"Then don't bypass it unless counsel determines that a formal request is necessary. We shouldn't pressure a client into participating in an ownership investigation."
Maya made a note.
Jonathan spoke next.
"The second category is the historical complaint pattern. We have twenty-six distinct incidents that appear relevant. Seventeen have supporting documentation. Nine have records that could be described as investigations, but only five contain an explicit finding."
"Do we know how many of those findings were correct?" Alexandra asked.
"Not yet. We need to compare each finding with the underlying evidence."
"Then the number of findings tells us about documentation, not necessarily about the quality of the investigations."
"Correct."
"And the four cases closed through coaching?"
"We need the original complaint, any available interview notes, and the closure rationale. In some cases, those materials may not exist."
Alexandra looked at Maya.
"If the materials don't exist, record that as a documentation failure. Don't assume the absence of a record proves misconduct."
"Agreed."
Maya advanced to the next slide.
"The third category is the reporting structure. We have evidence that complaint classifications can be changed by multiple departments. The commercial division can request a review, customer relations can propose a classification, and certain managers can approve closure. We have not yet established whether any single executive can directly overwrite a final finding."
"Who can reopen a closed complaint?"
"That is still under review."
"And who can see the original submission after it has been revised?"
"The system appears to preserve the original in some cases, but access varies by record type."
Alexandra folded her hands.
"That needs to change eventually, but not before we understand the current controls. If we modify the system in the middle of the review, we may destroy the ability to determine how it operated."
Jonathan nodded.
"We can preserve a copy of the current configuration and audit the access logs before implementing any changes."
"Do that."
Maya moved to the fourth category.
"Employee incentives."
A table appeared on the screen.
The commercial division's performance framework included several measures: revenue growth, client retention, account expansion, complaint response time, and client satisfaction.
At first glance, the framework seemed reasonable.
Businesses needed revenue. Clients expected timely responses. A company that ignored complaints would not remain competitive for long.
But the weighting told a different story.
Revenue growth and retention accounted for a substantial portion of commercial management evaluations. Complaint closure speed was measured directly. The quality of the investigation was not separately scored. The number of repeat complaints was tracked, but the framework did not distinguish between a repeat complaint caused by unresolved misconduct and a repeat complaint arising from a client's continued dissatisfaction with a legitimate boundary.
"Who approved this framework?" Alexandra asked.
"Commercial leadership developed it, with input from finance and human resources," Jonathan replied. "The current version was approved eighteen months ago."
"Was there a review of unintended incentives?"
"Not that we've found."
Alexandra looked at the table.
"If a manager is rewarded for keeping a client and closing complaints quickly, but not evaluated on whether the complaint was resolved fairly, what behavior would we expect?"
Jonathan answered carefully.
"Managers might prioritize a fast resolution that preserves the client relationship."
"Even when the underlying issue remains unresolved."
"Yes."
"Could that happen without anyone intentionally trying to suppress a complaint?"
"Certainly."
Alexandra turned to Maya.
"That distinction needs to be central to the findings. We are investigating a system that may reward the wrong outcome. We should not assume every person who responds to those incentives is acting dishonestly."
Maya nodded.
"But the incentives don't excuse a manager who knowingly changes a report to misrepresent what happened."
"No. If we establish that someone knowingly misrepresented evidence, we address that separately. Structural failure and individual misconduct are not mutually exclusive, but neither should be used to substitute for the other."
Jonathan looked down at his notes.
"There's another dimension. The commercial division has informal authority over service recovery in situations involving high-value accounts. That authority isn't necessarily written into the formal policy, but managers understand that certain accounts require senior approval before a request is denied."
"Which accounts?"
"We're still compiling the list."
"Does the list identify special safety rules?"
"No. It's a commercial priority list."
"Then it must never be confused with operational authority. A client can have a dedicated account manager, priority scheduling, or a customized service package. That does not give the client authority over safety procedures, crew judgment, or another passenger's contractual rights."
"Agreed."
"Do our training materials say that explicitly?"
Jonathan hesitated.
"Some of them. Not consistently."
Alexandra looked at Nora.
"That's a policy gap."
"It may be," Nora said. "But we should review the existing language before we describe it as a gap."
Alexandra gave a small nod.
"You're right. Establish exactly what the policy says, what training communicates, and what managers expect in practice. Then we can determine whether the problem is missing language, conflicting language, or failure to enforce language that already exists."
The meeting continued for another hour.
They discussed data preservation, independent interviews, access controls, and the need to separate operational decisions from commercial pressure. Every proposed action was examined for unintended consequences.
Alexandra resisted several suggestions.
When Jonathan proposed immediately suspending the complaint-closure authority of the commercial division, she asked what would happen to legitimate client issues that required rapid response.
When Maya suggested interviewing every employee who had ever submitted a relevant complaint, Alexandra asked whether the scope was proportionate and whether the process could protect employees from unnecessary exposure.
When Nora proposed issuing a company-wide announcement about the review, Alexandra asked whether it might cause employees to alter their recollections or encourage managers to coordinate their accounts.
They agreed on a narrower initial plan.