To the executives who celebrated the green dashboard,
I still remember the woman who called about the stair rail. She said it moved when her father leaned on it, and that he had begun taking the steps sideways because he feared falling. I entered the details into our repair system and promised that the maintenance team would contact her. Three days later, her case appeared in my quality report with a green mark beside it. The record said resolved. When I opened the notes, I found no evidence that anyone had repaired the rail. The final line read: “Customer did not respond to follow-up.”
The woman worked at a bakery that prohibited personal calls during the busiest hours. Our contractor had telephoned twice at midday and left a message without explaining how she could return the call after her shift. The system interpreted her silence as a completed contact attempt. A case could close without the unsafe condition changing. The difference between a person being unreachable and a problem being solved had been removed by a rule, then hidden behind a color that made managers comfortable.
I write from an imagined housing service. The contractor, tenants, workers, and repair system are fictional, not allegations about any particular company. I speak as the former quality reviewer who helped produce those green reports. My concern is not simply that a complaint went unheard. It is that the organization had begun to depend on certain complaints falling silent. Our performance improved when the people most burdened by a problem stopped appearing in the records.
The service began with reasonable aims. Residents needed one number for repairs rather than a maze of different contacts. Property managers wanted reliable evidence of outstanding problems. Contractors needed enough detail to schedule safe and efficient work. The new system assigned each request a reference number, categorized its urgency, and recorded every attempted contact. It made many ordinary repairs easier to coordinate. We were proud of replacing loose papers and unanswered office telephones with a shared record.
Our contracts rewarded timely closure of cases. Managers monitored how many requests remained open at the end of each reporting period, and the company used those figures when renewing agreements. At first the measure seemed fair. An old request should not be allowed to sit indefinitely. But closure had several meanings. A repair could be completed and verified, a request withdrawn, a complaint rejected for stated reasons, or a case closed because the tenant could not be reached during a narrow contact window. All four reduced the number of open cases.
The distinction mattered in practice. A tenant whose sink was repaired could reasonably be satisfied. A tenant who withdrew an unnecessary request might have no ongoing problem. But someone who could not answer at midday might still live with a broken rail at midnight. Our dashboard reported fewer outstanding issues without telling readers which kind of ending had occurred. We were counting the disappearance of a file as though it were the disappearance of a defect.
I raised the question with my supervisor. She said staff were already required to make two contact attempts and that the process was consistent for every resident. She was right about consistency. But two calls made at the same unsuitable hour are not necessarily meaningful opportunities to respond. We had turned a measure of our effort into a presumption about the tenant’s wishes. The process was easy to prove and hard to use for people whose schedules did not match ours.
I thought of the difference between procedural participation and genuine influence. We gave residents a channel, assigned a number, and made an attempt to contact them. The paperwork could show that every required step occurred. Yet the person experiencing the defect might remain unable to obtain a repair or challenge a mistaken closure. The institution offered participation on terms that made withdrawal the path of least resistance.
Then I learned why some people were reluctant to call again. A tenant named Amina told a support worker that she feared being described as difficult after repeatedly reporting a leaking ceiling. Another resident worried that asking for a second inspection could lead to conflict with a landlord during a lease renewal. We had no evidence that every manager punished complaints. The fear itself affected what people felt safe doing. Their silence could not be interpreted as approval without understanding why they had stopped.
Our service’s training materials instructed agents to remain polite and to discourage duplicate tickets. That was sensible when several residents reported the same fault. Yet staff were also discouraged from reopening an old case without fresh evidence, because reopened cases counted against the department’s resolution rate. The company did not need a written order telling workers to hide problems. The performance system made a difficult complaint administratively costly and an unanswered complaint conveniently complete.
One colleague began telling tenants to record calls with careful dates and names. She believed it was the only way to protect their requests from disappearing. A supervisor warned her against giving procedural advice beyond the approved script. She returned to the script, but afterward she often stayed late to check whether vulnerable residents had received a response. Her extra labor did not appear in the dashboard. The green result depended partly on work the system neither counted nor formally allowed.
I started examining a sample of closed cases. Many were properly resolved. Some were withdrawn after residents found another solution. Others had been closed for nonresponse, incomplete information, or an unsuccessful visit. The categories were all permitted by contract. What disturbed me was their pattern: people with irregular working hours, limited communication access, and unstable housing arrangements appeared repeatedly among the cases closed without confirmed repair.
The pattern was not proof of deliberate discrimination by a particular worker. It was evidence that our method of communication fitted some lives better than others. I thought of the way ordinary procedures can determine who reaches a service. A missed call can become a closed door when the institution treats its attempt to contact someone as more important than the person’s ability to respond. Every part of the process can be technically correct while the intended service remains out of reach.
