In 2020, a Trinidadian business owner called me in a panic. Her restaurant had been closed for three weeks. She needed "a website with online ordering" by the end of the month. Budget was tight. Timeline was tighter. She had already spoken to a developer who said they could have something live in two weeks.
I asked her one question before we went any further: had she spoken to any of her customers about how they wanted to order?
She had not. There was no time for that.
So we skipped it. We built the ordering system. It went live. And for about six months, it worked well enough. Then it didn't. Customers went back to calling in their orders. The system sat mostly unused. The developer moved on. The business owner was left with a subscription she wasn't sure she needed and a workflow nobody had actually asked for.
She is not the exception. She is the pattern.
1. COVID Didn't Create a Digital Culture. It Created a Digital Panic.
When the pandemic hit the Caribbean, businesses that had been comfortable ignoring digital for years suddenly needed a website, a WhatsApp Business account, and a checkout flow by Monday. The urgency was real. The pressure was legitimate. But the speed at which the region digitised in 2020 and 2021 compressed years of decisions into weeks, and most of those decisions skipped the one step that determines whether digital products actually work: understanding the people who were supposed to use them.
Banks built online account-opening flows without researching how Caribbean customers actually think about trust and financial institutions. Retailers launched e-commerce platforms without asking whether their customers would enter card details on a site they'd never heard of two weeks ago. Government services moved online without accounting for the significant portion of users accessing them on mobile data with inconsistent connectivity.
The result was a wave of MVPs that were minimum and viable only in the narrowest sense. They existed. They were technically functional. But they were designed around assumptions imported wholesale from markets with different users, different infrastructure, and different relationships with digital technology.
Many of those products are gone now. The teams built to maintain them were cut. The businesses that survived pulled back to what they knew.
The question the region is left with is the one that should have come first: what does a digital product actually have to do to work for Caribbean people?
2. The Assumptions We Skipped Are Now Our Technical Debt
Every product embeds assumptions about its users. When you skip research, you don't eliminate those assumptions. You just borrow them from whoever built the template you're working from.
In the Caribbean, that usually means a Silicon Valley template, a UK government service pattern, or an e-commerce framework built for markets where trust in digital payments is already established. These aren't bad starting points. But they carry embedded logic that doesn't always hold here.
Consider what "trust" means in a Caribbean digital context. For many users across the region, a business's digital presence is secondary to its reputation within a community. A user in Port of Spain deciding whether to complete a purchase on an unfamiliar platform is not making the same calculation as a user in San Francisco doing the same thing. The informal referral network is still the primary credibility signal. The WhatsApp group recommendation still outweighs the Google review. The colleague who says "I used them and it worked" still does more conversion work than any homepage hero copy.
When you build without researching this, you end up designing for a user who trusts the interface by default. That user does not exist here at the same rate. So your checkout abandonment is high, your completion rates are low, and your analytics tell you there's a UX problem when there's actually a context problem.
The fix is not a redesign. It's the research that should have happened before the first wireframe.
3. The Products That Survived Did One Thing Differently
Not everything built during that period failed. Some Caribbean digital products from 2020 and 2021 are still running, still growing, and still earning genuine user loyalty. Looking across those that held, a pattern appears: the teams behind them stayed close to users even when it was inconvenient.
Not through formal research programmes. Most of them didn't have the budget or the process for that. But they watched. They asked. They had founders or designers who were embedded enough in their communities to notice when something wasn't landing, and responsive enough to change it before the gap became a chasm.
One fintech product I know of built its entire onboarding sequence around a single insight surfaced in informal conversations with potential users: people didn't trust the app until they understood exactly where their money went if something went wrong. That is not a flow any standard onboarding template accounts for. It required someone to sit with a Trinidadian user, in a Trinidadian context, and actually listen to what was making them hesitate.
They rebuilt the onboarding to answer that question before asking for any personal information. Adoption improved. The product survived.
That's not a research methodology. It's not a design sprint. It's the oldest practice in this field applied to a specific place: pay attention to the people you're designing for.
The businesses that treated digital as a checklist item to be completed under pressure are largely gone. The ones that treated it as a relationship with their users are still here.
What the Region Actually Owes Its Users
The Caribbean digital landscape is not starting from zero. The panic builds of 2020 and 2021 left behind something useful alongside the wreckage: proof that the region can move quickly, that users are ready to adopt digital tools when those tools are worth adopting, and that there is genuine appetite for products that take Caribbean life seriously as a design input.
What the region owes the people who showed up and tried those early products is better. Not faster. Not cheaper. Better designed, in the specific sense that matters: designed by people who asked the right questions first, who did not assume that a pattern that works in London or Lagos would automatically work in Laventille.
The wrong-thing-fast era has passed. What comes next depends on whether the designers, product teams, and business leaders who lived through it are willing to treat the failure as data rather than bad luck.
The research is overdue. But it is not too late to do it properly.