The 30-Day Message Consistency Audit

Message drift is easiest to catch when you stop looking at the surfaces individually and start reading them side by side. A thirty-day audit is the shortest window that produces comparison data credible enough to act on. Less than that, you're reading snapshots; more than that, the market has moved and the audit is dated.

The structure below is the one we run internally at Stratridge and with clients. Four weeks, seven surfaces, same scoring rubric each time. Ship at the end of week four.

The seven surfaces

Pick these, not others. The temptation is to audit everything — every landing page, every sequence, every video. The seven below are the minimum viable set that surfaces real drift without producing a review artifact nobody can read.

Seven surfaces, one day of pulling, no more. If it takes longer than a day, someone is editing while they pull; stop and come back with the unedited versions.

The four-week sequence

The cadence enforces the discipline. Pulling, reading, scoring, and synthesizing on the same day collapses the thinking. Letting each week breathe produces observations the pull-and-score version misses.

The scoring rubric (same questions, every surface)

Five questions per surface. Score 1–5 on each. Evidence required.

• Does this surface use the same category noun as the positioning brief? (1 = different noun / missing noun. 5 = same noun, prominent placement.)
• Does this surface describe the same ICP as the positioning brief? (1 = different persona / no persona. 5 = identical audience language.)
• Does this surface quote or reflect the unique-value claim from the brief? (1 = missing or contradicted. 5 = same claim, same framing.)
• Does this surface use any banned words from the language guide? (1 = three or more. 5 = zero.)
• Does this surface contradict any other surface in the audit? (1 = direct contradiction. 5 = fully consistent with the rest of the stack.)

Twenty-five possible points per surface. Seven surfaces. 175-point ceiling. In our client work, first-audit scores cluster around 95–115. A score above 140 is rare and usually means the auditor went easy; verify the evidence.

What the score distribution usually looks like

In our audits across 30+ teams, the typical distribution across surfaces (scoring from 5 to 25) looks like this:

The pattern is consistent enough to predict. Surfaces most visible to leadership score highest; surfaces most distant from the PMM's editorial view score lowest. The help center and the sales deck are the reliable drift reservoirs. Most teams underinvest in auditing both because they sit outside the PMM's org chart.

The triangulation interviews

Week 3 is where the audit stops being a document review and starts being a data-gathering exercise. Three twenty-minute calls:

• The AE — "Walk me through how you described us in the last big deal. What did the prospect push back on? What did you end up saying that's not in the sales deck?" The improvisation is the signal.
• The CSM — "What do customers describe us as, three months in? What do they tell their peers when they refer us?" If the customer's description drifts from the homepage, your retention narrative has a leak.
• The recent customer — "When you were evaluating, how would you have described us to someone on your team? Has that description changed now that you've used the product?" The delta between the evaluation description and the current one is where messaging is most actionable.

Write each interview down. Quote the exact sentences. Paste them into the scorecard as evidence for question 5 (contradictions across surfaces). Usually at least one interview finding reshapes the ranked recommendations.

The output

One document, five pages maximum, shipped at the end of week four:

• Page 1 — Surface-by-surface score grid. Seven rows, five columns, totals. One sentence per cell of evidence.
• Page 2 — The three strongest drift patterns, named and quoted. "Sales deck describes us as a [category A] tool while every other surface uses [category B]." Direct quotes, surface attributions.
• Page 3 — Triangulation notes. Three interviewees, three quotes each, delta from the written positioning called out.
• Page 4 — Three recommendations. Surface, owner, shipping date. No more than three; more than three ships zero.
• Page 5 — The next-audit date. Ninety days out. Calendar invite attached.

Five pages is the point. The audit gets acted on because it's short; longer versions get deferred. Leadership reads page 1 and page 4 and nothing else — so both need to stand alone.

Why thirty days, not ninety

Quarterly audits feel disciplined and are actually too slow. Messaging drift compounds inside a quarter — the sales deck edited in month one produces a shift the homepage copywriter doesn't hear about until month three. A thirty-day audit catches the drift while the trail is still warm; a ninety-day audit is archaeology.

The real move, for teams that can afford it, is running the audit continuously rather than episodically — Stratridge's Message Consistency capability monitors the seven surfaces in the background and surfaces the score changes week-over-week. The manual version in this article is the starting point, and the right one for most teams before they automate.