Running a creative agency RFP for a mobile game
A working template for a mobile game creative agency RFP: the brief pack, a sample test, a scoring matrix, rights clauses and a realistic timeline.
A creative agency RFP for a mobile game should run in four stages. A brief pack that gives every agency the same context, a paid sample test scoped to a single concept sprint, a scoring matrix that weights craft and process evenly, and a contract stage that settles rights and data handling before any retainer starts, all inside a timeline of roughly four to six weeks from brief to decision.
Skipping any one of those stages tends to show up later as a mismatch: an agency that pitched well but produces work off-brief, or a contract renegotiated mid-flight because nobody settled ownership of the assets upfront.
What should a creative agency RFP include?
Start with a short list. Three to five agencies, drawn from genre-relevant work rather than general reputation. A longer list mostly adds coordination cost without improving the eventual decision. Send every agency the same brief pack at the same time, and set a fixed window for questions so that one agency does not end up with more context than another purely because they asked first.
Structure the process as three gates: a written response to the brief pack, then a paid sample test for agencies that pass the first gate, then a final decision based on the scored sample plus commercial terms. Keeping the gates separate stops a strong pitch deck from substituting for actual delivered work, which is the single most common failure mode in creative agency selection.
What goes in the brief pack?
The brief pack should give an agency everything it needs to produce relevant work without needing a call to fill gaps. That means the game's core loop and genre positioning, the player motivation the creative needs to speak to, and the specific metric the studio will judge success against. Add two or three examples of what has and has not worked in past creative, described in enough detail to be useful without requiring the agency to have seen the dashboards.
A good brief separates the fixed parts from the open ones. Game name does not move. Neither do art style constraints or any regulatory limits, such as age-rating restrictions on the creative. Hook direction is open, as is format mix, and tone is open too. The brief should say so explicitly, since an agency that treats an open question as fixed will pitch something narrower than the studio actually wanted.
The earlier piece on this site, "What a good creative brief looks like in 2026", covers the brief itself in more depth and is worth reading alongside this template.
Should you ask for a paid sample test?
Yes, and the studio should pay a fair rate for it rather than treating it as free pitch work. An unpaid speculative round selects for agencies willing to work for free, which is not the same as selecting for the best agency, and it biases the process toward larger shops that can absorb the cost rather than smaller specialists doing genuinely strong work.
Scope the sample test tightly: one concept direction, a fixed small number of variants, a defined turnaround. The goal is not a finished campaign. It is a fair comparison of how each agency interprets the same brief under the same constraints. Score the sample against the brief's stated goals, not against which agency's reel looked most polished going in.
What belongs in a scoring matrix?
A workable matrix weights four areas roughly evenly: how closely the sample matched the brief, the technical execution quality, the clarity of the agency's own rationale for their choices, and commercial fit, which means price together with team availability and contract terms rather than price alone.
Score each area independently before combining them. Involve more than one person in the scoring, which catches a single reviewer's bias toward a particular visual style. A matrix that only measures craft will systematically favour agencies with the flashiest reel over ones that reliably hit the brief, and that is usually the wrong trade for an ongoing relationship.
What rights and data clauses matter most?
Settle asset ownership before signing, not after.
The contract should state plainly whether the studio owns the final delivered assets outright and, separately, what happens to unused concepts and raw production files from the sample and any later work.
Data handling needs the same clarity. If the agency will see any performance data to inform iteration, the contract should specify what they can retain. It should also specify whether they can reference the work in their own portfolio, and whether that portfolio use requires the studio's sign-off before publishing. Studios that skip this step commonly find their own creative appearing in an agency's public case studies before they intended it to.
Add a clause covering exclusivity and category conflict as well, particularly for a studio operating in a crowded genre. An agency that also serves a close competitor is not automatically a problem, but the contract should say how the agency handles brief confidentiality across those accounts rather than leaving it as an informal assurance made verbally during the pitch. The same applies to subcontracting: if the agency plans to route any part of the work, such as UGC sourcing or voice talent, through a third party, the contract should say so and extend the same data and rights terms to that third party.
Finally, put a real timeline on the whole process and hold to it. A brief pack sent with a two-week response window, followed by a one-week sample sprint and then a scoring and negotiation window of a further one to two weeks, gets a studio from open RFP to signed agency in around four to six weeks, long enough to be fair to every agency and short enough that the studio is not left without creative support in the meantime.
