How do CPG brands test new products before launch?
TL;DR
- Five common gates: concept, product, pack, price and claims, then a volume forecast for the retailer pitch.
- Each gate has one job. Don't ask a concept test to set the price or a pack test to pick the flavor.
- How many variants to launch is its own question: a famous 2000 jam study found more choice cut buying, but a 2010 meta-analysis of 50 experiments found the effect averages out near zero.
- Neon Blue runs the early gates on a simulated audience of your shoppers, so live budget goes to the finalists. Run your first study free
Neon Blue builds simulations of human behavior. Consumer packaged goods brands test new products through a series of stage gates, each with its own study: a concept test to choose the idea, a product or flavor test, a pack test, a price test, a claims test, and often a simulated test market before committing to retailers. Each gate kills or reshapes ideas before the next, more expensive step. The hard part is that the calendar is set by retailer line reviews, so research has to fit the window. Running early gates on a simulated audience of shoppers lets more ideas through the funnel in the time available.
What are the stage gates for a new CPG product?
| Gate | Decision | Study | Guide |
|---|---|---|---|
| Concept | Which idea to develop | Monadic concept read against a control | Concept testing |
| Product | Which formula or flavor | Blind sensory or described-product test | Live product test with real shoppers |
| Pack | Which design | Shelf standout, findability, communication | Packaging testing |
| Price | Price point and pack size | Price range and demand at set points | Van Westendorp, then a demand ladder |
| Claims | Which claim leads on pack | Best-worst ranking, believability | Claims testing |
| Volume | Is it big enough for the retailer | Trial and repeat build-up | Demand forecasting |
The product gate is the one that needs physical product in real hands. Every other gate can start on a simulated audience and move to live research for the finalist.
How many flavors or variants should you launch?
In 2000, Sheena Iyengar and Mark Lepper published a study in the Journal of Personality and Social Psychology. At a grocery store tasting table, shoppers saw either 6 or 24 jams. More people stopped at the larger display, but 30% of those who stopped at the small one bought a jar, against 3% at the large one. The study became the standard argument for short lines.
A decade later, Benjamin Scheibehenne, Rainer Greifeneder and Peter Todd pooled 50 experiments on choice overload in the Journal of Consumer Research (2010) and found the average effect close to zero, with large swings depending on the setting. For a CPG team, that means line size is an empirical question for each category, not a rule. Test it: show shoppers a shelf with three, five and eight variants and see which one sells the most total units of your brand, the same shelf setup used in a simulated test market.
What does a CPG launch study look like on a simulated audience?
Hypothetical: a snack brand plans a protein chip line and has to choose between launching with three, five or eight flavors for a retailer review.
- Audience: US adults who bought salty snacks in the past month, split by people who buy protein or "better for you" snacks, brand buyers versus rival buyers, and households with and without kids. Shopper mix set using building a synthetic audience.
- Questions to field:
- Here is the chip aisle. Which product would you buy on this trip? [Shelf with three, five or eight flavors of the new line, by cell]
- If you chose one of the new chips, which flavor, and how many bags?
- Which of these flavors would you never buy?
- If this line were in your store, what would you buy less of?
- How likely would you be to buy it again after trying it at [price]?
- Design: three shelf cells on a simulated audience, the same 24-product shelf apart from the new line.
Example output layout (illustrative, not a real result)
| Cell | Chose the new line | Bags per buyer | Source of volume (own brand / rivals) | Flavors rarely chosen | Repeat intent |
|---|---|---|---|---|---|
| 3 flavors | __% | __ | __% / __% | List | __% |
| 5 flavors | __% | __ | __% / __% | List | __% |
| 8 flavors | __% | __ | __% / __% | List | __% |
If five flavors pulls the most total units with the least taken from the brand's own core chips, that is the line the retailer pitch is built on. The volume estimate feeds the go or no-go call.
How do CPG teams fit research into a retailer calendar?
- Work back from the line review. Know the date the sell-in story is due and schedule gates from there.
- Run early gates in parallel. Concept, claims and price can run on the same simulated audience in one cycle, so the product gate gets the most time.
- Save live research for the finalist. A live quantitative test and a home-use test on the chosen product give the retailer story real-buyer evidence.
- Plan the post-launch read. Track the new line with a brand tracking wave after it ships.
Food and beverage teams can see category-specific study designs in food and beverage product research.
Questions about CPG product testing
How do companies test new products before launch?
Most run a sequence: concept test, product test, pack test, price test and claims test, then a volume forecast or test market. Each step narrows the options for the next.
What is the difference between concept testing and product testing in CPG?
A concept test reads reactions to a description of the idea. A product test has people use or taste the real product. Concept comes first; product needs physical samples.
How long does CPG product testing take?
The full sequence usually spans months because of product development. Running concept, claims and price gates on a simulated audience shortens the research part.
How does Neon Blue help CPG brands?
Neon Blue runs concept, pack, price, claims and shelf studies on a simulated audience of your shoppers by segment, marks every result as simulated, and points live research at the finalist. Start a free first study.