Talabat – Outside-In Analysis

2-Minute Brief

Three arguments run through this piece. All figures are in US dollars, the currency Talabat reports in.

$9.4Borders handled in 2025, up 28%

51%of order value on the talabat app, from subscribers

+5% vs +41%Gulf vs non-Gulf growth, Q2 2026

Argument 01

The margin is real. Its source is the part under pressure.

Talabat kept 6.5% of every order as adjusted profit in 2025, more than twice DoorDash's. Its former CEO says more than half of that comes from advertising. Kuwait has capped platform fees, the UAE has new pricing rules, and Keeta is buying share with discounts. The profit fell 11% in the first half of 2026.

Argument 02

Talabat innovates constantly – on how it earns and how fast it delivers, not on how you choose or get help.

The subscription, the ads, pay-later and faster grocery are all recent and all working. The moment you choose dinner and the moment something goes wrong look much as they did. Support is the loudest complaint on Facebook and X, and second only to price in App Store reviews.

Argument 03

The biggest decisions are not Talabat's to make.

Its core technology is licensed from Delivery Hero. Uber has offered to buy Delivery Hero and plans to move the group onto its own systems, with the main migrations in 2029. If ordering moves from browsing apps to asking assistants, Talabat's answer will be set by Delivery Hero today and by Uber once the deal closes – not in Dubai.

⚑ The Model

What Talabat is, and where its profit comes from

Talabat started in Kuwait in 2004, passed to Delivery Hero in 2015 and now runs in eight countries: the UAE, Kuwait, Qatar, Bahrain, Oman, Jordan, Egypt and Iraq. Saudi Arabia is not one of them; Delivery Hero runs HungerStation there.

Restaurants and shops list, riders deliver, customers pay. Talabat earns from all three sides at once, and the side that pays most is the restaurant.

The restaurant marketplacesince 2004

Restaurants pay commission – about 17–18% on average, according to the former CEO – and customers pay fees on top. $6.65 billion of the $9.4 billion ordered in 2025.

The core

talabat martown grocery, since 2019

About 160 small warehouses Talabat stocks itself. It sells the goods, so the whole basket counts as revenue.

31% of revenue

talabat prosubscription, since 2022

A monthly fee for free delivery and perks, now in all eight markets. More than one in four active customers pay for it.

51% of order value

Advertisingsold to restaurants and brands

Paid placement in search and listings: 3.5% of every dollar ordered.

$323M, 2025

Talabat's revenue from every $100 ordered, Q2 2026

Commission is the base. Advertising is the profit.

About $39 of every $100 ordered became Talabat's revenue in the second quarter of 2026. The rest of an order's value passes to the restaurant or shop. The two biggest lines are talabat mart, where Talabat sells the groceries itself, and restaurant commission. Advertising is small – yet, by the former CEO's account, it makes more than half of the profit.

Talabat's Q2 2026 results, revenue lines as a share of order value. Advertising is paid by restaurants and brands on top of their orders, not out of the $100. The lines sum to $39.10.

None of the six other listed delivery companies that report profit on order value – Zomato, Uber's delivery arm, Swiggy, DoorDash, Grab and Delivery Hero – matched Talabat in 2025. Without advertising it still leads: about 3.0%, against 1.9% for Uber's delivery arm and 1.7% for DoorDash. The lead narrowed in 2026: 4.9% in the first half, against Zomato's 5.5% for July to September.

✓ What is genuinely strong

A rare delivery company that pays out cash

Adjusted profit before interest, tax and depreciation of $615 million in 2025, 6.5% of every dollar ordered, against 5.3% for Zomato, 3.9% for Uber's delivery business and 2.7% for DoorDash.
Net income of $464 million, $808 million of cash at the end of June 2026 and no external debt.
Talabat is listed in Dubai in its own right and pays out 90% of its profit: $531 million in dividends declared since listing, about 80% of it to Delivery Hero, plus a buyback of up to 5% of the shares.
Subscribers now account for 51% of order value on the talabat app, up from 37% a year earlier, and customers who use more than one category spend almost five times as much as food-only customers, according to the CEO.

