The questions I bring to a system that other people must keep using

In July 2024 we reviewed sixty-eight digital handover packs intended for continuing property services. Forty-four contained the material the receiving teams needed. Twenty-four did not: sixteen lacked usable maintenance references and eight lacked clear warranty ownership. The files existed, and the visual models could be opened, but those facts did not answer the first practical questions facilities colleagues would face after the project team left. A complete delivery list was concealing an incomplete service handover.

The project sponsor wanted to accept the full set and close the technology workstream, with missing details to follow through ordinary support. Property operations objected that this would transfer unresolved work without a dependable owner. I supported a seven-week correction programme costing SGD1.6 million. We agreed the acceptance questions with the receiving teams and assigned the missing references and ownership decisions to the people who could supply them. My function coordinated the digital material; it did not substitute for the specialists responsible for its operating meaning.

Twenty of the twenty-four deficient packs became usable for the agreed service purpose. Four remained held, leaving sixty-four accepted packs and four explicit exceptions across the cohort. I kept the distinction in the programme report rather than count the existence of every file as acceptance. The outcome did not establish statutory occupancy permission, building safety or completion of the maintenance described. Those were separate matters with their own authorities. This examination concerned the quality of the digital handover to the continuing service team.

The decision delayed a simple project closure and required the sponsor to retain attention on material that was less visually impressive than the model. It also made the transfer more credible. The facilities team could identify what it had accepted and where the remaining answer would come from. I wanted the delivery milestone to mark a usable change in responsibility, not merely the movement of documents from one repository to another while the receiving team inherited the task of discovering what was missing.

I would involve receiving facilities teams when defining the handover pack, before document production begins, so acceptance criteria reflect the questions they will actually inherit.

A December 2024 review examined 900 occupier-service account records before extending a common enquiry process. Eight hundred and ten had a consistent relationship between the service contact and the relevant property account. Ninety required further examination: fifty-four retained outdated occupier contacts and thirty-six involved shared agents whose role was not clear in the proposed record. The application could not responsibly resolve those relationships by assuming that a similar name or familiar email address supplied the missing authority.

The service-platform lead preferred to consolidate the records and let the operating team correct exceptions after launch. Darren Kelvane, our property-operations leader, wanted the unresolved relationships examined first. I supported a five-week programme costing SGD620,000 to clarify the record and the route for questions. The appropriate tenancy and service owners confirmed the relationships; technology made their decisions visible in the process. We did not alter underlying tenancy rights or physical building access merely because an enterprise service record required correction.

Seventy-two of the ninety disputed records were confirmed through the review. Eighteen retained a defined manual route, with the unresolved relationship stated and a responsible owner identified. Across the cohort, 882 records were ready for the agreed automated service route and eighteen needed that additional handling. The result concerned the disposition of the records, not a finding about every request a person might make or a legal conclusion about the property agreement. We preserved those boundaries in the introduction plan.

My choice left some work outside the common route and required the service team to maintain the manual arrangement. That was a visible operating cost, but it was more useful than making the application appear complete by assigning a relationship nobody had confirmed. A property account can outlast several contacts and agents. The system should support the people who must explain those changes, with enough context to handle the next enquiry responsibly rather than treating consolidation itself as evidence that the underlying relationship has been settled.

I would examine changes of agent and contact during the initial account design, rather than leave those relationships for reconciliation after consolidation has been scheduled.

An April 2025 service sample contained 1,200 contacts: 840 through the portal and 360 by telephone. The digital share was seventy per cent, which the adoption sponsor viewed as a reason to move rapidly towards a portal-only arrangement. I asked the team to examine what happened within each group. A first-contact disposition was recorded for 720 portal contacts and 180 telephone contacts. Those figures described a stated next step or outcome at first contact, not proof that every issue had been fully resolved.

The telephone group included questions that required explanation across property and occupier responsibilities. Removing the route would not necessarily turn those callers into straightforward portal users. Operations wanted an assisted option while the team learnt more about the remaining needs. I supported a twelve-week arrangement costing SGD360,000 for assistance and examination of the service journeys. The alternative was to treat the channel share as evidence that the residual group could be moved without a material change in the support required.

