A 30 / 60 / 90 day plan for the Account Manager, Ads role —
written before the conversation.
"Push play. Do you stay?"
— Ted Sarandos · co-CEO, Netflix · on the only metric that mattersHi —
I'd like to have a conversation about the Account Manager, Ads role. Rather than describe what I'd do in a first meeting, I've written it down. What follows is a thirty-sixty-ninety day plan built on a single thesis: don't "learn the process." Ship improvements inside it. My job is to make campaign delivery boring — in the best way — so the conversation moves from "did it run?" to "what's the next flight?"
Sarandos calls Netflix's core retention metric "push play, do you stay?" The same loop applies to advertising: press launch — do they renew? Predictable launches, stable pacing, clean billing, and wraps that earn the renewal — that's the role. The plan below is opinionated by design. The numbers and named systems are placeholders I'd replace, on day one, with whatever Netflix actually uses.
Read it as a memo from the operator I'd be in the seat — not as a pitch. If something is wrong about how Netflix Ads runs today, that's the most useful conversation we could have.
By Day 90, for my book of business —
| On-time launch rate | +15–25% improvement (baseline-dependent) |
| Asset-on-time rate | +20% via clearer specs + earlier enforcement |
| Escalation cycle time | Same-day closure priority-issue standard |
| Wrap turnaround | 48 hours campaign end → client-ready |
| Makegoods (ops error) | Near-zero preventable by Day 90 |
| Internal time saved | 3–5 hrs/wk recaptured via templates + automation |
— Shourjo Dasgupta
Melbourne · may 2026
An account manager's first thirty days at Netflix Ads should not be a tour of products. They should be an audit, then a contribution. The job is to draw the map of how a campaign actually moves from a signed contract to a first impression — and to take ownership of one operational slice fast enough that the team trusts the next sixty days.
Shadow Account Executives, Ad Ops, and the measurement partner workflow on real campaigns. By Day 10, take ownership of a low-risk operational slice — asset chase, spec checks, trafficking QA, or wrap drafting — whatever is most painful. Then build the Campaign Path map: every handoff from IO → trafficking → QA → launch → pacing → wrap → billing, with owners and failure points named.
Run a weekly fifteen-minute Launch Desk with AM, Ad Ops, and the measurement rep: what's launching, what's blocked, what's at risk. Replace Slack archaeology with one shared operating view of active campaigns: launch date, asset status, pacing, measurement readiness, billing flags.
Pick one bottleneck and fix it without permission theatre. Candidates: asset intake / spec compliance, a launch QA checklist, a pacing alert, or a wrap automation. The improvement should save measurable time or prevent a preventable error — small enough to ship in thirty days, real enough to point at.
A one-page "State of Delivery" map + the top 5 friction points ranked by frequency × impact × fixability. Plus one shipped pilot.
Velocity is what an account manager actually sells. Once the map exists, the next thirty days are spent removing the bumps in it. The objective is to make execution so unremarkable that the team stops talking about it and starts talking about the campaigns themselves. That's the only condition under which renewal conversations get bigger instead of harder.
A wrap a buyer can forward to their CMO. Four parts: Delivery — what ran, where, how it paced. Proof — measurement and verification readiness. What we learned — audience and context signals. What to do next — renew / expand logic. Build the template plus examples for two campaign types: standard, and premium / sponsorship.
Define pacing guardrails by campaign type. Set alerts and escalation thresholds — when do we notify the client versus fix quietly? Track root causes: late assets, targeting shifts, inventory constraints, trafficking error. The pacing playbook becomes a small exception log: what broke, why, how we prevented recurrence.
One area that creates repeat friction — billing-dispute triggers, makegood-process clarity, measurement-setup checklist, or creative-QA failure rate. Ship the change. Get it adopted by at least one adjacent team (Ad Ops, billing, measurement).
A Standard Wrap Template v1 + a Pacing Playbook + one shipped cross-functional change adopted by an adjacent team.
By Day 60, the team should be running a faster process. By Day 90, the process should be running with less of the team. The final phase is leverage — the systems, automations, and packaged knowledge that make the next thirty days require less of mine. This is where the work I did at Locky Dock with Claude Code becomes directly relevant: a tightly-scoped internal assistant the AM team can interrogate as if it were the most senior person in the room.
Five pieces a new AM can run on day one: a Kickoff checklist that prevents bad inputs; a Trafficking QA checklist; a Measurement readiness checklist; a Wrap generator (template + auto-fill fields); an Escalation ladder — who to pull, when, with what info.
A scoped internal helper that touches only approved docs: specs, naming conventions, measurement requirements, what "good" looks like in wraps, common troubleshooting flows. Sandbox prototype plus governance notes — what it can answer, what it can't, where it sources from. Permissions, ownership, sourcing — in that order.
Every wrap ends with a "next flight recommendation" tied to actual performance. A 90-day forward view of renewals and expansion targets makes the pipeline predictable. Wraps stop being deliverables. They start being sales tools.
An AM Ops Kit v1, a Knowledge Assistant prototype with governance, and a 90-Day Performance Scorecard showing cycle-time improvements + a renewal pipeline view.
Won't redesign core systems before running real volume through them. The map gets drawn from the inside, not from a slide deck.
Won't push AI into production without permissions, sourcing, and clear ownership. Sandbox first; ship second; govern always.
Every new step has to remove a bigger step. If a checklist doesn't save time or prevent error, it's added cost — that's not the deal.
By Day 90, my accounts run on a tighter machine: fewer fire drills, fewer surprises, faster wraps, smoother renewals. The work becomes boring. The results become easier to sell. Push play — do they stay? Press launch — do they renew?