Meal4U

The Founders Flew to Ukraine

Meal4U is a Danish food platform. Restaurants, configurable menus, mobile ordering. The concept was clear. The execution wasn’t.

The previous development team didn’t deliver. Features were unfinished or built wrong. The product existed in partially working code, but it wasn’t what the business needed. The founders had run out of confidence in the development process.

They flew from Denmark to Ukraine. Not to start a partnership. To see, in person, whether there was a path forward at all.

The First Week Was Scope, Not Code

We didn’t open the editor for seven days.

With the founders in the room, we worked through the product feature by feature. What works. What doesn’t. What has real business value. What was built because someone thought it should be there. What can come out without hurting the product.

We re-estimated scope roughly ten times in that week. Each pass followed the same logic: take the feature list, cut 30-50%, check whether the product still makes business sense, add back only the items the business model can’t survive without.

The final scope was about half of what the previous team had attempted. It was the right half.

This is where most failed software projects get fixed, and it’s where most failed software projects fail in the first place. The problem is rarely the code. The problem is a feature list that was always too large for the budget, with no one willing to remove items.

What We Built

The Configurable Menu Engine

This is where the engineering became interesting. Meal4U isn’t a delivery app where you pick a dish and it arrives. Every dish is a configurable object.

A single menu item has ingredients that can be added or removed, modifiers that change the base (size, preparation, extras), dependencies between options (selecting ingredient A unlocks modifier B), discount rules that fire on specific combinations, and a price that recalculates in real time as the user makes changes.

Multiply that by a full restaurant menu. Then put it in a cart where the user has multiple configured items, each with its own modifier set, each with its own discount logic applied. The cart total has to update instantly on every change.

Our iOS lead describes it this way: every dish isn’t a product but a configurable object with rules, dependencies, and personalization, and all of it lives in real time on the client.

Making it fast was the work. The combinatorial space of ingredients, modifiers, discounts, and pricing is large. A naive implementation recalculates everything on every user input and the interface lags. We optimized the calculation logic and the rendering path so that even complex orders with many configured items feel responsive.

iOS and Android Apps

Our iOS lead built both apps. No complete mobile design existed at project start, so he designed the experience himself: interaction flows, menu navigation, ingredient selection logic, ordering process.

This is what a product recovery on a constrained budget actually looks like. You don’t get a full design phase followed by a full development phase. The engineer building the app makes product decisions because waiting for a design round the budget can’t cover means the product doesn’t ship.

Our iOS lead worked as iOS engineer, Android engineer, mobile UX designer, and product owner for the mobile experience. Holding complex business logic behind a simple interface was the core deliverable.

Backend Integration

The mobile apps integrate with Meal4U’s backend API for restaurant data, menu configuration, order processing, and payments. Real-time sync means menu changes (an ingredient disabled, a price updated, a promotion launched) reflect immediately in the mobile experience.

Where the Project Is Today

The app earns 1,000+ organic downloads per month on iOS. Not launch-week numbers driven by paid acquisition. Steady organic interest from users who find the app and keep using it.

The download count isn’t the story. The approach is.

EltexSoft comes in after failed contractors more often than we come in first. Unfold (Apple Best of 2019) was the same pattern: a codebase inherited from prior teams, stabilized, carried forward for years. PropertyRate is an ongoing Kohana-to-Laravel migration on a live system. Meal4U was a recovery under a budget that had already paid for something that didn’t work.

What these engagements have in common: the client has already been burned. Trust is damaged. Budget is partially spent. The new team has to deliver more with less while rebuilding confidence in the process.

The thing that makes it possible is senior engineers willing to make product decisions, not just write code. Scope engineering that prioritizes business value over feature count. Honest conversations about what to cut. And the discipline to tell a client that a feature they want isn’t worth the budget it would consume.

Scope Engineering Is Engineering

Most failed software projects don’t fail in the code. They fail in the feature list. Too many things to build, not enough budget to build them, no one willing to say “we shouldn’t build this.” The previous Meal4U team tried to build everything. We built half of it, and the half we built works.

The reset week, where the scope got cut down to what the budget could actually support without losing the business model, was the decision that made everything that followed possible.

Who We Are

EltexSoft is a boutique food delivery app development company. 35-50 senior engineers. Headquartered in Lisbon, Portugal. Engineering team in Ukraine. Founded in 2015.

Our FoodTech clients include Meal4U, Ripe (acquired by Hungry), and Tableride. We also build with Laravel, React, Vue, Django, iOS, and Android.

5.0 Clutch rating across 30+ verified reviews. 200+ five-star Upwork reviews. Top Rated Plus and Expert-Vetted agency status (top 1%). Average client engagement: 3+ years.

30-minute technical call. Bring your stalled project, your inherited codebase, or your food platform idea. We’ll tell you what we’d build and what we wouldn’t.

Talk to us →