What we asked stores to do—and what we had to change ourselves.

In 2023, twelve stores helped us examine the combined work arriving from head office. Each function could explain its own requests. Merchandising needed a display changed, operations needed a check completed and commercial colleagues wanted information for a campaign. The problem appeared when we put those requests into the same working week. Forty-eight recurring instructions were reaching the trial stores, alongside receiving, customer service and the ordinary work required to keep the shop functioning.

I asked the central sponsors to make decisions about the requests themselves. Twenty continued unchanged. Twelve were withdrawn because we could no longer justify their contribution. Sixteen were combined into eight coordinated instructions, leaving twenty-eight weekly requests. The six-week trial cost C$186,000 in preparation, observation, coordination and cover. Some sponsors objected to losing separate reports because those reports made their own activity visible. I supported the withdrawal where the receiving work did not justify that visibility, while retaining obligations whose authorised owners required them.

The selected task-time observation moved from twenty-four to fifteen hours per store per week across the comparison periods. That figure concerned the defined requests, not every hour worked in a store. The periods were not a controlled experiment, and other conditions could affect the result. We did not convert the difference into a claim about staffing reductions, recovered cash or a universal productivity rate. Store managers also reported that the combined instructions needed clearer wording in two areas; fewer messages did not automatically mean less ambiguity.

The trial changed the conversation because head office had to account for what it was asking stores to absorb. A function could still propose additional work, but it needed to identify the receiving conditions and explain whether another request should move. The cost and the remaining ambiguity stayed in the report. I would involve store managers before recurring instructions accumulate, rather than use a periodic exercise to remove work that should have been questioned earlier. The operating choice was not to ask the twelve stores for greater endurance. It was to accept responsibility for the volume and sequence of the work we were sending them.

During 2015, eighteen stores repeatedly received deliveries during periods when the shop floor also needed concentrated attention. A delivery could be on time according to the carrier arrangement and still arrive at a difficult point in the store day. Managers were moving colleagues between receiving and customer questions, often with both jobs only partly complete. I asked the network team to examine the relationship between the delivery window and the receiving work, rather than treat punctual arrival as the entire operating result.

We worked with the authorised transport and commercial owners on revised windows for those locations. C$320,000 covered coordination, staff cover and changed service arrangements. Carrier colleagues explained that a convenient window for one store could be an inconvenient sequence for the route. Some central managers preferred to retain the existing arrangements and ask stores to adapt. I supported paying for the defined changes where the receiving account showed a material conflict, while leaving locations with different conditions outside the programme.

The revised sequence moved receiving activity away from the selected periods of concentrated customer demand. It did not remove the need to unload, check, move and record stock, and it did not establish one timetable that every carrier or store could use. Managers still needed clear ownership when a vehicle arrived outside its planned window. We kept that exception visible instead of describing the programme as if a new schedule had eliminated uncertainty. The specialist owners retained their authority over service terms and the conditions under which work could be performed.

The outcome was a more usable arrangement for the eighteen participating stores, bought at a stated expense and with continuing coordination. I resisted translating it into a sales increase because we had not established that causal account. What changed was the sequence of necessary work. Looking back, I would have asked for store-day observations when the original service arrangements were reviewed, before the repeated conflict became familiar enough to be treated as normal. A delivery commitment needs to consider the operation receiving it, not just the journey that brings it to the door.

In 2008, I approved a shelf-reset plan for eighteen stores in my region. The central description made the task sound compact: move a defined range, adjust the presentation and complete the supporting checks. I accepted the proposed window without testing enough of the actual work with store managers. The instruction overlooked differences in existing layouts and the time needed to deal with displaced stock while ordinary trading continued. Those omissions belonged to the plan I had authorised.

Thirteen stores completed within the original period. Five needed an extension, and we funded C$94,000 in paid cover and correction work. A colleague argued that extensions would weaken discipline around future deadlines. I understood that concern, but maintaining the date would not make the missing work disappear. I supported a revised sequence with named owners, communicated the change to the relevant functions and kept the additional expense in the programme account. No store was asked to make the report look complete while practical work remained unfinished.

