The Real Cost of an Expert Call: What I Learned at Goldman Sachs

✦

Senior analysts at top funds spend 14 hours a month on research logistics. Scheduling calls. Rescheduling when someone cancels. Vetting credentials. Taking notes. Chasing transcripts. That is an entire week per quarter not analyzing investments. I watched this pattern repeat for years. The invoice for five expert calls shows $6,000. The actual cost, once analyst time and failed calls are included, runs $8,250 to $10,500. Roughly 40% of calls produce nothing useful. The effective cost per genuine insight climbs to $2,750 to $3,500. Most firms never calculate this number. They see the vendor fee and miss everything else.

The maths on a five-call project

A typical expert network project runs five calls at $1,200 each. The invoice shows $6,000. The analyst time tells a different story.

Each call requires 4.5 hours of work. One to 1.5 hours defining the brief, reading background, and designing questions. Half an hour to an hour on scheduling, rescheduling, and chasing dial-ins. One hour on the call itself. One to two hours writing notes, extracting insights, and integrating findings into memos.

Five calls equals 22.5 hours of analyst time.

Fully loaded analyst cost at a top fund runs $100 to $200 per hour. Salary, bonus, benefits, seat, data, overhead. That 22.5 hours costs $2,250 to $4,500 in internal time.

Add the $6,000 vendor fee. Total cost: $8,250 to $10,500.

But 40% of calls end up off-target or fail to advance the thesis. Only three of five calls produce useful insights. Spread the total cost across three useful conversations.

Effective cost per useful insight: $2,750 to $3,500.

"The invoice shows $6,000. The actual cost runs $8,250 to $10,500."

The hours that disappear

Those 22.5 hours per project never appear on a P&L statement. They vanish into coordination work that produces no direct return.

I watched those hours get pulled from higher-value activities. Deeper modeling. Broader watchlist coverage. Better collaboration with portfolio managers. Stronger conviction on live positions.

The pattern was clearest during active trades. An analyst would be developing a differentiated view, building conviction, when they had to pause and manage three expert calls for a different investment. By the time they returned to the original analysis, momentum was lost.

An analyst running multiple expert projects each month loses 40 to 60 hours to logistics. That is 500 to 700 hours per year. The equivalent of three months of work spent scheduling, rescheduling, and taking notes.

"Three months per year spent on logistics instead of analysis."

What changes the maths

The firms doing this differently stopped buying access and started buying finished intelligence.

Instead of paying for introductions and doing all the work internally, they provide a 10-minute brief and receive verified answers from correctly profiled experts. Structured outputs. Fact-checked. Ready to integrate.

Analyst involvement drops from 22.5 hours per project to two or three hours reviewing and integrating findings.

That recovers roughly 20 hours per project. For teams running multiple projects monthly, the recovery exceeds 240 hours per analyst per year. Time that goes back into analysis instead of administration.

Closing thoughts

The invoice price for expert calls tells an incomplete story. The $1,200 headline misses the $2,250 to $4,500 in analyst time per project. It misses the 40% of calls that produce nothing useful. It misses the cognitive cost of context-switching between analysis and logistics.

The real cost per useful insight runs $2,750 to $3,500.

The question worth asking is simple. How much analyst time goes to research logistics versus research analysis? The split reveals whether the current model is working or whether hours are disappearing into coordination that compounds nothing.

I saw the pattern for years. The firms that measured it changed how they worked. The firms that did not kept paying three times what they thought.