The first ninety days decide the engagement

The first thing any competent agency asks for is access: analytics, ad accounts, the site, the CRM, the data warehouse if there is one. The speed at which that arrives is the single most predictive early signal we have, and it has almost nothing to do with IT.

When access takes six weeks it is rarely because the request was difficult. It is because nobody internally has enough authority or motivation to unblock it, which means the engagement has no real sponsor. That same absence will later show up as work sitting unapproved, decisions not being made, and eventually a renewal conversation nobody is advocating for.

If you are the client, the useful move is to treat access as a pre-kickoff deliverable owned by a named person with a date, not as an administrative task that runs in parallel with the work. If it is genuinely hard at your organisation, say so upfront — that is manageable. Discovering it in week four is not.

The most common cause of a failed first quarter is not underperformance. It is that the two sides were measuring different things and neither wrote it down.

Before any work begins there should be a short document naming what the first ninety days are expected to produce, and — more importantly — what will not have moved yet. Organic results will not appear in a quarter. A rebuilt measurement stack shows up as better decisions long before it shows up as revenue. Paid restructuring often looks worse before it looks better, because learning has been reset deliberately.

Writing down the things that will not improve is the uncomfortable half, and it is what protects the relationship in the month-three review. Without it, any honest explanation of why a number is flat sounds like an excuse invented after the fact.

Engagements move at the speed of their slowest approval. What determines that is not how senior the people involved are; it is whether one named person can say yes without convening anyone.

Committees are not the problem in themselves — plenty of good work goes through review. The problem is a committee with no owner, where feedback arrives from four directions with equal weight and no mechanism for resolving contradictions. The agency then either averages the feedback into something nobody wanted, or picks a side and creates a political problem.

The fix is dull and effective: name the person whose call is final, and route consolidated feedback through them. It costs one conversation at kickoff and it removes the most common source of drift over the following year.

A good first month looks slower than clients expect and is the highest-leverage period of the entire engagement. It is spent finding out what is actually true — whether tracking is trustworthy, where the money is really going, which assumptions in the brief survive contact with the data.

An agency that arrives with deliverables in week one is either recycling something or guessing. Both are worse than they look, because the resulting plan will be built on the brief rather than on the account, and briefs are frequently wrong about where the problem is. We have started engagements aimed at paid efficiency and found the real constraint in the conversion path more than once.

What you should expect at the end of that month is a findings document that tells you at least one thing you did not know and, ideally, at least one thing you did not want to hear. If the first month produces no surprises, either your account is in unusually good shape or nobody looked very hard.

Some things are recoverable and some are not, and the difference is usually whether the problem is capability or attention.