onboarding an agency properly.

agencies are judged on month three and largely determined by week one. the pattern is consistent: accounts that move quickly had access, a decision-maker and an honest briefing immediately; accounts that stall spent the first month waiting for logins and approvals.

none of this is difficult. it is a short list of decisions that nobody makes deliberately because the contract felt like the finish line rather than the start.

it is worth being blunt about the asymmetry: an agency can be excellent and still deliver a poor first quarter if access arrives late and approvals take a week. that outcome gets recorded as the agency failing, and the same pattern then repeats with the next one.

— 01access on day one, not week three.

everything, at once: ad accounts, analytics, search console, tag manager, page and channel admin, the website CMS, brand files, and any existing playbook or research. every day of delay is removed from the first month and never recovered, because the audit that should happen in week one is what the rest of the plan depends on.

give ownership-level access where possible rather than restricted roles. an agency that cannot see historical data or install tracking spends week two asking for permissions instead of finding the problem.

and hand over what has already been tried. failed campaigns, formats your audience disliked, the offer that did not work — this is the single most valuable thing you can supply and it is almost always omitted. without it, the agency rediscovers your history at your expense.

a practical tip: create a single shared folder with everything in it before the kickoff rather than emailing items as they are requested. it takes an hour and it removes a fortnight of back-and-forth.

— 02one decision-maker, named.

the fastest-moving accounts have one person who can approve work without convening anyone. the slowest have a committee, or worse, an unnamed founder who reviews things sporadically and overturns decisions after publication.

name that person in writing and give them authority over the calendar. everyone else advises. this is uncomfortable in businesses where several people have opinions about marketing, and it is the difference between a month of output and a month of revisions.

agree the approval window too — 48 hours is workable, a week is not. a calendar built on weekly approval cycles produces content that is stale before it publishes and a team that stops proposing anything time-sensitive.

decide what does not need approval at all. captions within an agreed voice, story replies, community responses inside a documented policy — pre-approving categories rather than items is what makes a fast cadence possible without anyone losing control of the brand.

— 03the workshop, with the right people in it.

the kickoff should include whoever talks to customers daily — sales, front of house, the person answering dms — not only the marketing owner. they know the objections, the questions, the reasons people choose you and the reasons they walk away, and none of that appears in a brand deck.

what the agency needs from the session: the three objections that kill deals, the words customers actually use, the profile of your best customer versus your most common one, and the honest reason people choose you over the alternative. an hour of that is worth more than a month of desk research.

include the awkward material. the complaint you get repeatedly, the part of the service that is weaker than you would like, the constraint you cannot change. good strategy accounts for those; strategy written without them produces promises the business cannot keep.

record the session or take proper notes, and share them back. it becomes the reference document for every new person who joins the account on either side, and it prevents the slow drift where the plan detaches from what the business actually does.

— 04what to agree before the work starts.

four things in writing: the monthly output floor, the reporting standard and date, the definition of a qualified enquiry, and who owns the accounts and assets. these are cheap to agree now and expensive to argue about later, and any agency that resists putting them down has told you something.

then set the review cadence — a delivery check at week six, a verdict at month three, with the numbers named in advance. that structure protects both sides: it gives the agency room to do foundation work without being judged on week-two revenue, and it gives you a defined moment to act if it is not working.

and tell them how you like to work. preferred channel, meeting frequency, how much detail you want, whether you want to see work in progress or only finished pieces. it sounds trivial; mismatched working styles cause more friction in the first quarter than capability gaps do. our own account setup starts with exactly that conversation, and the terms we publish cover the rest.