How to Plan Creative Capacity for Game Marketing.

A creative pipeline is the route a piece of content takes from idea to published post, and the set of stages it passes through on the way.

Briefs, gameplay capture, art and player clips go in at one end. Finished, approved, scheduled posts come out at the other.

Creative capacity is how much of that a studio can carry in a week without work piling up between the stages.

Most game teams can describe their pipeline. Very few can tell you its number. That number decides how often the game can publish, and publishing rhythm is what organic growth actually runs on.

What goes into a creative pipeline, and what comes out

The inputs are wider than ideas. A game’s pipeline is fed by the build itself: gameplay capture, patch notes, art, character reveals, live event dates, and whatever the community has already made and posted.

The outputs are narrower than most teams admit. A published post, and what the team learns from it. Everything else is work in progress.

Between the two sits a fixed set of stages:

• Brief: what the piece is, who it is for, what it has to do
• Capture: the footage, art or screen recording it needs
• Make: the edit, the design, the copy
• Review: the decision to publish, change or drop it
• Schedule and publish
• Read back: what it did, feeding the next brief

A calendar and a pipeline are not the same object. The calendar says when something publishes. The pipeline decides whether it will exist by then.

Capacity is the ceiling on how often a game can publish

Everything a studio wants from its own channels sits downstream of production. Consistency, format iteration, a series players come back for, a community that gets replies: none of it survives a pipeline that cannot feed the calendar.

When capacity runs short, nothing announces it. A slot slips, then a week goes quiet, then the channel gets relaunched six months later with a new look and the same problem underneath.

The case for why a game’s growth should not depend on the ad account is made in why mobile game ads need a new organic strategy. The operational half of that case is simpler and harsher. Owned channels only compound if the studio can produce at the rate compounding requires, including the weeks when nothing exciting is happening.

How many concepts a week does the game actually need

Work it out from the calendar, not from a feeling about how busy everyone is.

1. Count the publishing slots for the next month, per channel, per week. That is the demand.
2. Strike out the slots that do not need an originated concept: a community repost, a reply, a raw clip of the new boss fight, a screenshot of the patch notes. Those cost time, but they do not consume creative capacity.
3. What is left is the number of originated concepts the calendar is asking for.

Now put that next to what the team actually shipped in the last four weeks. Not what it planned to ship. What went out.

No industry benchmark helps here. The two numbers that matter are your calendar’s demand and your team’s real output, and you already own both.

Concepts and variants are not the same unit

A concept is an idea: a new format, a new angle, the next episode of a series. A variant is that idea cut again for somewhere else, another aspect ratio, a shorter version, a different opening, a different title.

They cost different things. A concept costs thinking and a review decision. A variant mostly costs edit time.

Counting assets per week mixes the two and hides the answer. Count them separately and the sizing becomes arithmetic.

Suppose the calendar asks for twelve slots a week, each concept yields three usable cuts, and three in four briefs survive review. Twelve slots divided by three cuts is four concepts a week to fill the calendar, and five or six started to land four.

Your numbers will differ. The method does not. Measure your own survival rate by taking the last twenty briefs and counting how many were published.

Who reviews, when, and for how long

An approver with no slot in the diary is the most common bottleneck in any pipeline, and it hides well, because it looks like a busy person rather than a broken process.

Four things fix most of it:

• One named approver per stage. A group is not an approver, it is a discussion.
• A recurring slot, at a time the work is genuinely ready. Two short sittings a week clear the queue twice; one long sitting clears it once.
• A stated turnaround and a stated default. Decide in advance what happens when a slot is missed, because silence is already a decision, just not one anybody chose.
• A decision, not a document. The reviewer is approving this stage only. Reopening a settled stage is a new brief, not feedback.

Batching helps more than it should. Twelve items in one sitting is faster than twelve interruptions, and the judgements come out more consistent, because the reviewer is comparing rather than reacting.

