experience keeps pointing at the same place: mid-market revenue leaks between systems, not inside them. The score makes that gap measurable before you spend a dollar fixing it.

This post documents the framework. We use it as the diagnostic for every discovery call at kratt. Operators who score themselves before the call save 30 minutes and ship 10x better follow-ups. The reason it works is boring: it forces you to measure the seams between teams, which is the one place no software vendor ever sold you a dashboard for.

1. Why score loops, not departments

Traditional business audits score departments: marketing, sales, ops, product. The pattern misses where mid-market revenue actually leaks - in the handoffs between departments. A great marketing team handing leads to a great sales team via a 17-minute callback gap loses just as much as a mediocre marketing team would. Each department reports green on its own dashboard. The number that is red lives in the white space between two dashboards, which means no one is paid to watch it.

This is why kratt scores loops, not boxes. Harvard Business Review's response-time research showed conversion drops 10x between minute 1 and minute 5 of a lead's life. That gap lives in no department's KPI dashboard. It lives in the loop. Bain's revenue marketing stack research reaches the same conclusion.

The economics of the seam are worse than most operators model. A lead that costs you 80 dollars to generate is not worth 80 dollars when it arrives - it is worth 80 dollars multiplied by the odds you convert it. Halve the contact rate and you have doubled your effective cost per acquisition without changing a line of ad copy. That is why a 2-point lift in one loop frequently beats a 20% lift in ad spend: the loop fix is free yield on traffic you already paid for. The same arithmetic shows up in onboarding and renewal seams, which is why Gartner's customer lifecycle marketing research keeps landing on lifecycle handoffs rather than channel performance as the real lever.

2. The 5 loops

The 5 loops cover roughly 90% of where mid-market businesses leak revenue. The framework draws on patterns documented in Gartner's customer lifecycle marketing research, but operationalised for the mid-market revenue band where one founder still oversees every loop:

1. Inbound Loop. The path from a stranger landing on your site to a real conversation. Owned by the voice agent + form layer.
2. Qualification Loop. The path from real conversation to calendar-booked next step. Owned by the script, CRM, and calendar handoff.
3. Brand Loop. The path from brand DNA to shipped assets across channels. Owned by the ad factory and brand system.
4. Ops Loop. The path from weekly operations to founder dashboard and decisions. Owned by closed-loop automation.
5. Content Loop. The path from one piece of content to compounded organic + repurposed reach. Owned by the SEO and content engine.

Each loop shares one shape: a start state, a handoff, and an end state that should trigger automatically. A loop is closed when the end state of one step fires the start of the next without a human remembering to do it. Most mid-market revenue leaks at the exact point where a human is supposed to remember something - return the missed call, log the qualifier, repost the asset, check the report. The score measures how many of those memory-dependent handoffs you have removed.

3. The scoring rubric

Each loop scores 0-5. Use the highest score that is honestly true for you. The word "honestly" carries the whole framework. If you score the aspiration instead of the reality, the diagnostic points you at the wrong loop and you spend 21 days fixing something that was never the constraint.

Score	Inbound	Qualification	Brand	Ops	Content
5	Under 90s to real conversation	Auto-routed with 2+ qualifiers captured	8 brand rules + 6 primitives + hundreds of creatives/mo	Closed-loop ops + founder dashboard	1+ post/week repurposed across 3 channels
4	Under 5 min	Auto-routed with 1 qualifier	Brand guidelines + 20 assets/mo	Most ops automated, weekly reports	Bi-weekly post + 1 channel
3	Under 30 min	Manual routing same-day	Brand guidelines doc + designer	Some Zaps + reports	Monthly post
2	Under 4 hours	Manual routing within 24h	Inconsistent brand, ad-hoc designer	Mostly manual + spreadsheets	Quarterly post
1	Same day	Lead-list dump + sporadic follow-up	No brand system; founder choices	All manual	Sporadic / random
0	Next day or worse	No qualification at all	Nothing recognisable	Spreadsheets only	None

Two scoring rules keep operators honest. Score the median case, not your best day: if a hot lead gets a 5-minute callback at 10am Tuesday but a Friday-night form sits until Monday, your inbound score is the Monday number. And a loop only earns a 4 or 5 if it runs without you. If the callback is fast only because you personally watch the inbox, that is a 2 or 3 in a costume. The score measures the system, not your willpower.

