handing over to a new agency.

the transition is the most avoidable expensive moment in an agency relationship. everything you need is easy to obtain while the outgoing team is still being paid and still cares, and unexpectedly difficult two weeks after you have told them it is over. that sequencing is the whole game.

the checklist below is what we ask for when taking over an account, and what we hand over when a client leaves us. it is worth running through it before you make any announcement.

the reason to be systematic is that the losses are cumulative and quiet. a missing pixel costs a month of clean data. missing raw footage costs a reshoot. a lost keyword map costs a rediscovery exercise. individually each is an inconvenience; together they are the reason switching agencies so often produces a worse quarter than staying with a mediocre one.

— 01do these before serving notice.

three things, in order. first, confirm ownership of every account — ad accounts, business manager, google properties, analytics, search console, page admin, the domain and the hosting. you should be the owner with the agency as an added user, not the reverse — it is the first line of our own terms for the same reason. if it is the reverse, fix it now, quietly, as an administrative housekeeping request.

second, get the raw files. edited exports are not the same as source footage, layered design files and original photography. an agency that has shot for you for a year holds a library that is worth real money to reproduce, and monthly delivery should have been happening all along.

third, ask for tracking documentation — what is installed, what fires, where conversions are defined and what the definitions mean. this is the least glamorous item and the one that most often gets lost, at which point your next agency spends month one rebuilding measurement rather than improving performance.

none of these requests are unusual and none of them signal that you are leaving. they are simply what a well-run account looks like, which is why asking for them is safe.

if any of the three meets resistance, you have learned something important about the relationship regardless of whether you were planning to leave. an agency that will not transfer account ownership or deliver source files is building exit friction deliberately, and that is worth knowing about a year before you need it.

— 02what to ask for during the notice period.

the calendar and the pipeline: what is planned, what is half-finished, what is scheduled to publish after their last day. brands routinely discover a gap in week one of the new relationship because scheduled content ran out and nobody said so.

the working documents: playbook, brand guidelines, templates, hook libraries, audience research, keyword maps, negative keyword lists. these were paid for by you and they are what makes the next team productive quickly rather than slowly.

and the account history: what has been tested, what failed, what the current strategy is trying to do. a thirty-minute call between the outgoing and incoming teams saves the new agency a month. it feels awkward to arrange and it is the single highest-value thing in the whole transition.

frame it neutrally when you ask. most professionals will take that call if it is presented as a handover rather than an interrogation, and outgoing teams often want the record to show that the account was in reasonable shape when they left. make it easy for them to look good.

— 03what the new agency should do in week one.

audit what they inherited before changing anything. that includes verifying tracking rather than trusting the documentation, reading the last six months of performance, and identifying which of the current campaigns are actually carrying results.

the thing to resist is the clean-slate instinct. a new team is tempted to restructure everything, partly because it is easier than understanding what exists and partly because it looks decisive. it also destroys learning and resets platform optimisation, which usually produces a worse month two than doing nothing would have.

the reasonable exception is measurement. if tracking is broken, fix it immediately — every other decision depends on it, and inherited measurement problems have a habit of being blamed on the new agency six months later.

— 04the overlap question.

two weeks of paid overlap between agencies is almost always worth it. it costs a fraction of a month's fee and it eliminates the gap where nothing publishes and nobody is watching the ad account. if budget makes overlap impossible, at least stagger it so the new team has access before the old one loses interest.

finally, write down what went wrong last time and share it with the incoming team, including your own contribution to it. approval delays, unclear priorities, a strategy nobody agreed — those recur unless someone names them. the transition is the one moment when everybody is motivated to be honest, and it is worth using.