Capacity to make is not capacity to approve

These are two different numbers, and almost nobody measures the second one.

Production capacity is items finished per week. Approval capacity is decisions made per week. A pipeline’s throughput is the smaller of the two, never the average.

Approval is usually the smaller number, and it does not scale the way production does. Adding an editor adds output. Adding an approver usually adds a round.

Measure it like anything else. Time one sitting, count what got cleared, multiply by the sittings in a week. If a producer clears an item in five minutes and sits down twice a week for half an hour, that is twelve decisions a week, and no amount of editing capacity gets a thirteenth post out of the door.

That split is the first thing NBK sizes when it takes on mobile game production. Not how many videos a week the studio wants, but how many decisions a week it can make.

Where the queue backs up, and how to spot it early

Count the work sitting in each stage once a week, on the same day. Six numbers, five minutes.

The stage whose number grows two weeks running is the constraint. That is the whole diagnostic.

Size alone will mislead you, so count age as well: the oldest item in each stage. Twelve items that clear in two days is a healthy queue. Four items where the oldest is three weeks old is a stall wearing a small number.

The arithmetic is unforgiving. The wait in front of a stage is the size of its queue divided by how fast that stage clears work. Ten items in front of a reviewer who clears five a week is a two-week wait, whatever the calendar says.

The early warnings are behavioural before they are numerical:

• The same item comes up in three consecutive stand-ups
• Work comes back from review with thoughts rather than a decision
• Yesterday’s post goes out today and nobody flags it
• Someone starts making unbriefed content, because it feels better than waiting

When the pipeline is the constraint, cut something deliberately

Three levers, in the order that costs the least:

1. Cut variants first when the format is already proven. You know it works, so the fourth cut of it is the cheapest thing you own.
2. Cut concepts when review is the constraint. A concept the approver skimmed will publish, underperform and teach nobody anything, which is worse than a smaller slate judged properly.
3. Cut review stages when a stage has never sent anything back. A stage that has never rejected anything is not quality control, it is a delay with a job title. Merge it or remove it.

The rule underneath all three is the same. Cut whatever costs you the least learning, and cut it on purpose.

The worst version is the accidental cut, where the slot is simply missed, nobody decides anything, and the team quietly learns that the calendar is optional.

The option most studios skip: publish less on purpose

Setting the calendar to what the pipeline sustains is not a retreat. It is the only way the calendar means anything.

A schedule the pipeline cannot feed produces gaps, and a gap costs more than a smaller cadence, because players read the gap and the team learns that the plan was fiction.

There is a quieter cost too. When the calendar outruns capacity, review is the first thing to get bypassed, and a bypassed review is how a channel ends up publishing a stale build, a wrong price or a spoiler nobody signed off.

What good looks like

A studio with its creative capacity under control can answer six questions without opening a spreadsheet:

• How many originated concepts do we publish a week, and how many do we start
• Who decides at each stage, and when do they sit down to do it
• How long does an item take from brief to published
• Which stage is the queue in this week
• How many decisions a week can our approver actually make
• What gets cut first when a week goes wrong

None of that needs new software. It needs somebody to own the numbers and say them out loud every week.

How NBK thinks about creative capacity in games

NBK’s team came up in publisher-scale production, 46,000+ posts shipped and 600M+ views a month, and the lesson from that volume is that the constraint is almost never the making.

So the sizing starts on the decision side. Who approves, how often they sit down, and how many items they can clear. Cadence is then set to that capacity, and raised only when the capacity genuinely rises.

What is being sized is the studio’s own output: the video, the community content and the series it publishes itself. That is the production line worth protecting, because it is the one the studio keeps.

Next step

If your team is producing plenty and still missing slots, the constraint is in the system rather than the people, and it is usually a decision that has nowhere to happen.

If review is where your pipeline stalls, the approval process template is a fast way to write down who decides what, by when, and what happens if they do not.

If your social process is slowing down good ideas, NBK can help rebuild the workflow behind the content.