Today we're speaking with Markus T., Head of Research at a mid-sized multi-strategy fund in Hamburg. Markus oversees a team of six analysts covering European and US equities. He has used earningscalls.dev for a little over a year. We talked about how his team does sector research now, and why he says the old way feels almost negligent today.
Interview recorded on May 22, 2026. Plans and prices mentioned are the ones the interviewee was on at the time; current plans are on the pricing page.
Markus, your quote on the landing page says you can search a whole sector for the quarter and pull every call you need in seconds, versus a week before. What did that week actually look like?
Painful. Two junior analysts would split the sector between them, go to each company's IR page, download whatever PDF or HTML transcript was available, and then manually extract the bits that mattered. The formatting was never consistent. One company's transcript would have speaker names in bold, another would have them buried in the text. We'd end up with a messy spreadsheet that took days to assemble and was already out of date by the time it was finished.
What changed when you moved to the API?
Full-text search across the whole archive. That was the big one. I can query a phrase like "demand softness" or "pricing pressure" across every transcript in a sector for a given quarter, and the results come back in seconds. But the real value is the speaker segmentation. I filter on executive segments only, so I'm not wading through operator chatter or analyst questions. I see exactly what management said, tagged and attributed.
How does that change the output your team produces?
We cover more ground now. Before, we'd pick maybe ten companies for a deep dive because that was all we had time for. Now we scan the whole sector, thirty or forty names, and let the search show us the outliers. Last quarter we caught a mid-cap industrial company whose CFO used the word "challenging" three times in prepared remarks after not using it once the previous quarter. That flag led us to dig deeper, and we ended up trimming exposure before the stock reacted.
What about the learning curve? Your analysts aren't developers.
The REST API is simple enough that anyone who can write a basic query can use it. But honestly, most of my team interacts through the web search on the site. The free reading tier means they can pull up any transcript, search within it, and read the relevant segments without touching the API at all. The API is for the automated sweeps I run myself.
You're on the Pro plan at $24.99. Does that cover your usage?
Easily. We're doing sector-wide searches a few times a week during earnings season, and the 3,000 requests go further than you'd think because each search returns aggregate results across many calls. If we ever hit the limit I'd upgrade to Ultra without blinking, but so far it hasn't been necessary.
What's the biggest limitation you've encountered?
Coverage depth for smaller European names. The platform covers 70 countries, but if I want a transcript for a Nordic small-cap, it's not always there. For large and mid-caps across major markets, it's been reliable. I've learned to check coverage before earnings season for any new names we're considering.
If a peer at another fund asked whether it's worth it, what would you say?
I'd ask them how many earnings calls they read per quarter. If the answer is more than five, they're wasting time doing it manually. The free tier alone is worth bookmarking. The API is what turns it from a convenience into a competitive advantage.
And the speaker tagging, how much do you trust it?
I spot-checked about twenty transcripts in the first month against the original calls. Every executive segment was correctly attributed, every analyst question tagged properly. At 93% named speakers, the coverage is strong enough that I don't think about it anymore. I just filter and trust the output.
One last thing. What does your team do with the time they save?
They read more. Sounds odd, but it's true. When collecting the data takes minutes, the hours go into reading and thinking about what management said.
Want to try the same workflows? You can search a whole sector, pull speaker-tagged transcripts as JSON or connect Claude through the MCP server. Plans and coverage are on the pricing page, and the full reference is in the API docs.