How foodpanda became an innovation lab for delivery hero

From Karachi to GIK to Shell: two very different origin stories

The episode opens with Muzamil introducing two guests from foodpanda Pakistan — Ibad ur Rehman Ahmed, who heads Q-Commerce and New Verticals, and Syed Maz Hashmi, Director of Finance. The pairing is deliberate: one guest to explain how the quick-commerce business actually works, and another to explain how a startup survives a brutal economic environment.

Ibad’s path is the more winding one. Born and raised in Karachi, he left the city for the first time to study electronic engineering at GIK — a move he describes as genuinely life-altering. “I would not have been so tolerant of so many different perspectives,” he says. “If I am to understand inclusion or inclusiveness, probably some of those initial routes came from GIK.” After graduating in 2008, he joined Engro in Daharki, then Shell, where he spent five years implementing SAP-based maintenance systems across Pakistan, Oman, and the UAE. That role eventually led him to an MBA at INSEAD — one of the top two business schools in the world by the Financial Times ranking at the time of recording — followed by six years in Saudi Arabia spanning business development across East and Southern Africa, corporate strategy and M&A, and finally a retail turnaround that took a demoralized nine-store chain to thirty-three stores.

Maz’s story runs through Bangladesh. His father was a textile engineer who relocated the family to Dhaka in 1996, where Maz spent a decade doing his O and A levels. He returned to Pakistan in 2006, chose ACCA over CA partly for practical reasons around family finances, and worked his way through a listed auto-components manufacturer, a BPO, and eventually into foodpanda in 2018 — when the company had two people in the finance function. Today he manages a team of sixty.

Why electronics engineers in Pakistan ended up in supply chain and real estate

Before the conversation moves to foodpanda, Muzamil and Ibad spend time on a question that matters beyond their personal histories: why do technically trained people in Pakistan so rarely end up doing technical work?

Ibad’s answer is precise. The problem was not temperament — it was application. “If there is no application out there in the industry, how will you find a relevant job?” Printed circuit board manufacturing did not exist at scale in Pakistan. Telecom boomed and then saturated almost simultaneously. Instrumentation and controls had real demand in oil, gas, and fertilizers, but the overall electronics landscape left most graduates with nowhere to go. The ones who genuinely loved the discipline left the country. The rest transitioned into management, supply chain, or — as Muzamil notes — real estate.

Maz adds a parallel observation from the finance side. The first five or six years of a finance career are spent preparing accounts and doing reporting. After that, the job changes entirely. “Then you are basically an integral part of running the company,” he says. “Business administration comes in. Thinking out of the box comes in. Facilitation comes in.” The ACCA qualification provides the base, but the actual work of finance in a growing company looks almost nothing like what the syllabus describes.

Building the finance function at a startup from two people to sixty

Muzamil asks Maz directly about the moment he joined foodpanda in 2018 — a company that was not yet well known, operating out of a small office, with a rider fleet of 150 to 200 people at most.

“When I shared with my friends, eyebrows were raised,” Maz recalls. “What company is this?” But he had a simple principle driving the decision: “Your life should not have regrets. Karke fail ho gaye toh it’s still a learning.” He was also sold on the vision articulated by the then-CEO. He joined as the person who laid the foundation of the finance department, and he has watched the company scale from that small office to tens of thousands of daily active riders across thirty-five cities.

The financial journey was not smooth. “I came from a manufacturing concern, a very orthodox company where ROI was very important, where you needed to show profit at the bottom line,” Maz explains. “Food Panda aaya toh we were burning money.” In 2019 and 2020, they burned even more. But that period taught him how disruption-model economics actually work — and how fast growth creates a race against your own systems. “The company is running faster than you and you’re just trying to catch up.” Today, he says, the numbers have moved from red to yellow to approaching green. Operationally, foodpanda Pakistan is profitable. The money being deployed now is for growth, not for keeping the lights on.

Why Panda Mart exists and what the shops model is actually for

Later in the discussion, Muzamil pushes on one of the more interesting structural questions in the business: why did foodpanda build its own inventory-heavy dark stores rather than simply connecting existing retailers through the platform?

Ibad explains that both models were tried, and both still exist — but they serve different purposes. The shops model has an inherent limitation: foodpanda cannot integrate with every retailer’s POS system, which means inventory accuracy is always a negotiation. “We have never technically asked for inventory counts. We have only said: is it there or not?” Some retailers treat the channel seriously; many do not. That inconsistency damages the consumer experience.

But shops is not a failed experiment. It is a data layer. “Shops is excellent for finding out what people are searching for, what they are going for, and if it is actually selling,” Ibad says. When a category proves itself on shops, Panda Mart can internalize it with full end-to-end control. The two models are complementary, not competing.