4. A worked example: 17/25 and where the money goes

Take a 6M-revenue home-services firm running paid search. Honest scores: Inbound 2 (calls returned within about 4 hours), Qualification 3 (coordinator routes same-day by hand), Brand 4 (clean guidelines, roughly 20 assets a month), Ops 3, Content 5. Total: 17/25 - "leaking" by the bands below. But the total hides the story. The minimum is the constraint, and the minimum is Inbound at 2.

Now put numbers on the leak. Say they spend 40,000 dollars a month on search at roughly 90 dollars per lead, so about 444 leads land monthly. With a 4-hour median callback, contact rate on inbound forms often sits near 35-40% before the lead has called three competitors. Drag that callback under 90 seconds - the difference between a 2 and a 5 - and contact rate on the same leads commonly moves into the 55-65% range. Same leads, same spend, more conversations reaching the pipeline. At a 4,000-dollar average job, recovering even 30 missed conversations a month dwarfs anything a 20% budget bump buys.

This is what the total buries and the lowest-loop rule surfaces. The firm's instinct is to celebrate the Content 5 and the Brand 4, because those are visible and flattering. The money sits in the Inbound 2, invisible because nobody sees the lead that quietly went to a competitor. Fix the 2 first.

5. How to interpret your total

Sum the 5 scores out of 25:

• 21-25 - Healthy. Polish the lowest-scoring loop. You are in the top 5% of mid-market operators on system discipline.
• 16-20 - Leaking. Fix the lowest 2 loops. The leak is meaningful but not catastrophic. 60-90 day window to recover.
• 11-15 - Hemorrhaging. The 21-day closed-loop deployment fixes 3+ loops at once. Stop investing in any individual function until the loops are wired.
• 0-10 - 2022 ops in a 2026 market. The fix is not 5 small projects; it is one closed-loop sprint with a single goal: lift every loop above 3 in 90 days.

The bands are a triage tool, not a grade. A total tells you how many loops need work and how urgent the timeline is. It never tells you which loop to touch first - that is always the minimum. The total sets the pace; the lowest loop sets the target.

6. How to pick the fix priority

The lowest-scoring loop wins, regardless of total. A firm scoring 22/25 with a 2/5 inbound loop should fix inbound first - the leak is concentrated. A firm scoring 14/25 with three loops at 3 should not "fix everything"; it should pick the loop with the highest revenue yield and fix that one in 21 days.

Revenue-yield rule of thumb:

• If you spend money on ads or SEO, Inbound and Qualification loops have the highest yield. Fix those first.
• If you sell to ecom or DTC and brand recognition matters, Brand and Content loops have the highest yield.
• If you are services with consistent inbound but operational drag, Ops loop has the highest yield. The 25 hours a week reclaimed convert directly to closed deals.

When two loops tie for lowest, the tiebreaker is perishability. A lost lead is gone for good; a delayed content post can still ship next week. So an inbound or qualification leak beats a content or ops leak at the same score, because the inventory it wastes never comes back.

Pair this with The 25-Hour Week for the ops loop, The 90-Second Inbound Loop for the inbound loop, and The Brand DNA Sprint for the brand loop.

7. The 4-step decision framework

Scoring is the easy part. Turning a score into a 21-day plan is where most operators stall. Run these four steps in order and the plan writes itself.