△ Where the margin comes from

Restaurants, mostly

Then-CEO Tomaso Rodriguez on a podcast in July 2025: "We have around 6.5% of GMV, which is our EBITDA margin, and… more than half of that actually is coming from advertising revenues." Without advertising and partner-funded discounts, he said, Talabat would "operate a marketplace at a very, very, very tight margin, if not breakeven."
Partners – restaurants and shops, but also banks and card partners – funded $588 million of customer discounts in 2025 – 6.2% of order value, and 7.2% in the first half of 2026.
Gross margin on order value fell from 11.2% to 9.8% between the second quarters of 2025 and 2026, as commission rates slipped (more grocery, and cuts to win big restaurant brands), free delivery spread with the subscription, and vouchers grew.
From 1 September 2026 Kuwait capped platform fees to restaurants at 17% of order value. By Talabat's own description, the cap covers commission, advertising and promoted placement together.

The honest version: Talabat's profit is not a trick – advertising is a real business, and restaurants pay for it because it works. But it means the margin depends less on delivering food well than on owning the screen where people choose what to eat. That screen is what regulators are now pricing, what Keeta is undercutting with discounts, and what AI assistants would change most. It is also why I think the opening is in the ordering experience itself: an app built around why people order, not around a shelf of restaurants, would be something Talabat owns outright.

The Shift Inside the Numbers

The Gulf slowed. Egypt is carrying the company.

Talabat reports two regions: the Gulf (the UAE, Kuwait, Qatar, Bahrain and Oman) and everything else (Egypt, Jordan and Iraq, with Egypt by far the largest). In 2026 Gulf growth fell to single digits while the rest kept growing fast.

Growth in orders by value, year on year

Talabat's 2025 and H1 2026 results, in US dollars. In the second quarter of 2026 alone, +5% against +41%. The Gulf was still 78% of orders by value.

Group growth slowed from 33% in the first quarter of 2025 to 21% in the fourth. The timing matches Keeta, Meituan's international delivery brand, which launched in Qatar in August 2025, Kuwait and the UAE in September and Bahrain in January 2026. Egypt has no Keeta.

The profit moved the same way. Talabat's own accounts split pre-tax profit by country group: the Gulf fell 28.6% in the first half of 2026 while its revenue grew 13%. Egypt's revenue grew 70.5%. Keeta is not the only cause: regional hostilities began on 28 February 2026, and Talabat spent $58 million of its investment programme in the half.

Pre-tax profit, first half, $M

Egypt went from zero to a real profit. The Gulf gave back $79 million.

Talabat's H1 2026 financial statements, segment note, which splits profit into the Gulf and Egypt. Egypt's H1 2025 pre-tax profit was $0.25M.

The other side: Talabat is not losing the Gulf. It budgeted about $55 million in 2026 to defend food against "discount-led competition," says it is spending less than that, and raised its 2026 forecasts – for net income twice, for adjusted profit once, to $535–565 million. The Gulf is maturing under attack, not collapsing. But a company that floated on Gulf growth now depends on Egypt for it.

By the Numbers

What Talabat reports, and who owns it

From Talabat's 2025 annual report (March 2026) and its first-half 2026 results (August 2026). Figures before 2026 exclude instashop; 2026 figures include it.

$9.4B Orders by value, 2025

Up 28% on 2024, excluding currency swings. Revenue was $3.76B.

6.5% Adjusted profit margin

$615M in 2025, on order value. 4.9% in H1 2026, and 4.6–5.0% guided for the year.

51% Order value from pro

June 2026, by value. 37% a year earlier.

+5% Gulf growth, Q2 2026

Against +41% outside the Gulf.

How the habit is built

About 7.7 million monthly active customers at the end of 2025, ordering 6.7 times a month, for 585 million orders in the year. Customers who buy in more than one category – food and groceries, say – account for 75% of order value. The CEO's phrase: "talabat is becoming the cupboard."

How much room is left

By Talabat's own consultants, a quarter of the addressable population in its eight countries orders food through an app, and 3% buy groceries that way. In mid-2025 Rodriguez put it as 0.4 delivery orders per person a month in a region where people eat about 90 meals. Food is maturing in the Gulf; grocery has barely started anywhere.

What the market, analysts and Delivery Hero say

Talabat listed on the Dubai Financial Market on 10 December 2024 at AED 1.60 a share, about $10.1 billion. Delivery Hero sold 20% and kept the roughly $2 billion raised. The shares fell as low as AED 0.627 in March 2026, after regional hostilities began, and closed at AED 1.09 on 23 October 2026 – still about 32% below the IPO price.