We retained both channels during the defined period and used the observations to decide which explanations and handovers needed improvement. I did not claim a staffing saving from the higher digital share. Nor did I combine first-contact disposition with final resolution or present the two channel cohorts as equivalent groups in a controlled experiment. They contained different needs and self-selected routes. The figures were useful descriptions of service demand, but they could not carry every causal or workforce conclusion attached to the adoption proposal.

The decision cost continued assisted capacity and postponed a simpler digital-only operating model. It also gave the programme a clearer subject than raising a percentage. We needed to understand how the enterprise would answer the remaining questions and who would own the difficult handovers. A portal can be an effective part of property service without being the only route every occupier should be expected to use. I wanted the next change to follow that understanding, with adoption serving the service proposition rather than replacing it.

I would investigate reasons for switching channels at the beginning of the portal trial, giving the next service decision evidence beyond the route through which a contact finally arrived.

A September 2025 supplier-exit rehearsal examined 5,000 document references associated with our property platform. Four thousand two hundred could be used in the proposed receiving arrangement. Eight hundred depended on the incumbent environment for interpretation or linked material. The procurement sponsor had expected the export facility to provide a straightforward exit option. It supplied files, but the existence of those files did not establish that the next team could use the record for its continuing property responsibilities.

I recommended SGD780,000 of conversion and examination work before treating the alternative as executable. The supplier-management lead preferred to use the export capability as sufficient evidence in the renewal negotiation. I wanted a more defensible position: identify what could move, what required correction and what would remain dependent. Property-service colleagues helped define the uses that needed to survive. My team tested the receiving arrangement against those uses rather than merely counting files copied from one location to another.

Six hundred and fifty of the eight hundred dependent references became portable. One hundred and fifty still required a specialist viewer. The final position was therefore 4,850 portable references and 150 declared exceptions within the tested cohort. We retained the incumbent's SGD1.2 million annual renewal while the remaining service dependency was considered. The conversion programme did not automatically save that renewal cost, and it did not establish that every other platform function was ready to leave. We had improved the evidence behind an option, not completed a wholesale transition.

The result was less absolute than the early procurement account, but more useful to the enterprise. We could discuss the cost and conditions of an exit with a concrete understanding of the remaining work. The choice also exposed a question for future contracts: how will the enterprise demonstrate that its material remains usable outside the supplier's environment? I wanted that examination to become part of planning, rather than a surprise discovered when a renewal deadline made a patient technical discussion difficult.

A June 2026 proposal offered a comprehensive property application suite at SGD6.4 million over eighteen months. It aimed to introduce several connected capabilities across the portfolio. The programme sponsor valued the coherence of one large change, while property operations questioned whether the receiving teams could absorb all of its dependencies together. I supported a narrower six-month phase costing SGD2.1 million for occupier-service requests at twelve properties, with a clear account of the responsibilities left outside that first stage.

The decision was not based on subtracting the two prices and calling the difference a saving. The programmes had different scopes and periods, and the full-suite proposal had not been approved. We examined what the initial phase could establish and what later investment might still be required. The technology team worked with the selected property teams on the request route, support ownership and training. Other capabilities remained under their existing arrangements while the phased approach was tested.

At the readiness checkpoint, ten of the twelve sites met the agreed preparation conditions. Two still needed training and remained outside the ready group. I reported that split to the sponsor rather than describe the phase as fully ready or extend its result to all thirty-eight properties. Readiness itself was not a claim of improved occupier satisfaction, completed requests or financial benefit. Those outcomes would need observation after introduction. The checkpoint supplied evidence about preparation for a defined part of the programme.

The compromise preserved a smaller next step while making the future obligations more visible. It required the enterprise to manage interfaces between old and new arrangements for longer and accept that one programme would not remove every difference immediately. I considered that cost preferable to asking unprepared teams to inherit a broad system whose support responsibilities were still being negotiated. The next decision would depend on the two remaining sites, operating experience at the ready locations and an honest assessment of what the wider portfolio would need before expanding the change.

I would assess each property’s support responsibilities before announcing a phase boundary, allowing the proposed scope to follow demonstrated readiness rather than an attractive location count.