Almost every agency owner has felt it, and very few have systematised it: by the time a client tells you they're leaving, they've usually decided months earlier, often in the first couple of weeks. On an episode of MerchantSpring Marketplace Masters, Scott Ohsman, VP of Digital Commerce at Quickfire, with more than 30 years in the game and over 200 Amazon brand launches behind him, made the case that the outcome of a client relationship is set in the first fortnight, not the first year. Here is where it breaks, and how to stop it.
The damage usually starts at the seam between sales and delivery. A salesperson (or an owner-operator wearing the sales hat) courts the account, builds the rapport, and learns the real problem, because you typically win the business by promising to fix something a previous agency couldn't. Then the baton is handed over, and the delivery team- Ad, account management, creative- has heard only that 'a good account is coming.' They weren't in the room.
So the kickoff call arrives, and it's loaded. On the client side, all the senior stakeholders who signed off are present, several of whom will never join a call again. Expectations are high on both sides. And then the account manager, who doesn't actually know the core problem they were hired to solve, has to wing it. Cue the awkward silence, followed by the after-call Slack drama: 'Nobody told me that was the problem; I haven't even looked at it.' Ohsman's hard-won rule from years of this is blunt: no surprises, unless they're good ones.
The fix is to stop treating the signature as the handoff point. Ohsman's approach, once his agency scaled to 30 or 40 accounts, was to intertwine sales and account management far earlier. Once a deal is roughly 70% likely to close, pull account-management leadership into the late-stage conversations: who will lead this account, what's the workload on the AM and ad teams, and, just as important, is the personality a fit? The single smartest move, where possible, is to get the intended account manager on a call with the client before signing. Many brands ask for it anyway, and it lets both sides sense the fit before anyone is committed.
One thing no process fully removes is client amnesia, and Ohsman is refreshingly honest about it. You can document in the sales process that a problem isn't fixable, or that 3X growth was never promised, and still, in week one, hear 'but sales told me you could do this.' You check, and sales has the email proving otherwise. You largely have to live with it, politely, which is exactly why sales and account management need to be joined up before onboarding, so the real expectations are set as early and as jointly as possible.
There's a widely cited onboarding benchmark that the kickoff call should happen within about five days of signing, because buyer anxiety compounds badly past the ten-to-fourteen-day mark. Ohsman's experience broadly agrees: get it in as soon as possible while enthusiasm is high. The client has taken real risk, moved budgets and stakeholders, and the decision-maker is excited to show off the new agency like a new toy. That honeymoon energy is an asset, so use it.
The caveat is that a productive kickoff needs real prep beforehand, which volume-stretched agencies often skip. Do the groundwork, then keep the call itself somewhat informal: this is the start, the moment to build enthusiasm and, as Ohsman puts it, get the juice going, not to drown everyone in process. The signing moment is the peak of client happiness; it's worth protecting rather than squandering.
The next fault line is the mismatch between who sold the account and who services it. A client courted by a senior strategist can be dismayed to find a junior manager juggling their brand alongside others. The economics often force some version of this, but it has to be hashed out honestly before signing, because putting a junior person nominally in charge of a seven, eight, or nine-figure business is unfair to the client and to the junior. When the fit is wrong, the client escalates fast, bosses get called, and the glow of the courtship is damaged in the first week.
Underneath sits the agency's hardest structural problem: the pod. The account manager is the quarterback who drives ads, catalogue, creative and the rest, and every pod has to pay for itself, so the big accounts effectively subsidise the small ones. Ohsman stresses that volume is only one input. A mid-tier account can be lighter on revenue but heavier in work, a hybrid 1P/3P setup, a huge SKU count, complex advertising, and the account manager, closer to the day-to-day than senior leadership, is the person who actually knows what a pod can absorb. When a match is wrong, fix it early, and make sure the difficult conversation is led by someone the client already trusts, not the new face they met four seconds ago. In the first fourteen days, you have to earn enough trust to have that conversation at all.
Reporting is where relationships quietly break, often around month three, when the client starts asking what the vanity dashboard is actually for. Ohsman's counter is to overdeliver from week one. Respond fast, make sure access and permissions genuinely work (nothing erodes trust like a new client emailing for dashboard access and hearing nothing back), and over-communicate. Part of what those first two weeks are for is training the client on scope, boundaries and what to expect, and generous early reporting is how you set that tone.
Content matters as much as cadence, and it comes down to audience. Know who you're reporting to: a C-level reader wants a subject line and three bullets that answer two questions: what does it mean, and why should I care? They want actionable insight, not a beautiful spreadsheet they have no time to read. And remember the client hired you as the expert, so the fastest way to damage the relationship is to get on a call and ask them whether they should run a Prime Day deal; they called you precisely so you would tell them. Assume everything is read on a phone, one clip away from the next thing, and write accordingly. For agencies, reliable client reporting is also a scale problem, which is why MerchantSpring's marketplace analytics built for agencies exists: standing up a client's reporting before the first engagement is exactly the week-one overdelivery Ohsman describes.
There's a second failure mode that isn't the agency's fault: the client signs, hands over access, then vanishes for weeks- no kickoff, no-shows, silence- before resurfacing two months later, unhappy. Ohsman treats a no-show as a spidey-sense signal that something is wrong, and his advice is to escalate immediately, back to the sales team and the senior relationship who closed the deal, and move it up the chain fast rather than waiting politely.
His most memorable story is the extreme version. Having exhausted every proper channel on an unresponsive new account, he sent an invoice for triple the amount with the email subject line 'Cancelled Service,' effectively firing the client within three weeks of signing. He got a phone call almost immediately and a meeting the next day, and they became a happy, long-term customer; the silence had been circumstances beyond their control, and they'd simply trusted the agency to carry on without them. He is clear it was a last resort he doesn't recommend, but the underlying lesson stands: you can't do the work if you can't get the client to engage, so surface the problem early and unmistakably.
Ohsman's week-one advice- respond fast, get reporting live, answer 'what does it mean and why should I care '- is hard to do by hand across a portfolio. MerchantSpring lets agencies stand up a new client's reporting before the first engagement, share white-labelled dashboards and scheduled reports, and hand roughly four hours a week back to each account manager. See how agencies use it, or book a walkthrough of your own client portfolio.