Flight plan template

A flight plan is one row per week of one line of spend. Where the media plan says what is being bought, this says when it lands: the phase the week belongs to, the share of the flight it is meant to carry, the creative in market, the dependency it is waiting on and the decision due at the end of it.

That is the whole shape. Everything else on this page is about what belongs in each column, and why the two columns that look like paperwork — the dependency and the decision point — are the ones that stop a flight from running to its end whatever the numbers say.

Write it after the media plan is approved and before anything is built in an ad account. The plan settles what is bought; this settles the order it arrives in, and it is the document a week is checked against later. A week that spent nothing like its share is worth knowing about on the Monday rather than at the end of the flight.

Use it again every Monday. Put the week that just closed beside the row that planned it, and read the gap. A line can be on budget for the month and have spent it in entirely the wrong weeks, and this is the only document that can say so.

The media plan and the flight plan answer two different questions about the same money. The plan says what is being bought: one row per line of spend, with a channel, a campaign, an objective and an audience around it. The flight plan says when it lands. One row here is one week of one line, and the column carrying the meaning is the date.

Teams that keep only the media plan find the gap in the second week. The plan said a number for October and the account spent a third of it in the first five days, and nobody can say whether that is wrong, because the plan never described what October was supposed to look like from the inside. A flight plan describes it: this weight in the opening week, this much when the offer opens, this much held back for the last ten days, and here is the week where somebody has already agreed to make a call.

That is the whole shape. A week per row, a phase per stretch of weeks, a share of the flight on each one, and a decision written into the row where it is due. Everything below is about what belongs in those columns and why the ones that look like paperwork are the ones that stop an argument later.

A phase is a run of weeks with one intent. Five cover almost every flight, and most flights use three of them.

A tease phase buys attention before there is anything to buy. It exists where the offer has a date — a drop, an open enrolment, an event — and its weight is small and its metric is not a cost per acquisition, because nothing is being acquired yet. Judge it on reach into the audience you will sell to, and say so in the row rather than letting it be measured against the launch phase's target.

A launch phase is the first week the offer is available, and it is the phase where the weight is most often wrong in both directions. Too little and the auction never learns who to show it to; too much and the account buys its entire audience before a single creative has been read backwards. The honest answer is not in any template: it depends on how long your buying cycle is and how wide the audience is, both of which you can read from your own account and nobody can read for you.

A sustain phase is the long middle. Flat weight, steady creative rotation, and the phase that quietly consumes most of the budget in most flights. The risk in sustain is not overspend, it is drift: the creative ages, the audience saturates, the cost per result climbs a few percent a week, and because no single week looks like a problem, nobody names it until the flight is over.

A peak phase is a deliberate concentration of weight around a date that matters — a sale, a deadline, a seasonal high. It is worth its own phase name even when it is a single week, because it is the week that will distort every average that includes it.

A wind-down phase is the tail. Spend falling, retargeting only, the last of the audience that started and did not finish. It is the phase most plans leave off, and leaving it off is how a flight ends abruptly on a Sunday with a queue of half-finished journeys nobody follows up.

The share column is the one that makes the shape visible. Fill it as a proportion of the whole flight for that line, and the row of shares down the sheet is the flight's profile: front-loaded, flat, or held back.

Three profiles cover most cases and each says something different. A front-loaded flight buys its audience early and lives on retargeting afterwards, which suits a product with a short consideration cycle and a fixed deadline. A flat flight spends the same every week, which suits an offer with no date on it and an audience that renews. A held-back flight keeps its weight for a peak, which suits anything with a calendar event at the end and is the profile most often ruined by an early week that overspent.

What no template can give you is the numbers inside the profile. The right weight for an opening week is a fact about your auction, your audience size and your buying cycle, and the only place it is written down is your own account history. Read it there, then write it here. A share that came with the file came from nowhere, and it would be defended for the length of the flight by people who had no idea where it came from.

Where a line is genuinely new — a new channel, a new offer, no history to read — leave the share blank on the first pass and write the reason in the notes. A row marked "no history, first flight, weight set after week one" is more useful than the same row with a confident proportion in it.

The column that makes a flight plan different from a spending schedule is the decision point, and the rule is that every phase carries at least one.

Write it as a question with a date attached: whether to widen the audience after the first full week, whether the sustain creative is tired enough to rotate, whether the peak weight still makes sense given what the launch cost. The point is not that the answer is knowable in advance. It is that the week is agreed in advance, so that the conversation happens on a Monday with the numbers in front of everyone rather than on the day somebody happens to look.

A decision point also gives a week a purpose beyond its spend. The week before one is a week you are gathering evidence in, and that changes what you do with it: you hold the variables still, you do not rotate three things at once, and you make sure the reporting window will actually close in time to read.

Flights without decision points run to their end whatever happens, because there is never a moment where stopping is somebody's job. That is the failure this column exists to prevent, and it costs one sentence per phase to prevent it.

Every flight depends on something outside the ad account: a landing page, a product feed, a shoot, a legal review, a price change. The dependency column holds the one this week needs, with the date it is due, and the date belongs before the week that needs it rather than inside it.

This is the least glamorous column and the one that saves the most money. A launch week that arrives with its landing page unfinished does not pause politely; it spends its budget sending people to whatever page was in the ad, and the cost of that week is invisible in the platform's reporting because the platform thinks it did its job.

Put the dependency on every row that needs it, not only on the row where it first appears. A shoot that is late affects the three weeks that were going to use the footage, and a plan that names it once names it in the wrong week.

At the end of every week, put the plan beside the account: planned share against actual share, planned budget against actual spend, row by row. Three gaps are worth naming out loud.

The first is pacing against the profile rather than against the total. A line that is on budget for the month and has spent it in the wrong weeks has not followed the plan, and the flight plan is the only document that can tell you so.

The second is the phase boundary that did not happen. A sustain phase that was supposed to end two weeks ago and is still running is not a plan being followed; it is a decision point that passed without anybody making a decision.

The third is the dependency that slipped and took a week with it. Record it in the notes on the row, not in a separate thread, because at the end of the flight the row is what gets read.

Six checks. Every week of every line has a row, including the quiet ones. Every phase names its first week, its last week and what it is for. Every share is a proportion of the same flight, and the shares sum to one whole per line. Every budget is an integer in minor units and every row names its currency. Every dependency has a due date earlier than the week that needs it. And every phase carries at least one decision point, written as a question somebody has agreed to answer.

A flight that passes those six is not necessarily a good flight. It is a flight that can be argued with in advance, which is the only kind worth writing down.