Seventeen minutes. That is how long, on average, a customer of Mermarlo Mutual Insurance spent waiting on hold last year before reaching an employee, according to internal performance reports reviewed by FFID News. The reports also show that the company has known since 2016 that its customers would prefer not to wait, and that it has spent the nine years since confirming this at regular intervals.
Over four months, FFID News obtained training decks, vendor contracts, internal emails and nine years of call-queue data from the carrier, which takes about 2.1 million calls a year at a facility outside Dayton. The documents were provided by two former employees and through a records request to the state insurance regulator, which receives copies of certain operational filings.
Asked to characterize the findings, one of the former employees said: “They knew. Everybody knew. There’s a slide.”
What FFID News Found
The earliest document is a 2016 slide deck for new call-center supervisors. Slide 4 is titled “What Our Customers Tell Us.” It contains one bullet: “They do not like to wait.” Slide 5 is titled “What This Means for Us.” It contains the same bullet, reworded. The deck was revised in 2019, 2021 and 2024. Slides 4 and 5 did not change. In 2021 someone added a line to the speaker notes under slide 4. It reads: “Pause here. Let it land.”
The company’s survey data says the same thing in more words. Mermarlo asks a sample of callers each year to rate the experience and leave a comment. FFID News reviewed the comment exports. The word “wait” appears in a majority of comments every year since the surveys began. The phrase “finally got through” appears 4,118 times. The phrase “pleasant wait” appears once, in 2020. An analyst flagged it with the note “verify not sarcasm.” It was not followed up.
The Paper Trail
Every quarter the company produces a chart plotting hold time against the share of callers who hang up. The chart has been produced 36 times. In each version the line rises from the lower left to the upper right. Its title, “Abandonment vs. Queue Duration,” has never changed. In the margin of the version from the third quarter of 2022, someone has written, in pen, “same.”
A 2018 email thread with the subject line “Music” shows a workforce analyst named Teodora Vashti-Lund proposing that the company shorten holds rather than improve the music. The thread runs to 41 replies. The final message records the decision: renew the music contract for three years and “revisit queue length at a later date.” The contract, also obtained by FFID News, licenses four instrumental tracks at a price the company redacted. The signature page carries a handwritten note: “the saxophone one is fine.”
Vashti-Lund left Mermarlo in 2020. Reached at her current job, she confirmed the thread and said she had nothing to add. She then added that she still hears the saxophone track in her head “maybe twice a week, usually in the shower,” and that she once heard it in a dentist’s office and had to leave. “I don’t know if it was the same song,” she said. “It didn’t matter.”
The Queue Improvement Working Group
In 2021 Mermarlo formed a Queue Improvement Working Group to study the hold problem. According to its charter, the group had nine members, a budget of $340,000, and a mandate to “develop a data-driven understanding of customer wait tolerance.” It met every other Thursday for fourteen months. Its final report is 96 pages. The executive summary is one sentence: “Customers prefer shorter waits, and this preference strengthens as waits lengthen.”
The report recommends a second phase of study to determine whether the preference holds across age groups, regions and times of day. The second phase was approved. It is budgeted at $410,000. It is scheduled to conclude in 2027.
The chair of the working group, a senior operations manager named Delphine Okoro-Haas, agreed to speak at length. She described the group’s work as “rigorous” and “overdue,” and she keeps a bound copy of the final report on her desk, which she referred to during the call. She said the group had considered more than two hundred data points. Asked whether any of them had pointed in an unexpected direction, she paused for several seconds.
“No,” she said. “But that’s not nothing. Confirming something is still a finding. We confirmed it very thoroughly.”
She added that the group’s methodology section alone runs to nineteen pages and that she is “genuinely proud of it.”
What the Company Says
Presented with a summary of the documents, Mermarlo did not dispute any of them. A spokesperson, Perpetua Ganz, said in a written statement that the company “takes customer time seriously” and “continually evaluates queue performance.” Asked whether Mermarlo had ever found that customers liked waiting on hold, Ganz said it had not. Asked whether it had expected to, she said no. Asked why the question had been studied for nine years at a combined cost of at least $750,000, she said the studies were “part of a broader program.” Asked what the program was broader than, she said she would have to check.
Ganz confirmed that the music contract was renewed again this spring and that a fifth track is “under review.”
Two current employees, who spoke on condition of anonymity because they were not authorized to discuss queue data, said none of this was news inside the building. One said supervisors are shown the abandonment chart during onboarding and that nobody has ever asked a question about it. The other said she keeps a printout of slide 4 taped inside her locker. She said it was not a joke. She said she was not sure what it was.
How We Reported This Story
Documents were compared against one another for consistency and checked against the regulator’s filings where possible. Where the records made claims about customer preferences, FFID News tested them by calling Mermarlo’s customer line six times in September. The average wait was 19 minutes. A reporter heard the saxophone track on every call.
Mermarlo was given eleven days to respond to a list of findings. It responded on the tenth, by email, after the reporter had been on hold with the press line for 14 minutes. The email was four paragraphs. The first three restated the findings. The fourth thanked FFID News for its patience.
The longest single hold in the company’s nine years of data was 2 hours and 6 minutes. The caller stayed on. The notes field reads: “Wanted to ask about a late fee. Waived.”
