A connection date has to survive the work at the customer’s premises.

In 2023 we reviewed sixty contracted business-service orders that had been discussed using a common thirty-day activation expectation. Thirty-eight could use existing access arrangements, while twenty-two required additional external work. The latter group depended on matters that the sales conversation had not yet established sufficiently. I did not think one confident interval was an adequate account of both groups, even though the commercial team valued the clarity of a simple proposition.

The sales director challenged my position because customers needed a date they could plan around. I agreed with the need, but wanted a date grounded in the actual service requirement rather than one chosen before the dependency was understood. We worked through the affected orders with the responsible delivery, commercial and customer teams. Existing contractual commitments were not erased by changing our internal description. Where an expectation needed revision, the account had to explain the applicable agreement and the resolution reached with the customer.

The programme incurred SEK 4.8 million in commercial remediation and coordination expense. I retained that amount rather than calling the revised approach a cost-free improvement in discipline. The thirty-eight existing-access orders and twenty-two requiring more work remained separately visible. I did not claim that every order in either group would necessarily take the same time or that a completed assessment removed all later uncertainty. The change concerned the basis on which we made and explained a commitment.

For future proposals, I supported service-specific committed dates after the relevant assessment instead of the universal thirty-day message. That gave up some commercial simplicity and required an earlier operating contribution. It also made the organisation more answerable for distinguishing an indicative expectation from the commitment actually supported by the order’s circumstances. The customer still needed a usable explanation, not an internal list of reasons why nobody could be responsible.

This fictional case describes a COO’s role in connecting the commercial promise with the work that makes it possible. It is not advice to vary a contract unilaterally or a claim about the legal effect of an installation date. The decision I own is the removal of an overgeneralised promise, together with the expense and customer discussion required to correct it.

A 2016 business-service programme covered twenty-four customer sites. At a review point, technical installation was complete at twenty-one and still pending at three. Of the twenty-one installed sites, seventeen had reached the agreed customer handover and four still required service configuration or acceptance work. The programme could therefore be described accurately in several ways, but it could not be presented as twenty-four completed customer services.

The installation team wanted its substantial progress recognised. The account team needed to explain what the customer could rely on across the full estate. I asked that both accounts be retained. Seventeen accepted handovers, four installed sites still awaiting the next stage and three with installation pending described the actual position more usefully than a single completion percentage. Technical installation and customer acceptance were connected stages, not competing claims from different functions.

We funded SEK 1.7 million in temporary parallel service and support while the remaining work was addressed. That was a disclosed consequence of the programme position, not a claim that the customer had received every intended service without inconvenience. Commercial treatment remained governed by the relevant agreements and authorised decisions; an internal status label did not settle when a fee could be charged or what remedy was due. I kept those questions with the responsible commercial and legal colleagues.

One alternative was to declare the main delivery complete and treat the remaining items as a separate follow-up list. That might have made the programme report easier to close, but it would have weakened the connection between the original customer undertaking and the work still required. I chose to retain the remaining sites in the delivery account. This did not imply that all twenty-four locations had identical needs or that each outstanding item was an installation defect.

The lesson was about the unit of completion. A site with equipment installed, a service technically configured and a customer handover accepted are not automatically the same state. I want an operating account to identify which one it is reporting and why that state matters. This is an invented service-delivery case, with no real customer or network detail, intended to show the evidence behind a senior leader’s completion claim.

In 2021 a review of nine hundred field jobs identified 126 return visits. The initial performance discussion treated the revisit total as a straightforward indication of partner installation quality. The field partner challenged that interpretation, and the customer team could identify examples in which a later visit concerned a requested change rather than correction of the original work. I asked for a more specific account before supporting a contractual or operating response.

The review classified fifty-four revisits as installation correction and forty-two as customer-requested changes. Thirty remained under examination for other causes. Those categories sum to 126, but they do not all carry the same implication. I did not use the unresolved thirty to protect the partner from scrutiny, nor did I label them defects merely to complete a convenient performance measure. The account needed to preserve what we knew and what still required evidence.

We incurred SEK 2.4 million in joint review and qualified rework expense. That figure describes the work undertaken, not a recovered penalty or a financial value assigned to every revisit. The partner remained answerable for supported installation corrections. Our own teams also had to examine how requested changes were specified and passed into the service process. A more accurate classification did not mean reducing the standard of the original work or changing a contract through an informal management discussion.

