What should I expect in the first 30 days of an outside ad-spend audit?

Week 1: access and scope

The first week is logistics, not analysis. The operator needs read-only access to the ad accounts under audit, read access to the analytics property, and visibility into the conversion path on the destination site. On Google Ads, that means MCC manager-level read access. On Meta, that means a Business Manager partner add at the analyst role. On Microsoft Advertising, an agency link with read permission. Conversion access is the part most founders miss. If the operator cannot see how conversions are configured in Google Ads, the audit will catch the symptoms but not the source.

The KPI alignment call is the other half of week one. The founder names the metric the business runs on at the leadership level: blended ROAS, contribution margin, customer-acquisition cost against payback period, lead quality at the sales-handoff stage. Whatever the dashboard reports is rarely the metric that decides whether a campaign is working. The alignment call is where the operator learns which number the founder will be measured against, which is the number the audit memo has to anchor to.

The business-context briefing closes out the week. What does the brand sell, who is the buyer, what is the AOV, what is the margin band, what does seasonality look like, and what changed in the last ninety days. An hour of context up front saves a day of guessing later.

Week 2: the diagnosis pass

Week two is read-only by design. The operator runs the full diagnostic without changing a single campaign. This is the audit-without-intervention principle, and it matters because every change made during the audit muddies the diagnosis of what was already there. If the operator pauses a campaign on day eight, the data for the rest of the audit is contaminated, and the founder loses the baseline to measure the fix against.

The diagnosis covers structure, match-type discipline, conversion configuration, audience layering, Performance Max guardrails, attribution model fit against the business model, and the cross-check between platform-reported conversions and the source-of-truth in Shopify or the CRM. Every finding gets documented with the report it came from, the dollar figure attached to the leak, and the proposed fix. Nothing gets implemented yet.

On a Shopify baby and kids decor brand I work with, the week-two read surfaced what looked like a healthy Meta account: three months of spend totaled a little over a thousand dollars and produced seven purchases at a blended 2.57x ROAS. Read in isolation, those numbers pass the eye test. Read against the Google PMax campaigns on the same brand, it was clear Meta was capturing a small fraction of the buying signal and the founder had been told to scale Meta for a year. The waste was not inside Meta. The waste was that Meta had budget at all. That finding only surfaces in a diagnosis that compares the channels honestly, which is what week two is for.

Week 3: the prioritized memo

Week three is the deliverable. The memo lands as a written document, not a slide deck. It names the three to five highest-impact findings, ranks them by recoverable dollars per month, and includes the implementation sequence for each fix. The ranking matters more than the count. A founder who tries to fix fifteen things at once fixes none of them. The memo’s job is to name which three to act on first, in what order, and which can wait.

Each finding includes the diagnostic question that surfaced it, the dollar figure of the leak measured against the trailing ninety days, the proposed fix, and the expected lift if implemented. The expected-lift number is directional, not promised. The point is to give the founder enough to prioritize against, not to manufacture certainty the data does not support.

The memo also names what was checked and found clean. A founder who only sees the problems does not know what was tested. Calling out the conversion setup, the negative-keyword hygiene, or the attribution model as already working is part of an honest delivery.

Week 4: intervention or handoff

The fourth week is the founder’s choice. The operator either implements the fixes directly on the account, hands the memo to the in-house team for them to run, or sends it to the existing agency with implementation notes. All three are valid endings to the engagement.

If the founder hands the memo to an existing agency, the memo should travel with enough specificity that the agency cannot argue around it. Each fix names the report, the dollar figure, the proposed change, and the expected outcome. That structure is what keeps the agency conversation from turning into a debate about whether the leak exists.

If the founder runs the fixes in-house, the memo doubles as a training document. The in-house marketer who runs the implementations learns the diagnostic pattern, and the next leak gets caught faster.

If the operator implements directly, the engagement either ends after week four or rolls into a separate scope. Either way, the audit and the implementation are priced separately on a clean engagement, and the founder always sees the memo before any implementation work starts.

What to ask for at kickoff

Three documents make the audit faster. The trailing twelve months of monthly spend by channel, the conversion definitions currently configured in each ad platform, and the KPI the business runs on at the leadership level. Send those in week one and the access setup completes in days instead of a week and a half.

Three questions to ask the operator before access is granted. What read-only role do you need on each platform, what is your protocol for handling the data once the engagement ends, and will you implement changes during the audit window or only after the memo lands. The answers tell you whether the operator runs the engagement on the audit-without-intervention principle or treats the audit window as live management.

Red flags to watch for during the 30 days

Three patterns to watch for in any outside audit. An operator who pauses campaigns or changes bidding during the diagnosis week is contaminating the baseline, and the memo at the end is partly a story about their own changes. An operator who pushes for a retainer sale before the memo lands is pricing the relationship ahead of the diagnosis, which inverts the incentive the founder hired them for. An operator whose findings would all happen to require their ongoing involvement is selling the next engagement, not delivering an audit.

The clean version of the four-week engagement is read-only access in week one, no changes in week two, a written memo in week three, and a founder-owned choice in week four about who implements what. Anything that breaks that shape is worth a question.

The hub at /wasted-ad-spend/ walks through the leak categories the memo usually surfaces. The free 25-page audit at /audit is the asynchronous version of the same diagnostic process, written so a founder can walk it section by section without paying anyone. The services overview and the pricing page cover what the paid version costs, and the contact form is where the conversation starts.