The Unifying Concept Test
Before adding a service to your app, check it fits the one mental model customers already hold of you
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 4
- Confidence
- 95%
Kevin Aluwi's rule for multi-product apps: a new service only compounds if it slots into the single unifying concept customers already carry in their heads about your business. At Gojek that concept was the driver — every service a person on a motorbike could plausibly perform cross-sold itself. Services outside that concept had to be re-sold from scratch, killing the promised super-app economics of lower CAC and higher retention.
Origin
Developed at Gojek by co-founder Kevin Aluwi through two UX-research findings: only 30-40% of customers knew Gojek offered mobile top-up despite it sitting on the home screen, and customers assumed their motorbike driver would personally deliver the massage service. Aluwi frames it as a corrective to the 'super app' narrative popular in VC and consultant strategy decks.
Core principles
- 01Customers hold ONE concept of your business, not a list of your services.
- 02Services inside the concept cross-sell for free; services outside it require a full re-acquisition spend.
- 03Home-screen placement is not distribution — awareness is not the same as shelf space.
- 04Super-app benefits (lower CAC, higher attach, cross-sell) are claims to be tested, not defaults.
- 05Every off-concept service also taxes design: you end up with a grid menu because nothing relates.
How to run it
- 1
Name the concept customers actually hold
Run UX research asking customers to describe what your product is, in their words. Do not accept the internal answer. Gojek's internal answer was 'on-demand super app'; the customer answer was 'the driver'.
Pro tip The tell is what customers assume about a service they have never used — Gojek learned its concept from customers asking whether the motorbike driver would give the massage.
- 2
Test the candidate service against the concept
Ask whether a customer holding your concept would predict this service exists without being told. Package delivery, food delivery and grocery all pass 'a person on a motorbike'. Mobile top-up, cinema tickets and loans do not.
Watch out Relevance is not fit. Mobile top-up was relevant to 95%+ of Gojek's users and still failed the concept test.
- 3
Audit awareness before you blame the funnel
For an underperforming in-app service, ask a sample of users whether they know it exists at all. If a large share do not, the problem is conceptual placement, not conversion or UI.
Pro tip Gojek's number was only 30-40% aware — for a button on the home page. Awareness gaps this big mean the service is invisible because it is unthinkable.
- 4
Price in the re-selling cost, or don't ship it
If the service fails the test but you still want it, budget explicitly for the advertising and customer education needed to re-establish it as a separate idea — and add that to your CAC model. If that spend kills the business case, kill the service.
Watch out Off-concept services also constrain your design forever — every new one pushes you toward the generic grid-of-icons home screen.
In the wild
Southeast Asia is overwhelmingly prepaid, so buying phone minutes and data was a product 95%+ of Gojek's customers needed. It sat as one of six buttons on the home page. Engagement was still low, so the UX-research team asked customers directly whether they knew they could top up on the Gojek app.
→ Only 30-40% of customers knew the product existed. The team concluded the issue was not placement but that top-up did not fit the driver-shaped concept customers held of Gojek.
Gojek launched on-demand massage: order a masseuse to your home. A recurring customer question was whether the Gojek driver would come into their house and give them the massage. Internally the question seemed absurd, but it exposed the mental model — Gojek was 'the app of driver-related services'.
→ Massage was shut down a few years later. Aluwi cites it as the clearest proof that a service outside the unifying concept confuses rather than compounds.
Common mistakes
Treating engagement and eyeballs as a licence to add services
Aluwi: it is not as simple as having lots of engagement and then assuming a super app makes sense for customers. Attention on one service does not transfer to an unrelated second one.
Believing the strategy-deck economics
Lower CAC, higher attach rates, easy cross-sell and upsell are what super apps sound like on a slide. In practice the off-concept services raise CAC because you must buy awareness and education all over again.
Assuming placement equals discovery
Gojek put top-up on the home page and still had majority-unaware customers. If a service does not fit the concept, no amount of prime real estate makes it visible.
Is it for you?
Best for
Founders and PMs at consumer apps under pressure to expand into a multi-service or 'super app' surface, deciding which adjacent services to launch
Not ideal for
Single-product companies still hunting product-market fit, or B2B tools where the buyer's mental model is set by procurement rather than habitual use
From the transcript
“there kind of needs to be a unifying concept across all of your services within the app for your users to be able to think…”
“oh is the driver gonna come into my house and give me a massage”
“that whole Nirvana of like high of lower CAC higher retention uh that are on these great strategy decks often don't pan out”
From the episode
Taxi mafias, cash vaults, and 100% MoM growth: The story behind Southeast Asia’s biggest startup
Kevin Aluwi (Gojek)