INTERVIEW
Migros’s Utku Genç on HoReCa logistics, cost-to-serve and building resilience in Türkiye
Utku Genç, Director – Supply Chain & Logistics OOH & Wholesale at Migros Ticaret, explains how HoReCa logistics differs from store replenishment, why cost-to-serve has become the metric he watches most closely, and where AI is starting to earn its place in Türkiye’s food wholesale supply chain.
Migros is best known in Türkiye for its supermarkets, but its Out-of-Home (OOH) Consumption & Wholesale business serves a very different customer: hotels, restaurants and caterers whose orders are shaped by lunch services, occupancy rates and the space they have in their kitchens.
Utku Genç leads supply chain and logistics for that business, covering direct deliveries to HoReCa customers as well as the Cash & Carry and Tansaş Gross store formats. He spoke to Leaders in Logistics about capacity planning, where to find savings, cost-to-serve and the technology that has – and hasn’t – delivered.
How does HoReCa logistics differ from supplying Migros’s own stores?
Migros Out-of-Home (OOH) Consumption & Wholesale extends Migros’s supply chain beyond our own nationwide warehouses and stores to professional customers such as hotels, restaurants and catering businesses. Their purchasing decisions are closely linked to how they run their businesses, so understanding their operations is an essential part of our service.
My role covers direct deliveries to HoReCa customers, alongside the logistics supporting our Cash & Carry and Tansaş Gross formats. Tansaş Gross is our price-focused store format designed for customers buying in larger quantities. We use separate warehouse operations and dedicated teams for direct customer deliveries and store replenishment. In some locations, these are based within the same warehouse site, in physically separate areas. This allows us to support different formats from one location while keeping each operation focused on its own requirements.
Store replenishment generally follows established schedules and relatively standard processes. HoReCa delivery is more customer-specific. A hotel, a restaurant and a catering company may order the same product, but in different quantities, at different frequencies and with very different delivery windows.
A simple example is a restaurant that needs an order before 11am to prepare for lunch. Delivering the complete order at 3pm does not solve its problem. The products have arrived, but the opportunity to use them has already passed. Delivery timing is therefore part of the value of our service.
‘Delivering the complete order at 3pm does not solve its problem. The products have arrived, but the opportunity to use them has already passed.’
Food logistics adds another layer of responsibility. One order may include ambient, chilled and frozen products, as well as fresh items with a short shelf life. Alongside availability, we must manage temperature control, handling and the remaining shelf life when the goods reach the customer.
These differences make customer segmentation essential. A standard service model for everyone can be expensive and still leave customers dissatisfied, because some delivery arrangements may not match their expectations or meet their needs. We therefore need to understand which requirements are essential and where there is flexibility.
Delivery frequency and timing can vary between customer groups. Food safety, product quality and reliability must remain consistent. Our job is to make those different service arrangements work efficiently, at a cost that is sustainable for the business.
How do you plan network capacity and decide where a new distribution centre makes sense?
In our wholesale and food service operation, capacity planning starts with the outlook from our regional sales teams. They are close to customers and bring local insight into how demand may develop. We assess those expectations alongside each region’s needs and socioeconomic profile, and use them to evaluate future capacity needs.
Our first preference is to make good use of the warehouses we already have. We look at whether they can support the expected growth and service requirements, or whether additional investment is needed. This assessment must consider the type of demand as well as the volume.
For example, we are seeing growing interest in frozen products, including frozen convenience foods and ready meals. If that trend continues, we need to assess further whether we have enough frozen storage capacity and where investment may be needed. A warehouse may have available space overall but still lack capacity in the temperature zone where demand is growing. Looking only at total warehouse capacity would miss that issue.
‘We may end up with a cheaper building and a more expensive operation.’
When a new location is needed, we consider proximity to both customers and suppliers. The site must support an efficient inbound flow while allowing us to meet customers’ delivery requirements. A lower rent can be attractive, but longer journeys, additional vehicles or difficulty meeting delivery windows can outweigh that saving. We may end up with a cheaper building and a more expensive operation.
We therefore assess a new DC based on its total cost and the service it allows us to provide. Regional growth expectations tell us where capacity may be needed; the changing product mix tells us what kind of capacity to provide. The investment needs to bring those two together.
Where do you find savings, and where do you refuse to cut corners?
Managing operations in Türkiye has taught us to look beyond the price we pay for a service and examine how well we use it. A better transport rate matters, but if a truck leaves half empty, there is still a cost problem inside the operation.