Analysts are positive but not on the IPO story. Of 14 covering the stock, 12 say buy, but the average target, AED 1.46, is below the listing price, and the median forecast in the consensus Talabat published in May puts its margin at 4.8% of order value in 2028, against 6.2% for 2025 on the same basis. The market already assumes the margin comes down.

To Delivery Hero, Talabat is both the proof and the cash. It made up more than half of the group's adjusted profit in the first half of 2026. Delivery Hero's September 2026 investor presentation sets Talabat's margin against Uber's, DoorDash's and Meituan's, and its annual report says it uses cash from "USD-pegged markets (e.g., talabat)" to service group debt.

At the 23 October close, Delivery Hero's 80% of Talabat is worth about EUR 4.9 billion – some 37% of the EUR 13 billion Uber is offering for the whole of Delivery Hero. At the IPO Talabat was sold as a growth platform. It now trades like a high-margin delivery company whose margin is expected to shrink. The case for a re-rating is the "everyday app".

Who owns it

2004 · Kuwait

Talabat is founded

2015

Rocket Internet buys it for about $170 million; Delivery Hero takes over the same year

2017–2022

Its own riders, grocery, Otlob in Egypt renamed Talabat, the pro subscription

December 2024

Listed in Dubai. Delivery Hero keeps 80% and licenses Talabat its core technology; Talabat paid it $156 million in shared group costs in 2025

November 2025

Toon Gyssels, its operations chief from 2017 to 2021, becomes CEO

July 2026 → 2029

Uber offers EUR 41.50 a share for Delivery Hero. Close expected in the second half of 2027; the group's move onto Uber's platform in 2029

Product Quality Audit · Public Surface

People love the range. They complain about support and price.

What I Found · First-Hand

Well made, and built for the way people ordered three years ago

I went through the app in October 2026, as a long-time customer in Alexandria. Nothing broke. Restaurant pages are clean, a group order is one tap away, and the basket prices your actual order twice – with and without the subscription – side by side. That is a sales pitch that respects the customer.

But little of it would surprise someone who last opened it three years ago, and where it fell short was the app failing to connect what I wanted with what it can already do.

The basket I couldn't keepA closed restaurant

A six-item sushi order from a restaurant opening in 75 minutes. Checkout suggested other restaurants instead of sending the same basket at noon. Adding one item from a second restaurant would have wiped it: one cart at a time.

Intent lost

The shelfHome and search

Every block on the home screen is a restaurant, a store or an offer. Nothing helps me decide what to eat: no dishes across restaurants, no customer photos, no assistant.

Restaurant-first

A sign in every windowSubscription and offers

The subscription was offered in seven different places, alongside a sticky "EGP 100 off" bar, a "first order" discount, a challenge voucher, a credit card and a points catalogue. My 5,454 points are worth about EGP 155, and 82% expire on the same day.

Crowded

The newer linesDine-out and big brands

The home screen promoted two unrated dine-in venues 165 km away, in Greater Cairo. The food flow is tuned; the newer surfaces are not.

Untuned

What worksCredit where due

Group orders, advance orders, "order again", fees and minimums stated on the restaurant page, and the with-and-without basket. In Egypt one app also carries ten-minute grocery, pharmacies, flowers, pick-up, dine-out, a wallet and pay-later.

Solid

I went in thinking the experience was good but suited to two or three years ago, and that held. The parts that change are the parts that earn – offers, subscription, cards, points. The path from "I'm hungry" to "it's ordered" has stood still.

Against the Other Apps

Behind Uber Eats on helping you choose, and behind Careem on paying you back. Level or ahead on most else.

The obvious yardstick is Uber Eats, though nobody in the Gulf or Egypt can use it at home: Uber closed it in the UAE in 2020, handing the business to Careem, and left Egypt the same year. As a benchmark the gap is real and mostly old: group orders with bill-splitting in 2022, order status on the iPhone lock screen in 2023, customers' dish photos in 2025, and in February 2026 an assistant that turns a typed list or a photo into a basket.

Against the apps people there can actually choose, the picture flips. Careem added scheduled food orders only in November 2025. Keeta has the lowest App Store rating of the group. Talabat ships faster than any of them – 25 iPhone versions between July and October 2026 – but every one carries the same release note.

