Message Consistency Audit: A Self-Serve Assessment

A full message-consistency audit across fifty content surfaces can take a week for a PMM working alone. The self-serve version below is the 90-minute triage: enough to find the worst drift, decide whether the full audit is worth running, and produce a defensible one-page report for the CMO. It is not a substitute for the full audit — it is the filter that decides whether you need one.

What you audit, and in what order

The audit covers five surfaces in priority order. Drift on surfaces higher in the list is more damaging than drift on surfaces lower, because higher-priority surfaces are seen by more buyers, more often, in more decision-shaping moments.

For each surface, you answer the same four questions. Twenty questions total, scored from 0 to 2 each, for a possible 40 points.

The four questions for each surface

These four questions catch roughly 85% of the drift patterns that matter. Use them verbatim; the wording is chosen to produce a binary answer that resists PMM optimism.

Question 1 · Is the category noun identical to the one in the positioning brief?

Score 2 if the exact phrase in the brief appears. Score 1 if a close variant appears ("sales enablement" vs. "sales readiness"). Score 0 if a different noun appears. This is the test for Layer 1 drift. A single varied category noun across two surfaces is the most common form of message drift, and it is the most expensive one — it tells buyers you don't know what you are.

Question 2 · Is the ICP described in the same sentence shape as the brief?

Score 2 if the audience sentence matches the brief's ICP sentence in meaning, even if the wording is different. Score 1 if it matches most of the dimensions but misses one (e.g., mentions company size but not role). Score 0 if it describes a different ICP — broader, narrower, or simply different. This is the Layer 2 drift test. ICP drift is how a product that sells to VPs on the homepage gets demoed to directors by the sales team.

Question 3 · Does the top differentiator match the one in the brief?

Score 2 if the brief's top differentiator is the first differentiator mentioned on the surface. Score 1 if it's mentioned but not first. Score 0 if a different differentiator leads. This is Layer 5 drift — usually a content author picked their personal favorite differentiator over the one the brief anchored.

Question 4 · Does the surface name the alternative the buyer compares against?

Score 2 if the surface explicitly addresses the comparison (directly or by framing). Score 1 if it implies the comparison but doesn't name it. Score 0 if the surface ignores the comparison entirely. This is Layer 4 drift, and it is the layer most content authors skip.

Scoring and interpretation

Total the twenty answers. The breakpoints below are calibrated against fifty audit samples from our own review work; they're not arbitrary.

32–40: Healthy. The surfaces are aligned; any drift is cosmetic. Continue with a quarterly check. No full audit required.

24–31: Drift forming. A layer is slipping somewhere. Look at which question has the lowest average score across the five surfaces — that's your weakest layer, and the one to address in the next content review. Full audit not yet needed.

16–23: Active drift. Multiple layers are out of sync across multiple surfaces. A full audit is due, with a PMM-owned remediation plan. Do not pretend the CMO doesn't need to see this.

0–15: The brief is not operating. The positioning doc exists but isn't shaping content. This is not a message-consistency problem — it's a strategic-context problem. The audit output should route to the CMO and the person who owns the brief, not to the content team.

The four drift patterns

Beyond the score, the shape of where you lost points matters more than the total. Four patterns recur.

What your answer distribution tells you

The 90-minute protocol

A PMM running this audit alone should allocate the time as follows.

Minute 0–10. Read the positioning brief. Write the four answers (category noun, ICP sentence, top differentiator, named alternative) on a scratch sheet. This is your reference.

Minute 10–30. Homepage and pricing page. Score each surface against the four questions. Eight data points.

Minute 30–50. Sales deck and top three blog landings. Score each. Sixteen data points.

Minute 50–65. Onboarding email 1. Score it. Twenty data points total.

Minute 65–80. Total the score. Identify the lowest-scoring question across surfaces. That's your weakest layer.

Minute 80–90. Write the one-page report. Three sections: current score, weakest layer with three example quotes from the surfaces, recommended next step (full audit, content review, or upstream brief rework).

The hand-off point

This audit works at fifty surfaces. It does not work at five hundred. A company with five hundred content surfaces cannot be audited by a single PMM in ninety minutes, and the rules-of-thumb above lose accuracy when the sample is that large. At scale, the audit becomes a systematic review — automated drift detection, categorized findings, and a remediation queue.

The PMM who runs this manual audit quarterly and sees the score move in a consistent direction — either up or down — has useful signal. The PMM who tries to run it on a surface count where every quarter's sample is different has noise. Manual audits scale to roughly fifty surfaces; past that, the work changes kind, not just magnitude. The honest handoff is to tooling designed for the larger sample, and to reserve the manual audit for the highest-priority surfaces where human judgment on context still matters.