{"id":13655,"date":"2026-11-09T10:24:13","date_gmt":"2026-11-09T09:24:13","guid":{"rendered":"https:\/\/salomons.coach\/en\/?p=13655"},"modified":"2026-11-09T10:24:31","modified_gmt":"2026-11-09T09:24:31","slug":"the-operating-plan-everyone-knows-is-impossible","status":"publish","type":"post","link":"https:\/\/salomons.coach\/en\/the-operating-plan-everyone-knows-is-impossible\/","title":{"rendered":"The Ultimate Guide to the Impossible Operating Plan"},"content":{"rendered":"\n

The uncomfortable truths of operational leadership \u00b7 Theme 1<\/em><\/p>\n\n\n\n

Executive summary<\/summary>\n

An operating plan should reflect how work actually runs, including uncertain volumes, rework, and the timing of arrivals. Incomplete measurements call for explicit assumptions and checks against actual shifts.<\/p>\n\n\n\n

However, experienced managers can protect service when plans fail. Their interventions may hide extra costs or defer work.<\/p>\n\n\n\n

Senior leaders need clear rules for immediate decisions and a consistent direction for improvement. Reducing errors and repeated recovery work can improve service and cost together.<\/p>\n\n\n\n

The aim is to build that capability over time. Today\u2019s disruption may still require extra capacity or a change to a service commitment.<\/p>\n<\/details>\n\n\n\n

Imagine the start of a shift in a distribution centre. The volume looks manageable and the roster appears to cover the work. Then an inbound vehicle arrives late. Two agency workers are new and need support. Part of a shipment has incorrect labels. The supervisor moves an experienced employee away from loading to sort it out.<\/p>\n\n\n\n

The outbound departure time has not changed. People are working hard, but the work is no longer moving as planned. By the end of the shift, the team may still get the vehicles away on time. Some other work will have waited, and a few people may have stayed longer.<\/p>\n\n\n\n

At the next review, we see that the departure target was achieved. Unless we ask how, we can easily conclude that the staffing and operating plan were right.<\/p>\n\n\n\n

Having led businesses and worked for many years in operations, I understand why people keep finding a way. Customers are waiting and the shift has to get through the work. What concerns me is when that effort becomes the basis for the next plan, without understanding what was needed to deliver it.<\/p>\n\n\n\n

First check how the hours were calculated<\/h2>\n\n\n\n

Suppose a planning sheet shows 1,000 available hours and 1,150 hours of work. It is tempting to conclude that we are 150 hours short. Before discussing extra people, I would ask how those two figures were calculated.<\/p>\n\n\n\n

To treat the gap as a reliable figure, we need a reasonably well-defined process and measurements that reflect its execution. What work is included? This forms part of the operating plan. This ensures consistency across reporting and assessment.<\/p>\n\n\n\n

What counts as completed? That definition must be clear. Which tasks and delays are missing from the calculation? However, in practice, that foundation may be partly in place. To avoid gaps, align it with the overall scope and timelines. That way, stakeholders can track progress clearly.<\/p>\n\n\n\n

An order-picking standard might cover picking and packing, but leave out replenishment, searching for missing stock and correcting wrong information. A machine standard might come from a long production run, while today’s schedule requires several changeovers. A staffing calculation may count everyone on the roster as equally available, although some are still learning and others are training them.<\/p>\n\n\n\n

Actual output divided by paid hours can give a useful starting point. But we need to know the conditions behind it. A good week with experienced people and straightforward orders is a weak basis for planning a week with a different mix. Historical averages may already include some losses, so adding separate allowances without checking can count them twice.<\/p>\n\n\n\n

The 150-hour gap may still be a useful estimate. It should be presented with its assumptions and the main uncertainties. We do not have to wait for perfect data to act, but we should be clear about what we have measured and what we are estimating.<\/p>\n\n\n\n

The same volume can create very different work<\/h2>\n\n\n\n

Think of two days with the same number of pallets. On one day, most are full pallets with complete information and predictable arrivals. On the other, there are mixed orders, damaged goods and several deliveries arriving together near the end of the shift. The volume total does not explain the difference in workload.<\/p>\n\n\n\n

The same applies in an office operation. Twenty complete customer requests can require less work than ten that involve missing documents, repeated calls and approval from another department. Counting completed requests alone tells us little about why one team needed more time.<\/p>\n\n\n\n

Even the number of exceptions can be misleading. Ten label corrections may be handled quickly. One quality hold may stop an entire order. I would want to understand how often each problem occurs, how much work it creates and where it interrupts the process. An order waiting two hours for approval does not automatically create two hours of labour, but it can still miss its delivery deadline.<\/p>\n\n\n\n

The timing matters as well. Spare capacity in the morning cannot always solve a problem that arrives just before dispatch. Extra people in packing will not clear a hold that only a qualified quality employee can release. A total number of hours needs to be checked against the skills and time windows that matter.<\/p>\n\n\n\n