– not found	Talabat8 markets	Uber Eatsnot in UAE or Egypt	CareemUAE	KeetaUAE and Gulf	DeliverooUAE, Kuwait
Assistant that builds a basket	–	Yes, Feb 2026	Groceries, Jun 2025	–	–
Customers' photos of dishes	–	Yes, 2025	–	–	–
Group orders	Yes	Yes, with bill split	Mentioned	–	Shared tracker
Scheduled orders	Yes	Yes	Since Nov 2025	–	Yes
Automatic credit when late	Vouchers, pro only	–	Yes, 2025	Promised	–
Delivery subscription	talabat pro	Uber One	Careem Plus	None – free delivery without one	Plus

Radar · Talabat vs Keeta, scored /10

The shape is the argument. Everything that makes money or builds habit scores high. Everything a customer meets when deciding or complaining scores low.

Scorecard

Selection and reachi

9

Monetisationi

9

Habit and frequencyi

8

Subscription depthi

8

Speedi

7

Choosing what to eati

5

Scope clarity / positioningi

4

Price clarityi

3

Getting help ⚑i

2

Scores are my own, from using the app, Talabat's reported numbers, its terms pages and public reviews across the App Store, Google Play, Trustpilot, Facebook, X and Reddit. Keeta's scores are my judgement from its own pages, app ratings and reviews, without having used it – ahead on price and simplicity, far behind on range, habit and subscription. Public reviews over-represent people with a complaint.

App Ratings and Reviews

High ratings, mostly negative written reviews

Both are true. The rating averages every tap over a decade; people who write a review this month are mostly those something went wrong for. They show what is breaking.

App Store, Egypt384,697 ratings

4.65 lifetime. Of the 250 most recent written reviews (September – October 2026), 160 gave one star and 57 gave five. The complaints: prices and fees, refunds, and support you cannot reach.

App Store, UAE406,944 ratings

4.58 lifetime. The 250 most recent written reviews split 121 one-star and 101 five-star – far more balanced than Egypt, with more complaints about late or cancelled orders, and nine that name a rival.

Trustpilot1,379 reviews

1.1 out of 5, and 94% one-star. A complaints board rather than a sample – against 7.7 million monthly customers – but useful for what goes wrong.

Ratings as of October 2026.

App Store rating, UAE, October 2026

Lifetime averages on the UAE App Store. Noon's and Careem's cover their whole apps, not only food. Keeta's is the youngest listing, with 12,387 ratings.

Topics in the 500 most recent written App Store reviews

Price first, then support, then money owed

250 Egypt and 250 UAE reviews, 3 September to 9 October 2026, counted by keyword in English and Arabic; one review can raise several topics.

Positive · App Store Egypt, September 2026

Not a day goes by without ordering something from this app whatever I need

Positive · App Store UAE, October 2026

I order everything on talabat and if I have any problems with restaurants or stores they refund me without delays!

Critical · App Store UAE, October 2026

Now that reaching an agent is behind a pay wall. Better to move to another app. 0 support with automated reply only

Critical · App Store Egypt, October 2026, a subscriber

After using the app for several years, I've decided it's better to order directly through your favourite restaurants' own apps – when a problem happens, you'll find someone to talk to.

The last review is translated from Arabic.

The Next Shift

Will people still want to browse for dinner in two years?

Ordering on Talabat today means scrolling restaurants, opening menus and building a basket – its App Store listing still says "Browse a multitude of restaurant menus." Elsewhere that is starting to change. Beyond Uber Eats's assistant, DoorDash launched "Ask DoorDash" in June. Just Eat has a voice assistant in eight European markets. Glovo, Talabat's sister company inside Delivery Hero, has a shopping assistant inside ChatGPT and Claude in Spain, Kenya and Nigeria.

Delivery Hero has built one for Glovo; Talabat has said nothing about one of its own: "AI," "assistant" and "agentic" appear nowhere in its last four results presentations.

The case that it changes everything

An app screen can show more than a dozen restaurants and several paid slots. A conversation usually offers two or three answers. If people ask instead of scroll, the shelf Talabat sells shrinks – and advertising is more than half its margin. Its then product chief named the trade-off in 2025: more ads earn more, but "you take a hit on your organic conversion." A Dubai Reddit user gave the consumer version in September 2026, explaining why they prefer Keeta: "it suggests dishes from restaurants instead of just name of restaurants which makes it quite easy to decide."