Warehouse consolidation is one way we improve productivity. Where it makes operational sense, bringing activities into fewer locations helps us make better use of space, workforce and equipment, and combine delivery volumes. We assess the effect on the whole network, because lower warehouse costs can be outweighed by longer transport distances or weaker delivery performance.
Another example is how we use vehicles throughout the day. We already operate morning and afternoon delivery rounds with the same vehicle where customer requirements and routes allow. A truck completes its early deliveries, returns to the warehouse and goes out again. This helps us spread the workload and reduce the number of vehicles needed at the morning peak.
Inside the warehouse, we also look at product locations, picking routes, repeated handling and staffing around the actual workload. These activities are repeated throughout the day, so reducing unnecessary movement or handling can make a meaningful difference.
‘A short-term saving can become a much larger loss if it damages customer trust.’
Where I draw a clear line is food safety, cold-chain control and the safety of our associates. Correct temperatures, safe equipment and proper handling are basic operating requirements. Rising costs do not change that responsibility.
The same care is needed when changing service arrangements. Fewer deliveries may reduce our transport spending, but if the customer cannot hold enough stock between deliveries, we have to find another way that meets their expectations without leaving them to carry unexpected costs.
A short-term saving can become a much larger loss if it damages customer trust. That is why we focus on removing avoidable work and making better use of resources while protecting the service customers rely on.
Which metric do you pay far more attention to now than when you started your career?
Cost-to-serve: the total cost of meeting a customer’s needs at the level of service we have promised.
Today, at Migros, I apply this across HoReCa, Cash & Carry, Tansaş Gross and wholesale e-commerce. Two orders with the same sales value can have very different costs to fulfil, depending on picking requirements, temperature conditions and delivery arrangements.
When the cost of serving a customer is high, our first practical step is to look at the minimum order value. We discuss opportunities to increase the value of each order with the customer. A small order still requires preparation, handling and a delivery stop, so increasing the basket value can improve the economics of the service. That discussion also needs to take account of what the customer can realistically buy and store. Together with our other efficiency work, our transport cost-to-sales ratio has improved by nearly half a percentage point.
‘Cutting the cost of one activity does not always make the whole service cheaper.’
This makes cost-to-serve useful in conversations between commercial and logistics teams. It helps us connect the value of an order with the resources needed to fulfil it, while keeping service reliability and customer satisfaction in view.
Throughout my career, sales project management helped me understand the operational effort behind commercial promises. Later, managing logistics in industrial kitchen equipment manufacturing, supporting exports to more than 100 countries, reinforced that perspective across different markets and delivery requirements.
Earlier on, the focus was more on the cost of completing each activity. But cutting the cost of one activity does not always make the whole service cheaper. Today, I pay much more attention to the total cost of keeping our promise to the customer.
What are the best strategies for building a supply chain in volatile times?
Demand, costs and customer behaviour can change fast, so frequent reviews have become part of running the business. Being able to act on what they tell us matters as much as the plan itself.
In HoReCa, changes in tourism, hotel occupancy, restaurant traffic and purchasing power can quickly affect orders. These changes do not reach every customer or product in the same way. We try to tell a short-term change from a lasting one, and adjust our purchasing, inventory and capacity decisions to match.
One important lesson is that more stock does not automatically mean more security. Excess inventory ties up cash that we may need elsewhere. With short-life food products, it also creates a risk of waste. We need to protect availability while keeping stock levels in line with sales, expected demand and shelf life.
We have started using AI-supported systems in demand forecasting. This is still a new process for us, and we need more time to assess the benefits. Our aim is to make better use of our data to support purchasing and stock decisions. At this stage, it is important to compare forecasts with actual demand and consider the observations of our commercial and operational teams. Their experience helps us interpret changes that historical data alone may not fully explain.
‘More stock does not automatically mean more security.’
Financial resilience also extends beyond our own inventory and costs. In wholesale, collecting payment is as important as making the sale. We can deliver accurately and efficiently, but the business still suffers if payment does not arrive.
Commercial, finance and supply chain teams need to work closely together. Decisions about which customers to work with and on what terms must reflect financial strength as well as sales potential.
The same thinking applies to logistics contracts and capacity commitments. An arrangement that works at one volume level may become a burden when demand changes. Adjustment options, fixed commitments and exit conditions therefore deserve attention alongside the initial price.
Supply chains can appear operationally strong while carrying too much stock, too much fixed cost or too much financial exposure. Building resilience means managing those risks together while keeping the business ready to respond.
Which supply chain technologies have delivered real results, and which haven’t lived up to the hype?
