Media planning · Marketing science platform
Kuwalyst helps agencies and brands build, monitor and optimise their media plans using models, experiments and audited benchmarks. You can use the platform on your own or work with our team. Every figure is traceable back to its source.
A media plan can lose relevance soon after approval as platforms shift, creative performance changes and saturation moves. Rather than relying on an annual refresh, teams can update their plan as the evidence develops.
The planner brings planning, monitoring and optimisation together in one continuous process, so budgets can be adjusted when the evidence warrants it.
It is built on open methods, with models your team can inspect, assumptions it can audit and decisions that remain traceable from start to finish.
Compare budget scenarios based on your data or audited benchmarks, with the projected return, uncertainty range and trade-offs made clear for each option.
Manage the plan in market with goals, pacing and alerts in one view. Each metric is identified as measured, projected or benchmarked, and deviations are accompanied by the supporting evidence.
Review recommendations that set out the hypothesis, proposed action, expected impact and confidence level. Shift budget along response curves, test what remains uncertain and reduce spend where channels are saturated.
Figures are identified as measured, projected or benchmarked, and recommendations show both their confidence level and the evidence that supports them.
Observed in your warehouse or through an experiment you ran. These figures provide the reference point for the other estimates.
A model's estimate, shown with its uncertainty range. When the range includes zero, the plan says the effect may be nil.
Drawn from more than 200 benchmark sources, each traced to its primary publisher. The registry excludes unsupported articles and makes any gaps visible.
Asked for a TikTok ROAS on a brand that has never run TikTok, the planner checks the warehouse, reports that there is nothing to measure, and offers the closest tracked channel instead of an invented number.
Start with audited benchmarks when data is not yet available, then move to a data-backed plan as evidence builds.
Each traced to its primary publisher: eMarketer, IAB, IPA and WARC, Nielsen, platform documentation.
Figures are identified as measured, projected or benchmarked, and the planner makes it clear when evidence is still missing.
We have entrusted our advertising to Kuwalyst. Their marketing scientists plan and run our Meta and Google campaigns on the Kuwalyst platform, and in the first month, traffic to our online shop grew about fivefold and online orders doubled. Each channel has a clear role: on Meta, we build audiences while testing videos; on Google, we reach people looking for Japanese rice. We also learned that videos showing our shop and our rice as they really are resonate more than promotional ones, which makes planning ahead much easier. We already have a first estimate of how much Meta and Google each add on top of the sales we would make without ads, and an incrementality test to verify it is planned. We set the following month’s budget based on these results.
We had handed our GeoLift incrementality analyses to an AI agent, and we wanted an independent view on its methodology using a real case. Kuwalyst ran its own analysis of a real client test, covering test design, statistical power, the estimated effect and its confidence level, then compared it with ours and with the agent's. For that test, we now know what holds up statistically and what still needs work.
Explore benchmark-based media plans and budget scenarios without connecting your advertising accounts.
A monthly platform fee set by the number of brands: one, up to three or up to ten. Every plan includes onboarding, a monthly review with a marketing scientist and automated execution under human oversight.
A management fee applies to the media spend Kuwalyst manages, at a lower rate on each higher bracket. Custom marketing science and data work is scoped separately.
The planner uses Scaleway Generative APIs, hosted in the EU (France). Privately hosted models are available on request, including on-premise and VPC deployments. Read where your data goes, and where it stays.
We document the methods used in Kuwalyst openly. Explore the book or a method explainer, and look out for our forthcoming paper on media planning with AI.
BookCo-founder Hajime Takeda's practical guide to causal inference, MMM, CLV, pricing and forecasting—the methods that inform Kuwalyst.
Get the free PDF →How agentic systems build, monitor and tune media plans, and the guardrails that keep every number tied to evidence.
Coming soonThe Media Planner is software licensed by agencies and brands. If you prefer a managed service, our team can operate the same platform on your account. In either case, you retain the models, plans and supporting evidence.
Platform dashboards report attributed sales. Attribution overlaps across channels, and reported ROAS stays high on channels that are already saturated. Kuwalyst connects to your data sources, so the same platform figures are available per channel and aggregated, for online and offline channels alike. It then combines them with a media mix model and experiments, and tags every figure as measured, projected or benchmark, so budget decisions rest on evidence.
Yes. Benchmark-led plans begin with your brief and an audited registry of published benchmarks, with claims checked against primary sources. Each figure is identified clearly, and the plan prioritises the tests that will allow your own evidence to replace benchmarks over time.
Many media plans remain spreadsheets that are revisited once a year. Kuwalyst supports a continuous process, combining scenarios with explicit trade-offs, weekly goal monitoring and recommendations that show their confidence and evidence. When the plan changes, the reason is visible.
No. Agencies are among the planner's primary users, and we do not sell directly to their clients. With the white-label option, plans carry your name and reports display your logo.
Your data is held in an EU-hosted workspace on Scaleway. By default, AI processing uses Scaleway Generative APIs in France. Privately hosted models, including on-premise and VPC deployments, are available on request.
Tell us where your media plan lives today—even a spreadsheet is a useful starting point. We will show you how planning, monitoring and optimisation could work for your own situation.
Talk to us