The case that it changes little, for now

Adoption is tiny. A broker estimate puts chatbot referrals at 0.4% of DoorDash's traffic; OpenAI scaled back in-chat checkout in March 2026; Google ran food ordering in its assistant for years and shut it. Uber's CEO expects "the majority of our transactions come direct." DoorDash's CEO doesn't think assistants change "our ability to advertise." Assistants also need what Talabat already has – restaurants, riders, payments and a subscriber base – which is why every operator is building them in, not being replaced by them.

What It Could Build

Keep the range. Stop making people shop it like a hypermarket.

Most delivery apps in the region improve the same thing: a bigger shelf, ranked a little better. They win the order, not the person eating it. I would start from why people order and build the trust a cheaper rival can't buy – earning from fewer, better answers, plans people subscribe to, and employer contracts.

01

Recommend, don't list

At 1pm on a weekday, show me the three meals I'm likely to want, not 400 restaurants. Careem says more than 64% of its UAE food customers order for one at lunch, and built set meals around it. The objection: the home screen is the ad inventory.

02

Eat for a goal

About one adult in four in Kuwait has diabetes; in Egypt, 13 million do. Add everyone training or dieting. Zomato scores dishes for health; Calo, a Gulf meal-plan company, says it has served 290,000 customers. The objection: Talabat doesn't cook the food, and its own terms disclaim allergen information.

03

Dinner on repeat

The same lunch every weekday at one, with a tap to skip or swap. Calo is built on it; Swiggy's daily meals – launched 2019, shut 2020, relaunched 2024 – show it is hard. The objection: a customer who never browses never sees an ad.

04

Feed the workplace

Talabat has no visible product for employers, while Careem offers meal allowances in the UAE, and Uber – buying Talabat's parent – agreed in October 2026 to pay $2.3 billion for ezCater, a US corporate-catering marketplace. I found no platform in the region serving hospital night shifts. A bolder version: lunch ordered on an employer shuttle like Swvl's, waiting at the office. The objection: it needs sales teams and catering operations Talabat doesn't have.

05

Pay before the customer asks

One of the loudest complaints is chasing refunds. talabat pro already gives subscribers a voucher when an order runs late; in 2025 Careem began crediting late orders automatically. Go further: settle missing or wrong items, to the card, before anyone opens a chat – Zomato's system settles up to 70% of photo-backed claims. The objection: it costs margin just as margin is under pressure.

How Good Could Dinner Get?

The order reaches 6. Choosing and complaining sit far below.

Start at five stars – the order that simply arrived as expected – then ask what six looks like, then seven, and keep climbing to an eleven that is deliberately absurd. Walking back down shows moves you can't see from five.

11 ★

Deliberately absurd, and that is the point. You never decide. Talabat knows the household, the week and the budget, and dinner turns up before anyone asks. The cupboard restocks itself. Ordering stops being an activity at all.

8 ★

You ask, it answers. "Something light, under 60 dirhams, here by eight" returns three dishes, not three thousand restaurants, and the total you see is the total you pay. If something is missing, the money is back on your card before you open the chat.

7 ★

Nothing to argue about. One price, fees shown before the basket; the subscription shows what it saved you; and a person is one tap away for anyone, subscriber or not.

~6 ★
the order

Everything, delivered. Food, groceries and more from one app, with free delivery on many orders for subscribers. This is why 7.7 million people open it every month.

~3 ★
when it breaks

A chat tree and a voucher. Scripted replies, refunds to the in-app wallet, and a case closed without a call. That is the pattern across the written reviews.

Next 24 Months

What could squeeze the margin

First the direction of travel: Talabat raised its 2026 forecasts – for net income twice, for adjusted profit once – and is spending less than it budgeted to defend food. Five risks follow, each measurable, ordered by how likely they are to bite first.

01 · Already biting

Regulators are capping what restaurants pay

High

From 1 September 2026 Kuwait caps platform fees to restaurants at 17% of order value (10% where the restaurant delivers) – by Talabat's own description, commission, advertising and promoted placement together – and the customer delivery fee at KD 1. Talabat sued; on 27 September a court struck out only the penalties article. The UAE has new rules on subscriptions and predatory pricing, and antitrust investigations are open in Kuwait and the UAE. Watch for: a cap in the UAE. Consequence: if the cap reaches advertising, as Talabat reads it, it reaches the part that makes most of the profit. One rival reads it as an opening: Jahez expects the rules to help Snoonu, its Gulf brand, sign up restaurants in Kuwait.