The more important realization, Ibad argues, is that grocery was never the right frame for the business. “This is technically a hyper-local mall at your fingertips.” The delivery radius is four to six kilometers. Within that radius, the question is not just which groceries can be delivered in under an hour — it is which services and categories can be made available that people did not previously think of as deliverable: pharmacy, selective apparel, perfumes, nicotine products, flower shops. “These are the categories we need to build so that a proper mall at your fingertips can exist.”

The recommendation engine gap and what is being built to close it

Muzamil raises a pointed observation: the Panda Mart experience, while clean and functional, largely delivers the same interface to every user every time. There is no visible profiling, no checkout upsell, no equivalent of a retailer knowing that a customer who buys diapers probably also needs infant formula.

Ibad does not dispute this. He explains the structural reason: foodpanda was a food delivery company first. Quick commerce only got its own dedicated global product team roughly twelve to fifteen months before this recording. “It was only around twelve to fifteen months back that Quick Commerce got its dedicated product team.” That team is based globally in Berlin, but it is now actively soliciting country-level inputs to build locally contextualized journeys.

The recommendation engine already exists in a limited form — “customers also bought” appears if you double-click into a SKU description — but the insight is that users do not double-click. “People are in a rush. We need to bring that to the forefront.” The next step is surfacing those recommendations at checkout, and moving from blanket discounts to hyper-targeted offers: “Instead of here is 30% off everything, you hyper-optimize — this consumer, this product, 15% off, and you get the return visit.”

Pakistan as Delivery Hero’s innovation lab and the export pivot

By the end of the conversation, Ibad reveals something that reframes the entire discussion. Pakistan is not a charity case for Delivery Hero. It is one of the top one or two quick-commerce markets across Delivery Hero’s eleven APAC countries by GMV.

“Adoption has been unnaturally different,” Ibad says. “To their surprise, adoption has been faster than many other markets.” Part of that is COVID acceleration, but not all of it. The constraints of operating in Pakistan — thin margins, currency pressure, infrastructure gaps — have forced the local team to innovate in ways that other markets have not needed to. The B2B business, which supplies raw materials directly to the restaurants on the platform, is one example. “We are not going into the wholesale market. We are not going to kiryana stores. We are focusing on the twenty-five thousand restaurants that are already on our platform.” That complementarity — lower procurement costs for restaurants, passed on to consumers, keeping the platform affordable — is a flywheel that Delivery Hero is now looking to replicate elsewhere.

The most unexpected development is exports. Muzamil asks about a word Ibad had dropped earlier in the conversation, and Ibad confirms: foodpanda Pakistan has begun piloting cross-border trade through Delivery Hero’s sister brands — Talabat in the UAE, Hunger Station in Saudi Arabia. The idea is to use trading hubs in Dubai and potentially Singapore to move Pakistani categories into those markets, and to bring imported categories back. “What we can do is proper cross-border. We kick-started a project. They saw the weight of it and said, okay, interesting.” It is early, but it is real.

The economic pressure and how the business is being stress-tested

Muzamil does not let the optimism go unchallenged. Fuel prices in Pakistan had risen roughly fifty percent in the months before this recording. Inflation was compressing consumer disposable income. The global sentiment around quick commerce had shifted sharply — investors who once celebrated twenty-minute delivery windows were now asking whether the model was a business at all.

Maz is direct: “It is a challenge. I won’t say it’s not.” But he frames the response in two parts. First, foodpanda’s addressable market in Pakistan is still far larger than its current penetration — there are still many people who have never used the app, which means growth is available even as the macro deteriorates. Second, the company has deliberately diversified its revenue base: B2B supply to restaurants, home chef options for price-sensitive consumers, fintech partnerships, and the Panda Go last-mile logistics service for third-party businesses.

On the rider side, Muzamil raises the social media criticism that the pay model is not people-friendly. Ibad declines to answer directly, noting that the director of operations and logistics would be better placed to address it — a moment of candor that Muzamil accepts, flagging it as a conversation worth having separately.

2050: what both guests actually believe

Muzamil closes every episode with a long-horizon question. He asks both guests how they see Pakistan in twenty-seven years.

Ibad draws on his time across Africa, Europe, the Middle East, and South Asia. “The amount of people I’ve met outside and their perception of Pakistan is that they are some of the smartest people in the world that they have worked with. We just need to become a team.” He is not dismissive of the diaspora — “baahar rehne mein burai nahi hai” — but he argues that contribution matters more than location. Remittances, he notes, have already played a stabilizing role that is easy to underestimate. “Forget the blip. Things will be good. Go for the long haul.”

Maz is similarly measured. He does not promise a magic fix. “Koi magic nahi hone wala.” Course correction is coming and it will be painful. But he believes the resources, the demographics, and the underlying quality of Pakistani talent are real. “If even from today we start working in the right direction, by 2050 Pakistan could be in a very different landscape.”