Some colleagues worried that the detailed review would weaken a clear accountability message. I thought it would make the message more credible. If the organisation cannot distinguish correction from changed demand, it risks directing effort at the wrong problem and making a dispute harder to resolve. I wanted the supported findings used through the proper process, with the unresolved cases still visible and the customer’s actual service requirement kept central.

The consequence was a better basis for the partner discussion, not a claimed causal reduction in future revisits. We had not isolated such an outcome. This fictional case illustrates how a COO can insist on a meaningful operating measure while retaining commercial accountability. It does not identify a real contractor, disclose customer information or imply that every contested classification can be settled by the executive who owns the report.

In 2009 I supported a central routing proposal for eighty planned field visits. The comparison relied too heavily on road distance and the apparent efficiency of grouping work through one dispatch arrangement. Thirteen visits involved island locations whose practical schedules depended on ferry timetables. Local colleagues had raised the distinction, but I had not given it enough weight before supporting the programme. The resulting plan was less workable than its distance comparison suggested.

The correction required SEK 1.1 million in additional travel, cover and rescheduling expense. I kept that cost in the account and acknowledged that the original proposal had used an incomplete measure of the journey. A shorter road distance did not necessarily mean a usable service sequence. The issue was not that every centrally coordinated route was wrong; it was that I had allowed one convenient basis of comparison to stand in for circumstances it did not describe.

We reviewed the affected commitments with local teams and the responsible customer contacts. Qualified staffing, working-time constraints and the permissions needed for the actual work remained part of the revised plan. I did not ask crews to make the original schedule appear feasible through faster travel or unsuitable hours. Nor did I claim that the corrective expense established a particular safety or customer-retention outcome. It described the practical work needed to revise our arrangement.

The central dispatch team reasonably wanted a repeatable method rather than a separate rule for every destination. The local team wanted an operating plan that recognised the journeys they actually made. I supported a revised approach that kept the relevant location constraints in the planning discussion instead of treating them as exceptions to be explained after a missed visit. I also had to admit that the earlier local warning had been available to me.

The case remains useful because it concerns an ordinary management abstraction that failed at a specific boundary. I had mistaken a neat comparison for a sufficiently complete account of the work. This is a fictional Nordic field-service example, with no live routes or customer locations. It demonstrates the consequence of an owned assumption and the expense of correcting it, rather than presenting a dispatch method as a universal solution.

In 2025 a proposed service arrangement for six selected business sites relied on two carriers as part of its continuity proposition. A qualified route assessment identified a common physical entry dependency at the sites under review. The supplier names were different, but that fact alone did not establish the physical separation the customer discussion had implied. I asked that the proposition be qualified before the wider commitment was treated as settled.

The commercial team challenged the additional work because the dual-carrier arrangement had already required substantial negotiation. Engineering colleagues wanted the physical dependency understood on its actual basis. I supported their assessment and coordinated the customer and supplier discussion around it. My role did not include independently certifying the routes or designing the technical alternative. The operating account had to distinguish the specialist finding from the commercial choices it made necessary.

The approved review and alternative-entry work cost SEK 3.6 million across the selected programme. I retained that amount separately from recurring service charges. It was not a valuation of every disruption the work might prevent, and it did not establish uninterrupted service under all conditions. The revised proposition needed to state which dependency had been addressed and which other factors remained outside that conclusion. A second carrier could still have value without supporting every claim initially attached to it.

One possible response was to leave the arrangement unchanged and rely on the supplier distinction as an adequate shorthand. I rejected that explanation because it did not answer the specific finding. Another was to treat the common entry as proof that the entire arrangement was worthless. I rejected that too. The responsible decision required a narrower account of the requirement, the supported alternatives and the cost of the chosen response.

This fictional case shows why a COO should ask what a continuity statement actually describes. It remains at the level of service responsibility and governance: no real site, route, configuration or sensitive infrastructure detail is disclosed. The result is a more defensible customer proposition, not an absolute reliability promise. Qualified technical assessment and the appropriate commercial agreement remain necessary parts of an actual service decision.