Four weeks to hire
an agency, not
four months.

Agency searches rarely go wrong on the choice. They go wrong on the calendar. A brand decides in March that paid social needs outside help, takes discovery calls through April, loses May to a packaging launch, revives the shortlist in June with two of the four agencies gone, and signs in July for an August start.

Five months on a decision that contained about four weeks of work. The fees were never the expensive part. The expensive part was two quarters of media running the way it always had while everyone waited for a process nobody owned.

I run this from the brand's side. I do not sell creative, media buying or development, so my interest is in the search ending quickly with the right partner rather than in being one of the options. A three-person team I worked with went from $8M to $42M in annual revenue over 28 months, and part of that was replacing creative and paid partners without pausing growth while we did it. A hardware brand went from $6M to $28M in GMV in 18 months, which meant upgrading development, creative and paid partners more or less in parallel. Both of those ran on a calendar close to the one below.

Searches sprawl for
four fairly boring
reasons.

The first is that nobody owns it. A founder mentions it, a marketing lead takes two calls, an operations lead takes a third, and nothing moves because the search sits in the gap between three people's real jobs. The second is that no decision date exists, so every week is a reasonable week to keep looking.

The third is discovery calls in place of a brief. Each one resets the conversation, because you explain the business from scratch, the agency asks for numbers you have not prepared, and you leave with an action item instead of a response. Ten of those calls is two weeks of your working life and produces nothing you can score.

The fourth is scope drift. A search for a paid social partner becomes a conversation about the funnel, which becomes a conversation about the site, which becomes a rebrand, which becomes a project nobody has budget for. Deciding what the agency owns before you start is what keeps that from happening, and it is two hours of work.

Before the clock
starts, settle what
you are buying.

There are only three things an agency can sell you. Hands for work you can already specify, a team for execution at a volume your own people cannot cover, and judgment on the decision you keep postponing. Agencies do the first two well and the third badly, which is where most of the disappointment in this category comes from. The four kinds of help sorts that out, and the fractional advisor comparison covers the cost side. If the honest answer is that you need someone accountable inside the business, read the first operator hire before you brief a single agency, and whether outside advice is worth it at your size if you are not sure the spend is justified yet.

Week zero produces four lines: the metric this engagement is judged on, the person on your side who owns the relationship, the budget ceiling including media, and the date you will decide. If you cannot write the sentence describing what the agency owns, you are not ready to start, and starting anyway is how four weeks becomes four months.

Four weeks, four
outputs, one
decision date.

Here is the whole thing. Roughly seventeen hours of your time spread across a month, which is less than most brands spend on discovery calls alone.

Fix all four dates before you send anything, and put the decision date in the brief. It changes how agencies treat you. A real date signals a real budget, which moves you up their queue and gets you a senior person on the response instead of whoever writes the proposals.

The brief itself, the eleven questions and the scorecard the week 2 round depends on are all in the agency RFP for DTC brands. This post is the calendar around it.

One owner, one
decider, one
skeptic.

The owner runs the calendar, sends the brief, chases responses and books the sessions. The decider signs and is in the room for week 3. The skeptic is whoever will have to live with the output, usually the person closest to the numbers, and their job is to score independently and argue.

Three people can decide this in four weeks. Six people cannot decide it in twelve, because every additional voice adds a scheduling constraint and a veto without adding much judgment. Gather input widely and hand out votes narrowly. If your CFO needs to approve the spend, they are the decider and they attend week 3, or they approve the budget ceiling in week zero and stay out of the rest.

Tell the three of them the dates up front. Most searches that slip do so because the decider was traveling during the only week the finalists were available, which is a calendar problem masquerading as a strategy problem.

Week three is where
the decision
actually happens.

Both finalists get the same anonymized data and the same live decision, ninety minutes each, named leads only, no deck. You watch three things. Who they brought, and whether the person who will run the account said anything. Whether they push back on the brief, since the ones who agree with everything have not read it. And what they do when a number contradicts the plan they walked in with.

Run the reference calls the same week rather than after, so you are not adding a fifth week to chase voicemails. Ask each agency for a client at your size who left in the last year, alongside the three they want you to call. The ones worth hiring hand that over without much drama.

By Friday of week 3 you should have a ranked pair and a clear reason for the ranking that survives being said out loud to the skeptic. If you do not, the working sessions were run as presentations and the problem is on your side of the table.

The search ends at
the first weekly
review.

Week 4 is paperwork with teeth. Scope the pilot at 60 to 90 days with two dated outputs. Keep notice at thirty days once the pilot converts. Name the leads in the agreement and say what happens if they roll off. Settle access and ownership now rather than at the end, because the ad accounts, the pixel, the creative files and the raw footage are all much harder to recover during a breakup.

Book the first weekly review before anybody signs, and put the 30-day checkpoint on the calendar in the same sitting. The handful of engagements I have watched go badly from a good search all failed in the same place: the brand treated signature as the finish line and gave the agency four weeks of silence to invent a plan. The first ninety days of an outside engagement applies almost line for line to an agency, and the audit, sprint or retainer question is worth reading before you pick the shape of the pilot.

Call the agencies who did not win within two days, and tell them why in one specific sentence. It costs you ten minutes and it keeps the backup warm, which matters more than people think. Pilots do not always convert, and restarting from a second choice who still remembers your numbers takes a week instead of a month.

When a quarter is
already gone, run it
in two weeks.

Cut the longlist to three names you got from operators you trust, skip the qualifying calls, give four working days for written responses, and run both working sessions on the same day with references called that afternoon. Sign a 60-day pilot instead of a contract. It is tighter than I would like and it works.

Three things survive any compression. The written diagnosis question, because it is the only part of the process that predicts anything. The named team with hours attached. And one call to a client who left. Drop those and you are choosing on chemistry, which is how brands end up rehiring the same problem with a different logo on it.

Two more things worth having on hand before you start, whichever version you run: current numbers by channel so the brief takes twenty minutes instead of a day, and a view on what you would fix first. Channel benchmarks give you the first, and the order of operations I audit in gives you the second.

Questions from
operators mid-
search.