Experience as a compass<\/h2>\n\n\n\n

In my operations years, the engineering team often missed actual volumes by more than 30%, in both daily and weekly forecasts. A weekly total could also hide a difficult shift. If volume that was expected on Tuesday arrives late on Monday, Tuesday\u2019s spare capacity cannot protect Monday\u2019s departures.<\/p>\n\n\n\n

This is where experienced operations managers made a real difference. They knew which signals to take seriously, who could move to another task and which departure had some flexibility. Sometimes their judgement saved the shift. Moving two experienced people to sorting early enough, for example, can keep an outbound vehicle on schedule.<\/p>\n\n\n\n

Sometimes it went badly wrong. A manager might call in extra people and then find that the expected volume never arrived. Or they might wait because a late vehicle usually brings a light load, only to find there is too much work left to meet the cut-off. Experience helps you see patterns quickly. The risk is assuming the pattern still holds when arrivals, the mix of work or the team have changed. And if one manager is the only person who knows how to respond, the next shift may struggle with the same problem.<\/p>\n\n\n\n

I would ask experienced managers what they saw, when they acted and what happened as a result. Compare that with actual arrivals, hours worked and departures. This helps the team learn which judgements to use again and which assumptions to challenge.<\/p>\n\n\n\n

Look at actual shifts before adding another allowance<\/h2>\n\n\n\n

When the data is incomplete, I would start with a few recent shifts, including one that went well and one that caused problems. Walk through what happened with the supervisor and the people doing the work. Compare their account with the roster, system records and the work still waiting. Use this to find gaps, then check a wider period before treating the pattern as representative.<\/p>\n\n\n\n

Keep the measurement simple enough to maintain. For each shift, record the work due by the relevant cut-off, people available with the required skills, work completed correctly, work carried over and the main reasons for the difference. Record significant rework and overtime separately. If the system does not capture something important, a short manual record can help establish whether it matters.<\/p>\n\n\n\n

Definitions need to be consistent. A shipment marked as processed but still waiting for a document is not ready to leave. A production order that needs rework is not finished output. Otherwise, we can report good productivity while unfinished work builds up elsewhere.<\/p>\n\n\n\n

This review should help distinguish different causes. If similar work regularly exceeds the capacity of a stable process, there may be a real capacity shortage. If missing information keeps stopping the work, we need to address that with whoever provides it. If performance varies because shifts use different methods, first understand the difference and agree a workable standard. More staff may help temporarily, but the permanent decision depends on what is causing the gap.<\/p>\n\n\n\n

Plan for variation that happens in real life<\/h2>\n\n\n\n

A recurring exception belongs in the planning discussion, even if the exact day cannot be predicted. The volume and impact may vary considerably. Some customer requirements are legitimate differences in the work. Others, such as repeated order-entry errors, are problems we should reduce. Both consume time until something changes.<\/p>\n\n\n\n

Hopp and Spearman’s work on production systems explains how variability creates a need for buffers in capacity, inventory or time. Removing one buffer can increase the need for another. [1] My practical question would be which protection the operation needs, given its service promises and the variation we actually see.<\/p>\n\n\n\n

In a distribution centre, this might mean trained people who can move between activities during a peak. In production, it might mean protecting the schedule at the bottleneck and keeping necessary material available. In a service team, it could mean reserving part of an experienced colleague’s time for complex cases instead of filling their whole day with standard work.<\/p>\n\n\n\n

I would avoid solving this with an unexplained percentage added to every plan. Review ordinary days, busy days and days with common disruptions. Check seasonal peaks and changing customer mixes as well. Then agree what the plan should absorb, what response is available and when a commitment needs to be reviewed. Otherwise, the supervisor is left to make those choices during the shift.<\/p>\n\n\n\n

When senior leaders switch between cost and service<\/h2>\n\n\n\n

With that level of variation, balancing operating costs and service was a daily decision. In my experience, VPs and EVPs often asked us to focus on costs when overall business results were under pressure. When service became the bigger business concern, the instruction could switch. Responding to an immediate problem could be necessary. But repeatedly switching the emphasis encouraged us to treat cost and service as competing priorities, without enough attention to what could improve both.<\/p>\n\n\n\n

A supervisor deciding whether to extend a shift feels the immediate effect. Approve overtime and the labour budget takes a hit. Send people home and some shipments may miss their departure. Over the following months, managers can start planning around whichever number was challenged most recently. Over time, some become reluctant to commit to a plan or raise a risk early. They have learned that a decision supported today may be judged against a different priority next month.<\/p>\n\n\n\n