I presented my findings at the monthly review. The operations director asked whether there had been a rise in formal complaints. There had not. He interpreted this as reassurance. I asked whether tenants whose cases had closed automatically knew they could challenge the closure. The contact letters contained that information in small print, but only some residents received and understood them. I asked whether the absence of an appeal meant the underlying problem had been repaired. The room became quieter than I expected.
A director told me I was making an ordinary administration problem sound sinister. I agreed that no conspiracy was required. That was precisely why the problem deserved attention. Our incentives did not require anyone to enjoy neglect. They required people to reduce open cases, and the easiest cases to reduce were sometimes those whose owners could not remain in contact. The system treated their absence as efficiency. It depended upon what it was failing to hear.
We tested a different process with one building. Before closing a case for nonresponse, the service offered contact times that included evenings and a method for replying without a telephone call. Notices explained plainly whether a repair had been verified or whether the contractor had been unable to arrange access. Residents were given a practical route to reopen the request without proving that they were entitled to ask again. These changes took staff time. They also exposed problems we had previously treated as finished.
The open-case count increased in the pilot. Several managers worried that the service had deteriorated. Yet the larger number did not mean more defects had suddenly appeared. It meant the record had begun to show defects that our old process concealed. A dashboard cannot be judged without understanding what the organization has chosen to include. More visible work can be evidence of better information rather than worse performance.
We revised the categories so a repair confirmed by the resident or appropriate inspection was distinguished from a request closed because contact was unsuccessful. We kept reasonable procedures for situations where a tenant did not wish to proceed or access genuinely could not be arranged. We stopped allowing all forms of closure to count equally toward the same success score. A case could remain administratively closed for legitimate reasons without being represented as a verified repair.
The change revealed the way managers had been pressured. They were expected to maintain high closure rates while controlling contractor costs and staff time. A manager who reopened difficult cases could be treated as less efficient than one who cleared them quickly. We had rewarded the appearance of resolution and then wondered why employees favored actions that improved the appearance. It was unfair to blame individual agents without correcting the rule by which their work was judged.
I thought of how organizations teach behavior through what they measure and reward. A worker learns quickly that some questions create trouble while others make a queue disappear. A tenant learns that a repeated request may be treated as an annoyance. A manager learns to regard a quiet building as a successful building. These lessons can accumulate until the institution’s calm depends upon the public’s reluctance to speak.
We introduced a quality review that sampled closed cases and checked whether their stated outcomes matched available evidence. We also examined repeat reports from the same location without assuming a tenant was being unreasonable. Privacy and fairness mattered: nobody should be exposed to retaliation merely because they raised a concern, and property owners deserved an opportunity to respond to disputed facts. The aim was reliable repair information, not a permanent presumption of guilt.
I wrote to the residents in the pilot building with a short explanation of the new process. One replied that she had stopped reporting a recurring fault because the previous messages made her feel foolish. I wanted to apologize personally. Instead I helped ensure that her concern was assessed and that the response described what had been checked. An institution repairs trust not only by saying it is sorry, but by changing the conditions under which a person expects to be dismissed.
The stair rail from that first call was eventually inspected and repaired. I cannot claim that the original silence caused an injury; none was recorded. The danger is not limited to injuries that happened. It includes the conditions an institution permits to continue when it mistakes the absence of a complaint for a safe result. Waiting for a tragedy before reviewing a misleading measure is a poor form of accountability.
The executives’ next dashboard was less uniformly green. It separated completed work, withdrawn requests, unresolved access issues, and cases awaiting verification. Some numbers were uncomfortable. The report invited more questions than the old one did. That was its value. A record that allows the people with authority to ask better questions serves them more honestly than one that rewards them for not knowing.
There are limits to what a repair service can do. Contractors cannot enter every property at will. A tenant may change contact details, decline access, or choose not to pursue a minor issue. Some requests are duplicates or unsupported by inspection. A fair system must handle these realities. It must not turn them into a blanket claim that silence equals satisfaction. The burden is on the institution to describe what it knows and what it does not.
I left the company after the new process had been adopted. My departure was ordinary and unrelated to a dramatic confrontation. I keep a copy of the revised reporting guide because it reminds me that a few words in a category can change what a whole organization thinks it has accomplished. “Closed for nonresponse” and “repair verified” are not interchangeable descriptions of success, however similarly they appear in a spreadsheet.
To the executives, I ask one final question. If every resident were able to speak safely, answer at a time that fitted their life, and challenge a wrong result without fear, would your system still appear as successful? If not, you have been measuring the silence you produced as though it were a service you delivered. A genuinely useful institution does not fear receiving more truthful information about its own failures. It needs that information to do the work for which it exists.
The woman with the stair rail told us, after the repair, that her father had begun holding the banister normally again. The sentence was small and ordinary. It was a better account of success than the first green mark beside her case. A quiet building is not necessarily a safe building. Sometimes its quiet is simply the sound of people who have learned that nobody will listen.
With respect for the people whose unanswered calls once improved our reports,
A former reviewer who no longer mistakes silence for repair



