Choosing how a grocer stays in town

Most of our store decisions are forced by an asset rather than a strategy: a refrigeration plant, a roof or a car park that needs renewal. I keep a register of every store's major renewal dates beside its catchment forecast, so that the choice between renewing, reducing, converting and closing is made two or three years before the money must be spent. The forecast uses residents within walking distance and within a fifteen-minute drive, the share of households without a car, and the opening plans of the drugstore chains that now sell much of the food in our region. My office proposes; the store-operations director and the district managers test the proposal against what they know of each town. The board sees the options with their costs, including the cost of continuing to trade a store we expect to lose money.

Our 24 vans each carry about 400 items and stop for ten to fifteen minutes at forty to fifty points a week. I insisted from the beginning that the programme keep separate accounts, so that its loss is visible rather than buried in the nearest store. Each route is reviewed twice a year on sales per stop, the number of regular customers and the driver's hours. We add a small per-item charge, which customers have accepted because it is explained on the van. Drivers report concerns about a customer's wellbeing to the town's welfare office under written agreements. I value that arrangement, but I do not count it as income. The programme still loses about ¥60 million a year, and the board has agreed to carry that cost while sales per route continue to meet the review levels.

A chain of our size cannot develop a private-brand range alone at prices that compete with larger chains from outside the region. Through Mitsurane Procurement, we buy jointly with two neighbouring chains and develop the shared Tsumugiba range. The design question is what to share and what to keep. We share national suppliers, dry grocery, household goods and the private brand. Each chain keeps its own fresh produce, its local makers and a protected shelf of regional products that no joint buyer may remove. I helped draft the rules on decision rights, and I sit on the procurement company's board as our nominee. The arrangement works when each chain can see that it gains from joint purchasing and keeps what makes its own stores familiar to their customers.

Our stores cannot recruit all the part-time colleagues they need, particularly in towns where the working-age population is falling fastest. I treat labour hours as the scarcest resource we allocate, alongside capital. That means asking which tasks a customer values when a person does them and which they will accept a machine doing. Checkout is a mixed case: younger customers want speed, while many older customers want a person to scan their basket. Prepared foods are clearer. Our customers buy more cooked dishes every year, and hours spent in the kitchen show up in sales. My office models each store's hours by task, and we move them deliberately, with the store manager's agreement, rather than letting vacancies decide where the cuts fall. The model is revised every quarter.

Two of our stores sit near ski and hot-spring areas in western Iwate, where winter brings visitors from Taiwan, Hong Kong and Australia staying in self-catering lodges. Their baskets differ from those of residents: ready meals, fruit, drinks and breakfast items, bought late in the evening. We extended opening hours during the season, added multilingual shelf labels and stocked more single-portion prepared food. I resisted proposals to redesign either store around visitors. Residents provide about 80 per cent of their annual sales, and a store that served the lodges well in February but felt unfamiliar to its regular customers in June would be the wrong trade. The seasonal plan is reviewed each spring, using till data for the visitor months and comments from the two store managers, who see an evening trade the figures miss.