
Recast, an incrementality measurement and
forecasting company, has made statistical experimentation accessible to all advertisers and marketers.
The launch of “GeoLift by Recast” on Tuesday provides a standalone
incrementality platform to determine the “true” cause or impact of marketing on a business.
The platform simplifies the process by allowing marketers to run and
understand the results of the data without needing an advanced degree in statistics or a team of engineers to explain the results.
“Media-mix models attempt to understand incrementality,
how it impacts the business, so by running these campaigns outside of MMM, it gives us another way to compare the data,” said Michael Kaminsky, co-chief executive officer at Recast.
Marketers can explore different experimental designs, understand historical variance, and run experiments without relying on outside consultants or data-science teams.
advertisement
advertisement
The platform also
helps validate MMMs and feed model refreshes with reliable calibration data.
For example, if a marketer turns off a brand search campaign in half of the country to see how much revenue drops,
it will provide an estimate of the true incremental return on investment (iROI). The marketer then compares those numbers with those from the MMM data. The outcome tells if the strategy works or if
it’s misleading, Kaminsky said.
“It compares and corroborates the data,” he said, to determine the cause of the sales and conversions. Test and control groups establish causation, not just correlation, and help marketers allocate budgets to avoid wasted ad spend.
The platform, which is
free to use for the first six months, is based on the core algorithm of the econometric methods developed at Harvard and UC Berkeley.
John York, analytics and strategy team lead at digital ad
agency Acadia, used the model and found it easy to use because the market selection tool saved the team time when identifying the best test markets.
York said it provided clear data to
validate decisions for Acadia’s clients, which led to more trust, transparency, and willingness to test into new channels.