run your own marketing audit.

every agency offers a free audit, and most of them are lead magnets — a crawl report with the logo swapped. the useful version is something you can run yourself, and the order matters more than the tooling. get it wrong and you spend a month improving content that nobody could convert anyway.

— 01why order beats tooling.

an audit is a sequence of eliminations. if your tracking is broken, every number downstream is fiction, so it goes first. if your landing page converts at a quarter of what it should, more traffic is the expensive way to fix a cheap problem. work from the foundation up and you will usually find the answer in step two or three.

there is a psychological reason for the order too. content is the most visible part of marketing and the most enjoyable to have opinions about, so audits that start there turn into critiques of individual posts. tracking and conversion are dull, unarguable, and where the money actually leaks. do the dull parts while you still have energy for them.

— 02the seven steps.

block half a day for the first three and an hour each for the rest. write one sentence of conclusion per step — if you cannot, you have not finished the step.

work on a real time window rather than all-time data. ninety days is long enough to survive one bad week and short enough to reflect how the business runs now. and pick the window before you look, not after — choosing dates once you have seen the numbers is how audits turn into justifications.

the step people skip is the third one, walking your own funnel on a phone as a stranger would. it takes twenty minutes and it is where the cheapest wins hide: a form that asks for a company name before an email, a whatsapp button that opens a dead number, a page that takes six seconds on 4g. no amount of content strategy compensates for any of those.

bring one outsider in for that step if you can — someone who has never seen your site. watch them try to enquire without helping them. every hesitation is a finding, and you will learn more in ten minutes of watching than in a month of heatmaps.

— 03what a good finding looks like.

"our instagram engagement is low" is not a finding. "our reels hold 28% of viewers to three seconds, and the three that held 60% all opened with a question" is a finding, because it tells you what to do next.

the same applies to paid. "meta is expensive" is a complaint. "our cost per enquiry on brand terms is a fifth of our cost on broad interest audiences, and 40% of spend is on broad" is a decision waiting to be made.

the same discipline applies to competitors. "they post more than us" is not a finding either. "they publish a pricing page and we do not, and three of our top search queries are cost questions" is — it names something you can ship this month. the test for every line in your audit is whether someone could act on it without asking you a follow-up question.

— 04the decision at the end.

an audit that ends in a list is a document. an audit that ends in a choice is useful. force yourself into one of three conclusions: fix the foundation before spending more, keep spending and improve the creative, or stop a channel entirely and move the budget.

if you get to the end and cannot pick, the missing piece is almost always measurement — which means step one was not done properly. that is not a failure; it is the finding.

and resist the urge to act on all seven steps at once. pick the single change with the largest expected effect, ship it, and give it a month before you touch anything else. audits that trigger fifteen simultaneous changes make the next audit impossible to read, because nothing can be attributed to anything.

one caveat on doing this yourself: you will be generous with the parts you built. founders defend their own landing page copy and marketing managers defend their own calendar, which is why the conversion step and the content step are where self-audits go soft. if you can only bring in outside eyes for one thing, spend it there — not on the technical crawl, which any tool will do honestly because it has no feelings about your work.

write the whole thing up in one page and date it. the value of an audit compounds when you can put this quarter's next to last quarter's and see which conclusions you acted on. a folder of undated pdfs is how good findings get rediscovered every year and fixed none of the times. if the write-up ends in three conclusions you cannot resource, that is the point to bring in a digital marketing strategy agency rather than another tool.

related on the dispatch: audit the page before the ads, auditing your brand voice, how to do a social media audit.