What is test marketing, and can you run a test market without a launch?
TL;DR
- Classic test market: real product, real stores, a few regions, months of sales data, and rivals can watch.
- Simulated test market: mock shelf or store, purchase choices plus follow-up, a model that projects volume.
- Pre-test-market models go back to Alvin Silk and Glen Urban's 1978 paper in the Journal of Marketing Research.
- Neon Blue runs the simulated step on a simulated audience of your buyers, so only the strongest plan goes to a live regional test. Run your first study free
Neon Blue builds simulations of human behavior. Test marketing means selling a new product, or a new price or campaign, in a limited area or to a limited group before rolling it out everywhere. A classic test market is a real launch in a few cities, run for months, with sales tracked against control areas. A simulated test market skips the real launch: buyers see the product and its competitors in a mock shopping setting, make choices, and a model projects trial and repeat. Classic tests give you real sales and alert competitors. Simulated tests are faster and private, and they forecast rather than observe.
What are the types of test marketing?
| Type | What happens | Typical time | What you learn | Main drawback |
|---|---|---|---|---|
| Standard test market | Full launch in a few regions | Months to a year | Real trial, repeat and sales | Slow, costly, visible to rivals |
| Controlled test market | Product placed in a set of stores by a research partner | Months | Sales under controlled shelf and price | Fewer stores, still visible |
| Simulated test market | Buyers shop a mock shelf, then use and rate the product | Weeks | Projected trial, repeat and volume | A forecast, not observed sales |
| Online or digital test | Launch to a slice of users or one region online | Weeks to months | Real conversion and retention | Only works for digital channels |
Teams often stack them: a simulated test to pick the plan, then a smaller live test to check it in market.
Where did simulated test markets come from?
In 1978, Alvin Silk and Glen Urban published a model for judging new packaged goods before any test market, in the Journal of Marketing Research. Their model, called ASSESSOR, combined a mock shopping trip, where people chose among the new product and existing brands, with a later follow-up after they had used the product at home. From those two steps it estimated trial and repeat and projected market share before any real launch. Most simulated test markets since have kept that two-part structure: a choice that estimates trial, and a use-and-rate step that estimates repeat.
The 1969 diffusion model that underpins many launch forecasts is covered in forecasting demand for a new product.
What questions does a simulated test market ask?
- Here is a shelf with [category] products and their prices. Which one would you buy on this trip? [Includes the new product]
- If you didn't choose the new product, why not?
- After trying it, how likely would you be to buy it again at [price]?
- What would you buy less of if you bought this regularly?
- How often would you buy it? [Every week, every month, occasionally]
Questions 1 and 3 feed the trial and repeat estimates. Question 4 separates new volume from cannibalized volume, a point also covered in purchase intent.
What does a simulated test market look like on a simulated audience?
Hypothetical: a beverage company wants to know whether a new low-sugar iced tea should launch at a premium price or at parity with its current line.
- Audience: US adults who bought ready-to-drink tea or juice in the past month, split by current brand buyers and rival buyers, by age band, and by convenience-store versus grocery shoppers.
- Design: two cells, one per price, each shopping a 16-product cooler image on a simulated audience, then a product description and repeat-intent read.
- Readout: projected trial and repeat for each price, source of volume, and the reasons people passed.
Example output layout (illustrative, not a real result)
| Cell | Chose new tea on shelf | Would buy again | Would replace own brand | Would replace rival | Main reason to pass (coded) |
|---|---|---|---|---|---|
| Premium price | __% | __% | __% | __% | Coded theme |
| Parity price | __% | __% | __% | __% | Coded theme |
If the premium cell wins trial nearly as well but takes more of its volume from rivals, the premium plan goes to a live regional test. The trade-off feeds the go or no-go call.
Can you run a test market without launching?
You can test the plan without launching the product. A simulated test market tells you how buyers would choose among the products they would actually see, at the prices you plan, before you print packaging or book shelf space. What it can't tell you is how distribution, stockouts and rival responses play out; only a live test shows that. For ideas that are earlier than a full plan, start with testing a product idea and concept test questions. For price cells, the price points can come from a Van Westendorp range. And for digital products, a staged live rollout is often cheap enough to run after the simulation, as in the growth team workflow.
Questions about test marketing
What is an example of test marketing?
A drinks brand launching a new flavor in two regional markets for six months, with matched regions as controls, before deciding on a national rollout.
What is a simulated test market?
A pre-launch study where buyers choose among the new product and its competitors in a mock store, then rate it after use, and a model projects trial, repeat and volume.
Why do companies skip classic test markets?
They take months, cost a lot, and let competitors see and respond to the product before the national launch.
How does Neon Blue run a simulated test market?
Neon Blue fields the shelf choice and repeat questions on a simulated audience of your buyers, in as many price or pack cells as you need, and marks every projection as simulated. Start a free first study.