Route optimisation is one example where we have seen practical benefits. It has made planning faster and improved its accuracy. In an operation with different customer delivery windows, that helps us build plans that better reflect the requirements we need to meet.
We continue to feed operational data into these algorithms, including information from using the same vehicle for morning and afternoon delivery rounds. The aim is to keep improving how we use our transport capacity and the economics of each delivery round.
Another project, now approaching completion, uses AI to support customer order entry in our HoReCa operation. Customers send orders in different formats, including emails and order documents generated by their own systems. Our colleagues then interpret these documents, identify the products and quantities, and enter the information into our system.
An error at this stage can affect the entire delivery process. A wrong product or quantity can lead to customer dissatisfaction, another picking operation, a return or an urgent extra delivery. A missed order means lost sales, and we end up spending additional resources correcting a problem that started before the warehouse received the order.
‘Some of the effort we expected to remove from the warehouse reappeared in managing the connection between the systems.’
The project uses AI to read incoming orders and supports their conversion into consistent order information, with human review for unclear cases. We are yet to measure results. As the project is completed, we need to assess order accuracy, processing time and the amount of correction work required. Those results will tell us whether we are achieving the expected benefits.
An experience that fell short of expectations came earlier in my career, in manufacturing. We were managing production through an ERP system and introduced a separate WMS to improve warehouse efficiency. We underestimated how difficult it would be to keep stock and production records consistent across the two systems. Every material movement and receipt from production had to be recorded accurately, which required additional checks and follow-up work. Some of the effort we expected to remove from the warehouse reappeared in managing the connection between the systems. Over time, we also committed further capital expenditure to improve integration.
That experience taught me to treat integration as a core part of the technology decision. I now pay much more attention to how information will move between systems, how exceptions will be handled and how much daily support will be needed. A system may have strong features on its own. The real test is whether it makes the whole process easier to manage and more reliable for the customer.
What will transform retail logistics in Türkiye over the next five years?
I expect AI-supported planning to have a major impact on retail logistics in Türkiye. It should help us recognise changes in demand earlier and understand what they mean for purchasing, stock and capacity. The challenge will be turning that information into a timely, practical response, including investment decisions.
I expect the greatest benefits when demand planning, operations and investment planning become more closely connected. Commercial teams bring customer knowledge, logistics teams assess the capacity implications, and finance helps evaluate what the business can support. AI can provide useful input, but the judgement and responsibility remain with those teams.
Some changes can be managed with the resources we already have. We may adjust purchasing volumes, reorganise warehouse work or revise delivery plans. When the response requires new equipment, infrastructure or warehouse capacity, however, the situation is different.
AI may show us that an investment is needed, but reaching that investment decision can take time in the current economic environment. The business must still assess the cost, the expected return and other demands on its capital. Once a decision is made, implementation also takes time. Better information does not remove those constraints.
This is why earlier visibility matters. If we can recognise a developing need sooner, we have more time to assess whether the trend will last, consider our options and prepare an investment proposal.
Over the next five years, I believe we will become much better at seeing what is coming. How much value we gain from that will depend on our ability to make the necessary operational changes and investments in time.
At Migros Wholesale & OOH, we measure the value of these efforts by what they mean for our customers. Whether we deliver to their business or they shop at our Cash & Carry and Tansaş Gross stores, they should be able to count on us. As we grow and invest in new capabilities, keeping that trust will remain our priority. In the end, it is earned through the everyday experience of finding the right products, receiving reliable service and knowing we will keep our commitments.
About Utku Genç
Utku Genç is the Director – Supply Chain & Logistics OOH & Wholesale at Migros Ticaret. He has more than 20 years’ experience in supply chain and logistics across the retail, manufacturing and automotive industries.
He leads integrated B2B and B2C end-to-end supply chain operations, covering demand and supply planning, inventory optimisation, national and international distribution networks, cross-border logistics and multi-site operational governance, with full P&L accountability for logistics and transportation.
His expertise includes leading large operational teams, managing domestic and overseas warehouse operations, cost-to-serve optimisation, capacity planning and network design. He focuses on operational excellence and digital transformation through WMS/ERP integration, automation and KPI-driven performance management.
About Migros
Migros Ticaret is one of Türkiye’s major food retailers, operating a nationwide network of stores and warehouses. Its Out-of-Home Consumption & Wholesale business supplies hotels, restaurants and catering companies through direct delivery, alongside its Cash & Carry and Tansaş Gross store formats and wholesale e-commerce.
Stay in the loop with Leaders in Logistics
Subscribe to our newsletter to receive news,
updates and special offers.