Don’t choose your product hero in a meeting room before launch. Put the whole menu in market, give every item the same marketing weight, and let real sales and real engagement choose for you. When we did this for Chickster, the winner was clear within three weeks.
A product hero is the one item a brand leads with. It is the face of the ads, the first recommendation at the counter, and the product people come back for and tell their friends about. For a new brand, it is also where marketing money compounds fastest. Pick the wrong one and every rupiah after that has to work harder than it should.
How do brands usually pick a product hero?
Traditionally, the hero is chosen before launch through focus group discussions. A small group tastes the options, talks about them, and the brand backs the favourite. It feels rigorous. There is a moderator, a report, and quotes that look good in a deck.
The trouble is that the decision gets made before a single real customer has paid for anything.
Why can’t a focus group decide it for you?
Focus groups don’t lie on purpose. They answer a different question from the one that matters. The question that matters is what people buy, and buy again, on an ordinary day. A focus group tells you what people say they like, in a room, in front of others. The gap between the two is where the bias lives.
- Saying is not paying. “I would order this” costs nothing. Paying for it, with a familiar item sitting right next to it, is a different decision.
- The loudest voice wins. One confident participant can pull a small group toward their favourite.
- People are polite. Opinions get softer when the people who made the food are listening.
- The setting is not real. Tasting on a quiet afternoon is not ordering in a lunch rush with a queue behind you.
- The sample is small. A handful of people stands in for a whole city.
None of this makes focus groups useless. It makes them a poor final judge.
What did we do for Chickster?
Chickster is a quick-service chicken restaurant that opened in 2026. As a new brand it had no sales history to lean on, which is exactly the moment brands usually reach for a focus group.
We chose to run an A/B test at product level instead. Every item on the menu was marketed with the same weight, so each one had the same chance to earn attention and orders.
Within three weeks, sales showed a clear winner: Ayam Nori. It had the best content engagement and the best sales at the same time. That made it the product hero, and the item we now double down on in marketing.
The test was clean for one reason. Taste and price were already fixed. The only thing that differed between the items was the product itself, so the result told us exactly which product people chose. Not which recipe a panel preferred, and not which price sounded fair on the day.
Why does a product-level test reduce bias?
It measures what people do instead of what they say, and it measures it in the real place where the choice gets made.
| Focus group before launch | Product-level test in market | |
|---|---|---|
| What it measures | What people say they prefer | What people actually order |
| Setting | A room, a moderator, a small group | The restaurant, the feed, an ordinary day |
| Price | Imagined | Real and already fixed |
| Who decides | A handful of participants | Every paying customer |
| Signals | Opinions and quotes | Sales and content engagement, side by side |
| Output | A favourite | A winner you can double down on |
How do you run a product-level test?
- Lock taste and price first. If the recipe or the price is still moving, you are testing several things at once, and the result will not tell you which one mattered.
- Give every candidate equal weight. Same content format, same posting rhythm, same paid support, same prominence. Equal weight is the whole experiment. If one item gets a better video, you are testing the video, not the product.
- Track two signals for every item. Sales show what people choose. Content engagement shows what people want to see and share. You want a hero that wins both.
- Read the numbers every week. Sales and social data now arrive fast enough for weekly reads, so the decision takes weeks rather than a quarter.
- Decide when the winner holds. The sign to look for is the same item leading week after week, on both signals. Chickster reached that point in three weeks. A brand with fewer daily transactions may need longer.
- Double down. Put the hero at the front of your ads, your creative and your recommendations. Keep the rest of the menu visible, and keep reading the numbers.
What if engagement and sales disagree?
For Chickster they agreed, which made the call easy. When they don’t, read them as two different jobs.
- High engagement, low sales. The item is a great hook but not yet a great seller. Use it to earn attention, and let another item close the sale.
- High sales, low engagement. A quiet favourite. People buy it and come back, but it may not be the face that stops the scroll. Protect it, feature it where people order, and test new creative before making it the hero.
- High on both. That is your hero.
When you have to choose, sales carry more weight. A hero has to sell, not just get liked.
When is a focus group still useful?
Before taste and price are settled. A focus group is a good tool for exploring options, killing weak ideas early, and understanding why people react the way they do. Use it to shape the options. Then let the market choose between them.
What changes in a real-time world?
The old reason to decide before launch was that real sales data took too long to arrive. That is no longer true. Sales, stock and social engagement can be read weekly, even daily, so a brand can launch with a full menu and let the first weeks of trading make the call.
It is the same closed loop we describe in Closed-loop marketing: what the data says this week changes what runs next week. It is also why we build dashboards that put sales and marketing numbers in one place, which is exactly what Cirya is for.
FAQ
How long should a product-level test run?
Until the same item leads on both sales and engagement across more than one reading. For Chickster that took three weeks. Lower traffic usually means a longer test.
Does equal weight mean equal budget?
It means equal treatment: the same budget, format, frequency and prominence for every item. Anything uneven becomes part of the result.
Isn’t it risky to launch without a hero?
You still launch with a full menu. The bigger risk is spending the launch budget behind the wrong item because a small group liked it.
Does this only work for restaurants?
No. Any brand with several products and sales data that arrives quickly can run it, whether the candidates are flavours, variants, bundles or app features.
Launching a menu or a new product line? Send us the brief. Three lines is enough.
