"How much does a playable ad cost?" is asked as a per-unit question and almost never answered usefully that way. The unit price is the smallest part of the real cost, because playable advertising is an iteration business: the creative that works is the fourth or fifth version, not the first. This guide covers the production routes, what each really costs, and why turnaround time usually dominates the maths.
The three routes
Custom development, creative agency or studio, and in-house with a builder tool. They differ less in headline price than in what happens after version one.
1. Custom development
A developer builds the playable from scratch in HTML5, wires each network's SDK requirements by hand, and packages per network. This gives complete creative freedom and the highest per-unit cost. The part that is consistently underestimated is the per-network work: each network wants its own click call, packaging shape and size ceiling, and that work is repeated for every revision, not just the first build.
Turnaround is typically measured in weeks for a first version and days for a revision. It suits genuinely bespoke mechanics that no template approximates — and it is expensive to use for the variant testing that actually drives performance.
2. Agency or creative studio
Specialist playable studios sell per creative, often with tiered pricing by complexity, and frequently with a minimum engagement. You get professional creative direction and no internal production burden. The costs that surprise teams are on the revision side: variants are usually billed separately, and every change goes through an external queue.
This is a good fit when you need a small number of high-craft creatives and have no internal capability. It fits badly with rapid variant testing, because the cost and latency of each iteration discourage exactly the behaviour that improves results.
3. In-house with a builder
A subscription tool, and your existing designers or UA people produce the creative. Per-unit cost approaches zero after the subscription; the real inputs are the designer's time and the learning curve. Ceiling on creative ambition is lower than custom development, though modern builders cover most of what performs — configurators, mini-game mechanics, video with interactive breaks, and end cards.
The decisive advantage is iteration latency. A variant is minutes rather than days, which changes what you can test.
The number that actually matters
Not cost per creative — cost per test. If a variant costs several hundred and takes a week, you will run a handful per quarter. If it costs effectively nothing and takes twenty minutes, you will run dozens. The team running dozens finds the winner, and that gap dwarfs any difference in unit price.
The hidden costs nobody quotes
- Rejections. A rejected creative costs the resubmission cycle and the campaign start date, which is often the expensive part.
- Per-network variants. "One playable" usually means a package per network, and if that is manual it multiplies with every revision.
- Localization. One creative per market is one maintenance burden per market, unless language is a property of the project rather than a copy of it.
- Orientation. Portrait-only creative loses landscape inventory quietly, and a second orientation built as a separate project doubles maintenance.
- Drift. Several near-identical files across markets and networks, and no certainty which is current. This costs real time and is never in a quote.
What to actually budget
Budget for a series, not a creative. A realistic first campaign is one concept, three to five variants of its opening seconds and end card, and a refresh cycle every few weeks as performance decays. Whichever route you choose, price that shape of work rather than a single unit — it is the only comparison that reflects how playables are actually run.
A reasonable default
For most teams: build in-house with a tool for the volume and the testing, and reserve custom development or a studio for the one or two flagship concepts where a bespoke mechanic is genuinely the point. That combination keeps iteration cheap where iteration matters and buys craft where craft matters, instead of paying studio rates to test a button colour.
A worked example
Consider a quarter with one concept, tested properly. That means the base creative, roughly four variants of the opening seconds and end card, both orientations, three localized markets, and packaging for four networks. Then a refresh at week six as performance decays, which is another base plus two or three variants.
Counted as deliverables rather than as "one playable", that is somewhere around a dozen distinct exports and two rounds of creative work. Price each route against that number and the comparison changes shape: a per-unit price that looked reasonable for one creative rarely looks reasonable multiplied by twelve, and a subscription that looked expensive against a single unit usually does not.
The same arithmetic explains why teams under-test. If each variant carries a real cost and a week of latency, four variants is a large commitment and most teams quietly ship one. The creative that would have won is never built, and its absence never appears in any report.
When to switch routes
- From agency to in-house — when you are commissioning variants of an existing concept rather than new concepts. Variants are exactly the work a builder does well and an agency prices poorly.
- From in-house to studio — when your concept genuinely needs a mechanic no tool approximates, and you have data suggesting the concept is worth the spend.
- From custom development to a builder — when most of the developer's time is going into per-network packaging and revisions rather than the mechanic itself. That is the clearest signal the cost has moved from creation to maintenance.
Ask about revisions before you sign
Whatever the route, the question that predicts your real annual spend is not the price of a creative. It is: what does version two cost, and how long does it take? Get that answered in writing before committing, because that is the number you will be paying repeatedly.
The costs that sit outside production
Two more line items belong in any honest comparison, and neither appears on a production quote.
The first is media waste while you are learning. Early creatives underperform, and that spend is part of the cost of arriving at a working one. Teams that can iterate cheaply shorten this period because they reach the winning creative sooner; teams on a slow cycle pay for it in media for longer, which frequently exceeds whatever they saved on production.
The second is ownership. If an agency or studio holds the source project, every future change goes through them — including small ones like a seasonal headline or a new store URL. Ask explicitly who owns the working files and in what format they are delivered. "You own the output" and "you own the project" are very different answers, and the difference only becomes apparent when you need a change and the relationship has ended.
Questions to ask any supplier
- What does a variant of an approved creative cost, and how long does it take?
- Which networks is the deliverable packaged for, and is a further network a new charge?
- Are both orientations included, or is landscape a separate creative?
- Are localized versions included, and what happens when a string changes afterwards?
- Who owns the source project, and in what format is it handed over?
- What happens if a network rejects it — is the fix included, and how quickly?
Create playable end cards in minutes—no code required.
Open Playable Ads Maker