1. Find the minimum. Circle the lowest loop. On a tie, apply the perishability tiebreaker above. This is your one target.
2. Price the gap. Estimate the monthly dollars escaping that loop. For inbound and qualification, multiply leads by the contact-rate gap by close rate by average value. For ops, count the hours and ask what they convert to. A leak you cannot size is a leak you will misprioritise.
3. Pick the closure mechanism. Decide what removes the memory-dependent handoff. Inbound means a voice agent or instant-routing layer; qualification means CRM rules plus a calendar trigger; content means a repurposing pipeline. The mechanism is what makes the loop run without you.
4. Set the 21-day exit test. Define one number that proves the loop closed - median callback under 90 seconds, 100% of qualified leads auto-booked, one post on three channels weekly. No exit test, no start.

The discipline is one loop per 21 days. The 90-day arc lifts three loops cleanly because each fix has a clean start, a clean exit test, and no contention for attention with the next.

8. Who this is NOT for

The framework is calibrated narrowly, and using it outside that band wastes your time. It is not for businesses under 2M in revenue, where founder attention still closes the loops by hand and the right move is to sell more, not systematise. It is not for enterprises above roughly 30M, where each loop already has a dedicated team and budget; at that scale the seams are inside functions, not between them, and you need a deeper instrument.

It is also not a substitute for unit economics. If your average customer is unprofitable, closing every loop perfectly only helps you lose money faster. Fix the math first, then close the loops. And it is not for operators hunting a single silver-bullet tactic: the score refuses to tell you "just run more ads," because its premise is that your constraint is a seam you have not been measuring.

9. Common misconceptions

"I can fix all 5 loops at once."

You cannot. Trying to lift 5 loops simultaneously fragments attention and moves none of them. Pick the lowest, deploy in 21 days, then move to the next. The 90-day arc lifts 3 loops cleanly. The 6-month arc lifts all 5.

"My total score is the only thing that matters."

It is not. A 21/25 with one loop at 1 is worse than a 19/25 with all loops at 3-4. The minimum is the constraint. Theory of constraints applies here too: throughput is set by the slowest step, so improving anything but the constraint adds nothing at the system level.

"My business is too small to need this."

The framework is designed for $2-30M revenue mid-market. Below $2M, simpler diagnostics apply. Above $30M, you have specialised teams running each loop and need a different framework. The score is calibrated for the band where one founder still needs to see all 5 loops on a single page.

"A high score means I am done."

A high score means the loops are closed today. Loops reopen. A rep leaves, a tool changes its API, a new channel arrives without routing rules. Re-score quarterly. The operators who stay healthy treat the audit as a recurring check, not a one-time certificate.

What to ship this week

Run the Closed Loop Score in the next 5 minutes. Score every loop honestly - inflated scores cost you the diagnostic. The output gives you a single fix priority and the 21-day deploy path.

Then book a 30-minute system review if the lowest-scoring loop is below 3. The 30 minutes is enough to scope a 7-21 day fix and answer whether it is the kind of thing one of the founders should ship for you, or something your existing team can run with the framework.

Mid-market businesses do not lose because they pick the wrong CRM or hire the wrong VP. They lose because the loops between functions leak revenue at every handoff and nobody owns the loop. Score the loops. Fix the lowest. Compound the rest. The 5-minute audit is the cheapest diagnostic in business; the 21-day fix is the cheapest intervention. Both happen this month or they do not happen.

Related resources

• Run the Closed Loop Score - the interactive 5-minute version with prioritised fix list.
• The 90-Second Inbound Loop - the inbound-loop fix.
• The Outbound Voice Agent Script - extends inbound to outbound qualification.
• The Brand DNA Sprint - the brand-loop fix.
• The 25-Hour Week - the ops-loop fix.
• The 7-Day Vibe-Coded Website - the foundation underneath the content and brand loops.
• Book a 30-minute review - if your lowest loop is below 3, scope a fix sprint here.

Last updated: 2 May 2026 · Updated as the framework evolves with new audit data.

Frequently asked questions