The correction did not mean every delay was beyond a manager’s influence. It meant the receiving conditions had to be understood before individual performance was judged. We reviewed the five locations separately, including what had been prepared, what the instruction had failed to explain and what support the remaining work needed. The other thirteen locations provided useful comparisons, but their completion did not prove that the original plan was equally workable everywhere. A common deadline can conceal materially different starting conditions.

I changed how I sponsored later store programmes. Before approving a window, I wanted a small set of receiving managers to explain the task back in their own sequence, including stock movement, customer access and the ordinary work that would continue around it. That was a planning responsibility, not an invitation to postpone any difficult change. The episode remains in my account because it explains a judgement I had to revise. I had mistaken a concise instruction for a short job. The C$94,000 and the five extensions were the practical price of discovering that distinction after the plan reached the stores.

A 2021 review found that our operating discussion treated returned stock too quickly as available stock. The physical item had come back, but that did not settle whether it was ready for another customer. I asked for a defined cohort of 1,200 units to be followed through the receiving and assessment process. The question concerned what the business could responsibly count as ready for resale, with the authorised product and finance owners retaining their respective decisions.

The reviewed cohort contained 820 units assessed as ready for resale, 230 requiring repair review and 150 excluded from resale. C$76,000 covered handling and review expense. A commercial colleague wanted the full intake recognised quickly in availability because it was already physically present. I disagreed with using location as a substitute for disposition. The 230 units awaiting further review were neither ready stock nor automatically lost stock, and the 150 excluded units needed their own authorised handling route rather than a quiet return to the shelf.

Customer refund decisions were separate from this stock account. The exercise did not establish that every returned unit became cash, that a repair would necessarily be worthwhile or that all units had the same financial value. We also kept the cohort boundary intact. A clear account of these 1,200 units could improve our understanding of the process without proving that the same proportions described every return across the network. Finance retained responsibility for valuation and recognition, while relevant specialists owned product suitability and other acceptance decisions.

The work made the uncertainty useful. Instead of one attractive intake total, managers could see the stock already assessed, the work awaiting a decision and the items excluded from resale. That distinction gave receiving teams a more precise instruction and prevented a commercial availability figure from outrunning the underlying assessment. I would now ask for that separation when a returns process is designed, rather than after physical presence has become an accepted shorthand for saleable stock. The lesson is about the next promise to a customer: we should know what has actually been established before allowing an item’s return journey to become evidence that it is ready for another sale.

For four store refits in 2017, the initial preference was to remain open while work proceeded through twenty nights of partial access. I understood the appeal. A closure would inconvenience customers and interrupt normal trading in a visible way. The alternative nevertheless needed a complete operating account: how the store would receive stock, serve customers and restore its working arrangement between successive periods of access. Remaining nominally open did not answer those questions.

We compared that proposal with nine-day closures and chose the latter for the four locations. C$410,000 funded temporary customer-service arrangements and staff cover, separate from the capital refit programme. Commercial colleagues objected to the clearer interruption in trading. I supported acknowledging that interruption rather than assume the longer phased arrangement would preserve an ordinary store day. Technical, property and other authorised specialists retained their decisions about the work and the conditions under which premises could be used.

The closure choice required specific preparation. Customers needed accurate information about the dates and alternative service arrangements. Colleagues needed a defined account of their work during the period and a receiving plan for reopening. We did not describe those measures as eliminating inconvenience. Nor did the comparison establish a universal rule that a closure is always safer, cheaper or better than phased access. The choice depended on the actual work, the receiving conditions and the limits of each proposed sequence at those stores.

I took two lessons from the decision. First, continuity has to describe the service that remains usable, not simply whether a door stays open. Second, a visible interruption can be more honest than a longer series of constraints presented as normal trading. The C$410,000 remained an expense we had chosen, not a saving against the capital programme. I would require the same discipline if the evidence favoured phased access in another setting. My responsibility was to make the forms of disruption comparable and help the business choose deliberately, with enough preparation that customers and colleagues understood what would change and when.