Concept testing sounds fairly straightforward; you have an idea, you show it to some potential customers and you ask what they think. If the response is positive, then surely you’re onto a winner? However, to get this research right, there’s a fair bit more to consider.
We’ve worked on concept tests for everything from new products and services to pricing models, features, messaging, and propositions. Sometimes clients come to us with one fairly developed idea, and other times there might be four or five options on the table and a team that’s struggling to decide which one deserves further investment.
Either way, the aim is usually similar; to understand how an idea is likely to land with the people it’s designed for, while there’s still time to use the findings to impact what happens next.
When should you test a concept?
Ideally you want to test a concept before you’ve become too attached to it or it becomes too hard to change. Concept testing tends to be most useful when an idea is developed enough for customers to understand what they’re being asked to consider, but there’s still some flexibility internally to make amendments.
Test too early and people may struggle to give meaningful feedback because the idea is too vague. Leave it too late and you risk discovering something important after significant time and money have already been spent.
You don’t necessarily need a finished product, polished creative, or final pricing. Instead, a clear description, visual, mock-up, or prototype can often be sufficient to give potential customers enough to evaluate.
What should you actually put in front of customers?
This is where being disciplined really helps. It’s tempting to want to show everyone everything at once or put every single option on the table for consideration. But this will likely be too overwhelming and result in poor quality data. Limiting to a maximum of five concepts is usually wise.
And if you are comparing several concepts, they need to be presented consistently. One shouldn’t have beautiful finished creative while another is represented by three lines of text. Otherwise, you risk measuring the quality of the presentation rather than the strength of the idea.
It’s also worth thinking carefully about how much information customers would realistically have in the real world. Giving someone several paragraphs explaining every benefit might make a proposition easier to understand in research, but would they usually read all of that when choosing between products?
Essentially, the closer the test is to the real decision, the more useful the results tend to be.
Go further than “Do you like it?”
Positive feedback is always nice to hear, but it will only tell you so much. Someone can like an idea without wanting to buy it, and they can describe a feature as useful without being prepared to pay more for it.
A good concept test looks a little deeper; Is the proposition clear? Is it relevant? Does it offer something different? What would encourage someone to choose it? What’s putting them off? How does it compare with other options?
Depending on the project, you might also want to explore willingness to pay, purchase intent, which features matter most, or how different elements of the proposition influence choice.
Context matters
One mistake that’s easy to make is testing a concept in a vacuum. Customers rarely make decisions that way as usually they’re comparing your idea with competitors, what they already use, other ways of solving the same problem, or simply doing nothing at all.
An idea might perform very well when it’s the only thing on the screen. However, if you put it next to three established alternatives at realistic prices you may get a very different result.
That doesn’t mean every concept test needs to recreate the entire buying journey, but it is important to think about the context customers will eventually encounter and including enough of it to make the research meaningful.
Look at the differences between customers
An overall score can hide some of the most useful findings; a concept that gets a fairly average response across the whole sample might perform extremely well among a particular customer group or segment. Another might be popular with existing customers but do very little to attract new ones.
Instead of looking at the results at purely a total market level, at Clusters we always recommend to break the sample down into either customer segments, or different target groups. This way you can gain a better understanding of who the proposition resonates best with and why.
Understanding those differences can influence everything from the proposition itself to pricing, targeting, and how you eventually take it to market.
Good concept testing should help you decide what to do next
The best concept testing projects won’t just finish with a score and a recommendation to go ahead. They should help explain what’s working, what’s acting as a barrier, and what could be improved. Sometimes the answer is to move forward with one concept, and other times it’s to take the strongest elements from several ideas and develop something new. Occasionally, the findings can suggest going back to the drawing board.
All of these outcomes are useful and can save you a lot of time and money in the long run.
Ultimately, a good concept test should leave you with a clear idea of what to do next. If it only tells you that ‘72% of people liked your idea’, then you probably haven’t asked the right questions.