Lugmety, Saudi food delivery apps

Lugmety isn't a single app. It is a delivery ecosystem with a customer app that covers food, groceries, personal care and restaurant reservations, a driver app for the people doing the deliveries, a tablet app for the restaurants themselves, and an internal admin app.

Each app had been built fast to get the business moving. By the time I joined, the four did not feel like they belonged to the same product. Patterns were different, language was different, the same action would live in three places depending on which app you opened.

The client wanted research-led recommendations and redesigned flows for the most painful parts of each app, with enough documentation that their in-house team could keep building once we were gone.

I started with a structured heuristic audit of all four apps on iOS and Android. Nielsen's heuristics plus a mobile-specific checklist for native patterns, hit targets, content hierarchy and accessibility. Every issue was logged with a screenshot, a severity rating and a recommendation.

On top of that I ran moderated usability sessions on the customer app with people who had never used Lugmety, plus interviews with two drivers and one restaurant owner already using the older apps. Small sample, but enough to confirm the audit findings were real and not just designer preference.

Only then did I move into redesign. Every change I proposed traced back to either a heuristic violation, a research observation or a stated client problem. Nothing was changed for taste.

The very first thing the customer app asked people to do was pick a country and a language, on a splash screen layered over food photography. Useful, but the visual treatment buried the controls and the get-started button competed with the imagery for attention.

I flagged this as a low-cost win: keep the same flow, but lift the controls into a calmer composition, restore contrast on the wordmark and make the primary action obvious. Below are the original screens, side by side at real proportions.

The first interaction after the splash was a notification permission prompt. People had no idea what the app did yet and were being asked to let it ping them. The carousel that followed showed dots that suggested swiping but only the next button worked, and there was no skip.

In research, the notification prompt was the single most common reason people backed out of the flow. The carousel was tolerated but not read.

The redesign moved the permission prompt to after the user understood the app, swapped the static carousel for friendlier illustrations with a working swipe and a clear skip, and put account creation and sign in directly on the final screen instead of hiding both behind the side menu.

The old home arranged its four categories as tiny icons inside a pin shape, with thin gold labels on a near-black background. People in testing did not realise groceries, personal care and reservations were even available, and the white wordmark over the dark sky failed contrast checks for low-vision users.

The redesign reorganises home around large category cards with a header, image and short description, sized for one-handed use. The layout follows patterns people already know from other delivery apps in the region, so the learning curve goes away.

iOS users were being asked to navigate from a left-hand burger drawer, which is not how native apps work on that platform. Profile, orders and account creation were all stacked together inside the drawer, so people could not find them.

The redesign replaces the burger with a bottom navigation bar, the pattern people already use across other delivery apps. Orders, account and profile move to the bar where people expect them, and account creation gets promoted to a prompt on the home screen so first-time users do not have to go hunting.

The driver dashboard is what the delivery rider sees when they open their app: are they online, how many orders have they completed, how much have they earned, and an XP score from a gamified loyalty system the business wanted to keep.

In the original, all four metrics sat in a stack with no hierarchy. Drivers told us they did not know which number to look at first, and the XP score implied a reward system the dashboard never explained.

I redesigned the dashboard around hierarchy: visually distinct cards per metric, plain-language labels, and a clear availability toggle at the top. The XP system was flagged back to the client as something that needed a real reward loop to be worth keeping rather than a number on its own.

Restaurants run the Lugmety app on a tablet at the pass, not on a phone, so the redesign was sized for that. Staff need to see foods, stock, prices, orders and reservations without juggling screens, ideally while their hands are busy.

The original forced them to jump between sections to piece together what was happening, with no global navigation and no signposting for the things they needed most, like a live order's status.

The redesign keeps the tablet layout but reorganises it around the two jobs staff actually do: manage what is on sale, and track what is happening right now. Status uses colour the way road signs do so a busy manager can read state at a glance.

Marketing pushed back on moving the notification permission later in the flow. Their argument was that allow rates would drop if people had more time to think about it.

I went back to the usability sessions. The data was clear: people who hit the prompt before they understood the app either denied it or abandoned the install. People who hit it after onboarding allowed it more often, even though the conversion window was later.

Sharing the recordings rather than just the summary was what shifted the conversation. We kept the permission prompt, but moved it to after the user had seen what the app does. Allow rates went up. Drop-off in onboarding went down.

This was a redesign and handover project, not a metrics-led growth engagement, so the points below are usability outcomes from testing, not live analytics. The client confirmed the redesigns shipped and that their in-house team continued from the patterns we left them.

When you are redesigning four apps that share a brand and a backend but not much else, the biggest risk is solving each app on its own and ending up with four good apps that still do not feel like one product.

I learned to make consistency a deliberate output, not a side effect. Picking one navigation pattern across customer, driver and restaurant. Picking one status-colour language. Writing the same way in every app. Those decisions made the redesign feel like an ecosystem rather than four parallel rescues.

The other takeaway was how much weight a small, well-run round of usability testing carries in a stakeholder conversation. Five people on video, finishing or failing a real task, ended every disagreement about taste much faster than a deck ever did.

Handover was treated as a deliverable, not an afterthought. I worked with the client's in-house designer and lead developer for the final two weeks to walk through every flow, explain every decision and make sure the documentation was something they would actually open later.

The pack covered the redesigned screens, the patterns behind them, the heuristic findings that justified the changes, and a prioritised list of next steps. Everything was in one place, named the way the team named things, so it could plug into how they already worked.