02 · Already biting

Keeta's parent can outspend everyone

High

Meituan held about $23 billion (RMB 168 billion) of cash and short-term investments at the end of June 2026, and its new businesses lost about $540 million (RMB 3.85 billion) in the half. In Saudi Arabia, where Keeta arrived first, Jahez's local margin halved in a year. Watch for: Keeta in Egypt, where there is still no storefront or announcement. Consequence: the one market carrying growth would come under the same pressure as the Gulf.

03 · Dated

A year or more of waiting for an owner

Medium

Uber's offer has an acceptance period to 5 November 2026 and an expected close in the second half of 2027. Uber would hold at least 40% of Talabat indirectly, "potentially well beyond," and whether minority shareholders get an offer is unresolved. Watch for: a regulator ruling on the minority offer, or on Uber also controlling Careem. Consequence: big decisions on technology, Egypt and new categories wait for a new owner.

04 · Internal

Core technology it licenses

Medium

Talabat has 436 product and technology staff of its own, building on top. But the platform underneath – ordering, dispatch, rider scheduling, the rider and restaurant apps – is Delivery Hero's. The shared group costs Talabat pays Delivery Hero, which cover it, rose 31% to $156 million in 2025. Watch for: what Uber says about the 2029 migration for Talabat specifically. Consequence: Talabat does not control the systems that would turn its customer data into the next product.

05 · Internal

Trust erodes quietly, then all at once

Medium

Support is the top complaint on Facebook and X, and customers in three countries dispute subscription charges. Riders, hired through third-party firms, describe automatic penalties with no way to appeal. Habit keeps people ordering – "people will still use the app" – until a rival offers the same range with a person who answers. Watch for: the 20-point retention lift the CEO reports for subscribers; if it narrows, habit is weakening. Consequence: half of order value sits with subscribers, so the subscription carries the trust risk.

Competitive Intelligence

Attacked from the cheap side and the broad side

Talabat's rivals come at it from two directions. Keeta attacks on price, with a much larger parent behind it. Noon and Careem attack on breadth, folding delivery into apps people already use for shopping or rides. In Egypt the challenge comes from grocery.

Keeta

Meituan's international brand · UAE, Kuwait, Qatar, Bahrain

Aug 2025First Talabat market (Qatar)

3.68App Store UAE (12,387)

$23BMeituan cash and investments (RMB 168B)

10,000+UAE brands, self-reported

Cheaper for the customer and funded from China: in Saudi Arabia its average delivery fee was SAR 11.91 against HungerStation's 15.66 in January 2026. In a Dubai Reddit thread on its first year, five commenters had dropped Keeta or never tried it, against four who used it regularly.

Keeta's pull is discounts, and deal-seekers are not loyal to anyone. The danger is the day it adds what Talabat lacks.

Noon

Gulf e-commerce · Noon Food, Noon Minutes

4.61App Store UAE

10%Commission for Emirati small restaurants, year one

UAE onlyWhere Noon delivers from restaurants

UAE, KSA, EgyptQuick grocery

The Gulf's homegrown shopping app, with quick grocery in three countries and restaurants in the UAE. In April 2026 it offered Emirati small restaurants 10% commission in the first year, rising to 20% by the fifth.

The everyday app Talabat wants to become already exists, built from the other end.

Careem

Ride-hailing super-app · Uber-controlled again since July 2026

AED 19Careem Plus a month

4.54App Store UAE

$100MUber's 12.5% buy of Careem's app business from e&, closed July 2026

UAE, JordanFood markets shared with Talabat

A cheap membership with free food delivery, bundled with rides and payments. If Uber closes on Delivery Hero, it controls both Talabat and Careem's technology in the UAE – an overlap no regulator has addressed yet.

Today a competitor. In 2028, possibly a sibling.

Breadfast

Egyptian grocery app · restaurants since September 2026

4.83App Store Egypt (154,074)

4.65Talabat on the Egypt App Store

$50MRaised Feb 2026, at ~$403M

47 + 35Fulfilment centres, cafés

The reverse of Talabat's strategy: a grocery habit that restaurant delivery, since 14 September 2026, now rides on. More than a million orders a month, its own production sites and private label, 35 cafés and a prepaid card. Rated higher than Talabat on Egypt's App Store. Talabat's answer so far is supply: a 27,000-square-metre talabat mart hub opened in April 2026, serving 12 Egyptian cities.