I would ask senior leaders to agree which service commitments must be protected, how much flexibility operations can use and who can approve an exception. If a temporary change is needed, explain why, say how long it applies and make the consequences explicit. At the next review, judge the decision against what the manager knew and what had been agreed at the time. Keep the longer-term improvement priorities consistent, so managers can build capability beyond the latest cost or service instruction.<\/p>\n\n\n\n

Improve cost and service together<\/h2>\n\n\n\n

The longer-term task is to improve how the work is done. Correct shipment information can prevent both a missed departure and time spent chasing documents. Getting an order right first time can avoid a complaint, a second delivery and the handling that goes with it. Better coordination of arrivals can reduce both waiting and the overtime needed to clear a late peak. These are practical ways in which reliable service and lower costs can build on each other.<\/p>\n\n\n\n

There is a research basis for exploring this. Ferdows and De Meyer proposed that manufacturing capabilities can develop cumulatively: improvements in quality and dependability can support lasting cost efficiency. [3] That gives us a useful starting point for examining our own operations. It does not mean that every higher service promise will cost less, or that improvement happens without investment.<\/p>\n\n\n\n

For example, training people to handle several tasks costs time now. If it reduces mistakes and the need for emergency cover, it may improve service and costs over the following months. We need to check both results. Track the total cost of delivering the agreed service, including rework, repeat deliveries and recovery effort. Lower costs in one department tell us little if the work and expense have moved elsewhere. Time released by an improvement becomes extra capacity; it becomes a financial saving only when spending actually changes.<\/p>\n\n\n\n

I will return to this in theme 11, How reliable service and lower costs build on each other. That article will examine the research, practical examples and the conditions under which this works. Here, the implication for the operating plan is already clear: agree how to handle today’s disruption and protect the improvements that can reduce tomorrow’s need for recovery.<\/p>\n\n\n\n

Use an Obeya room to steer the shift<\/h2>\n\n\n\n

The morning meeting should connect today’s work to today’s conditions. What must be ready, and by when? Who is available with the right skills? What is already waiting from yesterday? Which known issue could prevent completion? The answers should change the plan where necessary.<\/p>\n\n\n\n

Consider a parcel hub where each supervisor handles late arrivals differently. One calls in overtime early. Another waits and then pays for an extra vehicle when the scheduled departure is missed. The weekly figures show that labour costs are high and departures are unreliable, but the team cannot see which decisions led to which result.<\/p>\n\n\n\n

The hub introduces an Obeya room, where shift supervisors, planning and transport spend ten minutes together before each shift. On one board they show the inbound arrivals, the work due by each departure, the people available with the right skills and the backlog from the previous shift. They put on-time departures and labour cost per parcel next to each other, with a target for each. They record overtime, rework and extra transport separately so they can see what it took to protect service.<\/p>\n\n\n\n

One evening an inbound trailer is due 40 minutes late. Planning confirms which parcels are on it and transport identifies the departure at risk. The supervisor moves two trained people from returns to sorting for 45 minutes and agrees with the next shift when the returns work will be picked up. They check progress 30 minutes before departure. The parcels make the scheduled vehicle without extending the shift or booking extra transport. If the work were still at risk at the check, the transport manager would decide whether to approve recovery transport or revise the service commitment.<\/p>\n\n\n\n

At the next handover, the team checks the departure, the hours used and the returns backlog. The same supplier has arrived late three times that week, so the transport manager takes ownership of reviewing the collection slot. The team then checks over the following weeks whether late arrivals and emergency spending fall, on-time departures improve and returns stay within their own target. The same routine across shifts makes decisions more consistent. Fixing the recurring late arrival is what can move cost and service towards their targets together.<\/p>\n\n\n\n

Ask what people gave up to meet the target<\/h2>\n\n\n\n

In High performance can hide overload<\/a>, I wrote about giving capable people more work because I trusted them to deliver. I have done this myself. We can make the same mistake with an operation that keeps meeting its targets through extra effort.<\/p>\n\n\n\n

The supervisor covers a vacancy instead of coaching a new team leader. Maintenance is postponed to finish an order. The improvement session is cancelled because everyone is needed on the floor. Each choice may be understandable that day. Repeating them leaves less time to make the operation work better.<\/p>\n\n\n\n

Repenning and Sterman’s research describes how pressure for immediate output can pull time away from improvement and weaken future capability. [2] This also connects to What gets seen, gets repeated: Firefighters vs Preventers<\/a>: the rescue gets attention, while the work needed to prevent the next problem is postponed.<\/p>\n\n\n\n

\n

Before we ask people to deliver more, we need to understand what they already have to do to deliver today.<\/p>\nJan Salomons<\/cite><\/blockquote>\n\n\n\n

What must be in place before you approve the plan<\/h2>\n\n\n\n

When a plan shows a 150-hour shortfall, the leadership team needs to establish what is driving it and decide how to cover the risk while it does so. The next plan needs four things.<\/p>\n\n\n\n