What is Simulated Market Testing?
Launching a new product is expensive, and rolling it out nationally only to watch it flop is one of the more painful ways to lose a marketing budget. So how do we get a read on whether a new product will actually sell before we commit to a full launch, without tipping off every competitor in the category and without waiting a year for the answer?
That’s the problem simulated market testing was built to solve.
What Simulated Market Testing Actually Is
Simulated market testing, usually shortened to STM (simulated test market), is a research method that estimates how a new product will perform in the real market without actually putting it in stores nationwide. Instead of running a live test market in a handful of cities, which is slow, expensive, and visible to competitors, researchers recruit a sample of target consumers, expose them to the product’s advertising or concept, and then let them “shop” for it in a controlled setting, such as a mock store, an online store simulation, or a mall intercept. Their reactions get fed into a mathematical model that forecasts sales volume for the first year or two after a real launch.
We’re essentially trying to answer two questions before spending millions on a national rollout: will people try this product, and once they’ve tried it, will they buy it again?
Where the Idea Came From
STM isn’t a new idea. Commercial systems started appearing in the 1970s, once researchers realized that traditional test marketing, actually shipping product to a few representative cities and watching what happens, had some serious downsides.
The two names most associated with the origins of STM are Assessor and BASES. Assessor was developed at MIT’s Sloan School of Management in 1973 by researchers Alvin Silk and Glen Urban. BASES followed a few years later, created by Lynn Y.S. Lin at Burke Marketing Research in 1977, and it’s now owned by Nielsen, where it remains one of the most widely used STM systems in the packaged goods industry today.
Both systems were built around the same basic insight: if we can accurately measure trial and repeat purchase intentions in a small, controlled sample, we can mathematically project those numbers up to the size of a real market.
How the Process Actually Works
Most STM systems run in two stages, and it’s worth walking through them because the logic explains why the method works.
Stage One: Measuring Trial
Researchers recruit people who fit the target market for the new product. These participants are shown advertising or a product concept, similar to what they’d encounter in the real world, and then given the chance to “purchase” the product in a simulated shopping environment alongside its competitors. This tells researchers how likely someone is to try the product in the first place, and it also reveals how the new product’s shelf presence, pricing, and packaging perform next to what’s already out there.
Stage Two: Measuring Repeat Purchase
Participants who choose the product take it home and actually use it. After a set period, researchers follow up and ask whether they’d buy it again, how satisfied they were, and how it compares to what they used before. This repeat-purchase data matters enormously, because a product that gets a lot of curious first-time buyers but no repeat business is not a viable long-term product. Think about it this way: a mediocre product with clever packaging might get plenty of people to try it once. Whether they buy it a second time tells us whether the product itself can carry a business.
Combining the trial rate and the repeat rate, along with assumptions about distribution and marketing spend, lets the model estimate first-year and second-year sales volume. Good STM systems also generate diagnostic feedback along the way, flagging whether the concept is seen as new, believable, and appealing, and whether the price feels fair relative to the value being offered.
A Practical Example
Let’s say a mid-sized snack food company has developed a new line of protein-based chips and is deciding whether it’s worth the cost of a full national launch, with all the manufacturing, distribution, and advertising spend that involves.
Rather than shipping the product to test cities and waiting months to see what happens, the company runs an STM. They recruit a sample of consumers who already buy snack chips regularly, show them the packaging and a short ad concept, and let them choose between the new protein chips and the usual competitors in a simulated store aisle. A portion of participants pick up the new product. Those people take a bag home, and a few weeks later they’re asked whether they’d buy it again and how it compares to their usual snack.
Suppose the trial rate looks strong, plenty of people are curious enough to pick it up once, but the repeat rate is weak. That’s a useful and fairly common outcome. It tells the marketing team that the concept and packaging are doing their job, but something about the actual eating experience, maybe the taste or texture, isn’t holding up. That’s a much cheaper problem to discover through an STM than after a national launch, when the company has already spent heavily on manufacturing and advertising for a product people won’t buy twice.
Why This Matters to Brand Managers
For a brand manager or product manager, STM data is often the deciding factor in whether a product moves forward, gets sent back for reformulation, or gets killed before it ever reaches a shelf. That’s a lot of weight to put on a research method, so it’s worth understanding both what it’s good at and where it falls short.
STM is valuable because it’s relatively fast (results in weeks rather than the six to twelve months a live test market can take), far cheaper than a full regional rollout, and confidential. Competitors don’t get to see the product sitting in stores months before a national launch, which matters a great deal in categories where copying a successful new product is easy and fast.
It also forces a level of discipline. A concept that sounds great in a boardroom sometimes performs poorly once real consumers are asked to choose it over what they already buy, and STM catches that before it becomes an expensive lesson.
Limitations Worth Knowing
STM isn’t perfect, and it’s worth being honest about where it can mislead a management team.
Because it happens in an artificial setting, it can’t fully capture how a product will perform once it’s competing for real shelf space, real distribution deals, and real advertising budgets against competitors who may respond aggressively to the launch. A simulated environment also can’t easily account for word of mouth, social media buzz, or a competitor’s reaction, all of which can meaningfully change how a real launch plays out.
There’s also a dependency on the quality of the underlying model and the sample. If the recruited participants don’t genuinely represent the target market, or if the model’s assumptions about distribution and awareness don’t match reality, the sales forecast can be badly off, even though the process looks scientific and precise.
Because of this, most companies treat STM results as an important input rather than the final word. A strong STM result is usually followed by a smaller, real-world test before a full national commitment, and a weak result is usually enough on its own to send a product back to development.
Bringing It Together
Simulated market testing exists because launching a new product for real is risky and expensive, and traditional test marketing carries its own costs and exposure. By measuring trial and repeat purchase in a controlled setting and feeding that into a forecasting model, companies get a reasonably reliable early read on whether a new product is worth the investment, and where it might need work before it reaches real customers.
Key Points to Take Away
- Simulated market testing (STM) forecasts how a new product will sell by measuring trial and repeat purchase intentions in a controlled research setting, rather than launching it in real test markets.
- The two best-known systems, Assessor (MIT Sloan, 1973) and BASES (Burke Marketing Research, 1977, now owned by Nielsen), were among the first commercial STM tools and remain influential today.
- STM typically runs in two stages: measuring trial through a simulated shopping exercise, then measuring repeat purchase after a home-use period.
- The method is faster, cheaper, and more confidential than a traditional test market, since competitors don’t see the product on real shelves before a national launch.
- Its main limitation is that an artificial setting can’t fully capture competitive response, word of mouth, or real-world distribution, so results are usually treated as one input rather than a guarantee.
