Skip to main content
VantaSoft
Back to Insights

Why 'Just Make It Like Uber' Is an Expensive Miscommunication

When you say 'like Uber,' your development team hears 'build Uber.' Budget overruns start in that gap, and one conversation before kickoff prevents them.

Working with developers 6 min read

Picture a kickoff meeting. You have spent weeks thinking about your product, and you know exactly what you want. Then you say the sentence that ends up costing you far more than you planned:

"We want something like Uber, but for dog walkers."

Everyone nods and the meeting ends. From that point on, you and your technical team are working toward two completely different products.

What you meant vs. what they heard

When you said "like Uber," you meant a clean booking flow. A customer can request a service, see who is available and pay through the app. That is a reasonable product, well within reach for a funded startup or a growing SMB, and maybe a few months of focused development.

When your technical team heard "like Uber," they started scoping real-time GPS tracking with sub-second location updates. They added a matching algorithm that optimizes for distance and availability at the same time, and surge pricing logic that adjusts rates based on demand. Then came a dual-sided rating system, push notifications and in-app messaging. Payment processing with split payouts to contractors followed. So did background checks, driver onboarding flows, admin dashboards, analytics pipelines and a support ticket system. That list describes Uber itself. A large engineering organization has spent more than a decade building it, and it is far more than a few months of work. The gap between what you meant and what they heard is where budget overruns begin.

Why this keeps happening

Pointing at a reference app feels like an easier way to communicate. You point at something that already exists and say "like that," which feels precise and efficient. You assume the other person will pick out the relevant parts and ignore the rest.

A technical team cannot do that, because it has no way of knowing which 10% of Uber you care about. Its job is to turn what you said into a system that works. If you said "like Uber," it will scope something that does what Uber does. That is the team following your instruction, and it has nothing to do with padding the estimate.

Better listening will not fix this, because the problem is structural. The person describing the product thinks in outcomes: "customers book a service easily." The person building the product thinks in systems: "what does the booking system need to handle, and what happens when it fails?" Sometimes the only bridge between those two views is a product with hundreds of features. Then the person building it has no way to know which features you consider essential and which ones you have never thought about.

What the reference quietly adds

Each of those implied features is a project of its own, and most products do not need them on day one.

Real-time GPS tracking with live map updates needs websocket infrastructure, a geolocation pipeline, map rendering and battery-efficient location polling on mobile. Most founders who say "like Uber" do not need real-time tracking. They need a status indicator that says "your dog walker is on the way."

A matching algorithm weighs signals such as distance, estimated arrival time, ratings and current demand. If you only need to assign the closest available person, that is a database query rather than a matching algorithm.

Surge pricing that responds to real-time demand requires a data pipeline, a pricing model and careful UX so customers do not feel cheated. Most early-stage products do not need dynamic pricing at all, because a price list does the job.

Split payment processing with contractor payouts is easier than it used to be, since Stripe Connect or a similar service handles the plumbing. But compliance, tax reporting and edge cases such as refunds, disputes and delayed captures add real complexity. If you pay your team a salary, this entire workstream disappears.

Together, these features can turn a few months of work into a far larger build. A reference implied every one of them without a single requirement asking for it.

How to brief a technical team instead

You do not have to stop referencing other products. References are useful visual shorthand, as long as a reference never stands in for a requirement. Four habits help:

  • Describe the outcome rather than the product. "A customer should be able to book a dog walker within 30 seconds and know when they are arriving" is a requirement. "Like Uber" is not.
  • Name what you do not need. When you are scoping, exclusions are worth more than inclusions. "We do not need real-time GPS. A text notification when the walker starts and finishes is enough" saves your team weeks of work they would otherwise have assumed was required.
  • Walk through the workflow instead of showing a screen. Describe what happens step by step from the customer's side: "They open the app, see available walkers this week, pick a time slot, enter payment, and get a confirmation email." Your team can estimate those five steps. It cannot estimate "like Uber."
  • Ask what assumptions they are making. After the brief, ask your technical team to list what they think you want, and expect to find things they inferred that you never intended. Catching those assumptions costs nothing before development starts and a great deal after three months of building. When the quote arrives, look for the same assumptions and exclusions in writing; they are among the first things to check when you evaluate a software development proposal.

The conversation that should have happened first

Better briefing helps, but the larger issue is who is in the room when the product is being defined.

A non-technical founder briefing a development team is translating a business vision into a technical specification alone. Most founders do not have that skill, and there is no reason they should. The most expensive mistakes happen in that translation rather than in the development itself.

A technical partner catches the "like Uber" trap before it reaches a scope document. That means someone with senior technical judgment who understands both the business goal and the system that has to support it. They ask "what are you trying to achieve?" before they ask "what do you want us to build?" The answer to the first question almost always leads to a simpler, cheaper and faster product than the one you would have gotten by pointing at a billion-dollar app and saying "like that."

VantaSoft has that conversation with every client before anyone writes a line of code. Your next product brief may say "like Uber," "like Airbnb" or "like Shopify." If nobody in the room is asking what you mean by it, the most expensive part of your project has already started.

Five quick questions about where your team's time goes. You'll get the one agent we'd start with, and what it would take off your plate.

Take the quiz

Tell us whatyou’re building.

Senior engineers and fractional CTO leadership, onshore in Irvine. It starts with a conversation.