Audit sounds like something done to a company rather than for one. The exercise underneath is simple: work out where revenue leaks between a lead arriving and cash landing, then put the leaks in order of what they cost.
The seven areas
These cover the full path a deal travels. Most companies leak in two or three of them, and almost nobody leaks in all seven.
Demand and routing
Lead arrives to owner assigned
- Median and 90th-percentile time from arrival to first human contact
- Share of inbound that never receives a response at all
- What happens to leads arriving overnight and at weekends
- Whether routing rules match how territories are actually run
Qualification
Owner assigned to opportunity created
- Whether a written ICP exists and whether reps apply it consistently
- Conversion from lead to qualified opportunity, by source
- How much time goes into accounts that were never going to buy
Pipeline definitions
The one that causes the most downstream damage
- Do stages have written exit criteria, and are they observable facts
- Agreement rate when two reps stage the same five deals independently
- Share of open deals that genuinely meet their current stage criteria
- Median time in stage, and how many deals sit well beyond it
Handoffs
Marketing to sales, sales to onboarding
- Whether handoffs are structural or conversational
- Information collected twice because it did not travel
- Time between closed-won and onboarding actually starting
Data quality
The quiet tax on everything above
- Duplicate rate across accounts and contacts
- Required fields that are empty on live records
- Number of places the same record can be created
- How much of the database has gone untouched for eighteen months
Reporting and forecasting
Whether anyone believes the number
- Whether the weekly number requires an export to produce
- Variance between committed forecast and actual, last four quarters
- Whether the CRM and the board deck agree without manual reconciliation
Stack and spend
What you pay for and what gets used
- Tools in the go-to-market stack, and seats actually active in each
- Overlapping functionality across subscriptions
- Integrations that have silently stopped running
Attaching a number to a leak
A finding without a number rarely gets fixed, because it has to compete for attention with things that do have numbers. Most leaks can be costed roughly, and rough is enough.
Slow lead response is the easiest. Take monthly inbound volume, your lead-to-opportunity conversion rate, and average deal size. The Harvard Business Review audit of 2,241 companies found firms responding within an hour were nearly seven times more likely to qualify a lead, so even a conservative uplift assumption on a portion of your volume produces a defensible figure.
Stale pipeline is the second. Total the deals failing a staleness test, then apply your true win rate. That gives you the forecast error you are currently carrying.
Duplicate data costs time rather than deals, so cost it in hours: duplicates multiplied by minutes spent reconciling, multiplied by loaded hourly cost.
Show the working in every case. A founder who can see the arithmetic can argue with it, and a number that survives an argument is one that gets acted on.
The afternoon version
You can find most of the big leaks yourself with four exercises.
- The stage test. Take five open deals. Ask two reps independently which stage each belongs in and why. Disagreement above 20% means your pipeline data is not comparable across people.
- The response test. Submit your own web form on a Friday evening from a personal address. Time the reply. This one number surprises more founders than any other.
- The shadow test. Ask your two best reps what they keep outside the CRM and why. Their answer maps precisely where the system stopped helping, a pattern covered in why your CRM became admin work.
- The staleness test. Count open deals whose close date has moved more than twice. Total their value. That figure is the gap between your pipeline and your pipeline.
The part that is harder to do in-house is calibration. You have only ever seen your own pipeline, so it is difficult to know whether a 34% disagreement rate on staging is terrible or typical. That comparison is most of what an outside read adds.
What the output should look like
Short. Three or four pages covering the leaks found with their locations, a cost estimate where the arithmetic allows one, a sequenced fix list, and an honest note on what is fine already.
That last section matters more than it sounds. An audit that finds problems everywhere is usually describing the person who wrote it rather than the company. Naming what works tells a founder the assessment was real.
Our version of this is the GTM teardown: forty-five minutes on your live pipeline, a written document three days later, and no obligation attached to any of it.
Source: Harvard Business Review, The Short Life of Online Sales Leads, an audit of 2,241 US companies.
Frequently asked questions
How long does a RevOps audit take?
What access does a RevOps audit need?
Can we do a RevOps audit ourselves?
What should a RevOps audit produce?
Zachary works with founder-led B2B SaaS companies between seed and Series B, as a fractional VP of Sales and Marketing and through the team that builds their revenue operation on Throughline.
Get a free GTM teardown