The strongest challenger in Egypt, the market carrying Talabat's growth.

What happened next door: Saudi Arabia is the best evidence of what Keeta does to a listed incumbent. Jahez's own results show its Saudi margin falling from 10.6% of net revenue in the first half of 2025 to 5.1% a year later, and to 2.7% in the second quarter, with Saudi revenue down 12.6%. Delivery Hero's HungerStation, with a deeper subscription, reported its margin recovering. In its markets, Talabat is the larger player.

The Moat, Stress-Tested

Density is real. Lock-in is not.

✓ What is real

More restaurants mean more customers, which mean more riders kept busy, which mean faster and cheaper delivery – and by its consultants' count Talabat is the largest food platform in each of its eight markets. The subscription turns occasional customers into weekly ones: it carries half of order value, and the CEO says it lifts retention by 20 percentage points.

△ What is thin

Customers move between Talabat, Keeta, Noon and Careem for the better deal. Restaurants list everywhere, rate Talabat's partner app 2.0 on the UAE App Store, and Kuwait now bans exclusivity. Riders work through third-party firms and can carry a Keeta bag tomorrow. What holds customers is habit and the subscription – both of which a richer rival can price against.

Delivery platforms become hard to displace by owning something beyond the order: the supply (Talabat's own grocery stock), the money (as Meituan and Grab did with payments and credit), or the restaurant's back office. Talabat has a start on all three – its own grocery stock, pay-later and a card, and a partner portal where restaurants track orders and buy ads – but none yet holds customers or restaurants in place.

Forward View

Three ways the next two years go

Demand is not the variable – the region keeps ordering. Three other things are: how far regulators and Keeta compress the margin, whether Egypt keeps carrying growth, and what Uber decides once it owns the parent.

Bear

The Jahez path, at scale

Trigger: Keeta enters Egypt, or a UAE fee cap

Keeta opens in Egypt while the UAE follows Kuwait on fees. Advertising and partner-funded discounts shrink together, the margin falls toward the 2–4% of global peers, and Talabat stops looking like a dividend stock. Signal: adjusted margin below 4% of order value for a full year.

Base

Egypt carries it, the Gulf holds

Most likely on current evidence

The Gulf settles at low growth with a margin near 5%, Egypt keeps growing in local terms, and grocery adds slowly. Uber closes in 2027 and Talabat stays listed and well run, waiting for 2029. Signal: the half-year Gulf growth rate staying positive while Egypt's profit keeps rising.

Bull

The everyday app becomes real – through Uber

Trigger: Uber closes and builds on Talabat

This path mostly depends on the deal. With it, Uber controls rides (Uber and Careem) and the region's biggest delivery apps (Talabat, HungerStation and Careem Food), and becomes an everyday company whether or not one app ever does everything. Talabat brings the habit: customers who buy groceries as well as food, pay-later and the card. Whether Talabat becomes the everyday app itself, or one part of Uber's, is open. Signal: a named customer assistant in the app, and grocery growth disclosed on its own.

Closing Thought

Talabat is the reference for delivery across the Gulf and Egypt. It reaches 7.7 million people a month, kept more of every order in 2025 than any listed delivery peer, and pays most of it back to shareholders. So far Keeta has cost it less than it cost Jahez in Saudi Arabia, where the local margin halved. What it has not shown is that it can change how people order rather than how it earns from them. Its margin depends on owning the shelf where people choose dinner, at the moment regulators are pricing that shelf and assistants may replace it. And the decisions that would answer the threat – technology, AI, new categories – sit with Delivery Hero today and Uber tomorrow. The question that decides whether Talabat becomes an institution is not whether it can beat Keeta. It is whether anyone gives it the room to reinvent the thing it already won.

Public sources only: interviews with former CEO Tomaso Rodriguez, former chief product officer Yi-Wei Ang and product director Islam Amin; Talabat's filings, investor presentations and website; Delivery Hero's, Uber's, Meituan's and Jahez's filings; the app stores, Trustpilot, Facebook, X, Reddit and Glassdoor; and press coverage. No internal data, financials, or proprietary metrics were accessed. Review patterns are thematic, not verbatim. Everything here is open to challenge.