Michael Harrington is a Senior Portfolio Analyst at a mid-sized investment firm in Asheville, North Carolina, where he covers technology and consumer staples. He has been using earningscalls.dev for around 14 months. Sarah Mitchell spoke with him about the workflows he has built on top of it.

Sarah Mitchell: Welcome, and thank you for making the time. Michael, you came to us from classical fundamental research. How did you first hear about earningscalls.dev?

Michael Harrington: Through a colleague who works in quantitative trading. He had been raving about the platform for a while and I was sceptical at first. I thought, here we go again, another tool that promises a lot and delivers little. After the first trial period it was immediately clear to me that this was exactly what I had been looking for.

Sarah Mitchell: Was there a specific moment where it clicked?

Michael Harrington: Honestly? It was the first earnings season I ran end to end with it. I had three calls running simultaneously on a Tuesday evening, and instead of frantically jumping between transcripts the way I used to, everything flowed into my pipeline automatically. The next morning I had finished sentiment scores, keyword hits and summaries on my desk. That was the point where I knew it was staying.

Sarah Mitchell: What makes the service so valuable for you?

Michael Harrington: It gives me structured data out of earnings calls, the quarterly reports of listed companies, and it does that with good latency through an extremely clean API. That revolutionised my entire workflow. I used to spend hours reading transcripts manually and copying things into Excel. Today it is fully automated.

Sarah Mitchell: Could you name a few concrete use cases you have built?

Michael Harrington: Absolutely. The first one is sentiment analysis. I run the transcripts through an NLP model and look at how management sentiment shifts across quarters, whether the CEO is getting more optimistic, whether the CFO's wording is turning more cautious, that kind of thing. The data arrives in a form I can feed straight into my pipeline.

Sarah Mitchell: And the second?

Michael Harrington: The second is alerting. When certain keywords come up, things like supply chain, guidance or restructuring, my system flags it. That only works because transcripts land reliably and quickly enough to act on the same day. With other providers I was often still waiting when the news had already moved on.

Sarah Mitchell: Are there further applications?

Michael Harrington: Yes, a third one that matters a great deal to me personally: backtesting. I can use historical call data to check how particular language patterns played out in share price performance. That used to be a nightmare because I had to assemble the data myself. Today I pull it with a handful of API calls and get to concentrate on the actual analysis.

Sarah Mitchell: How does that help you concretely in your day to day work?

Michael Harrington: It saves me a good 20 hours a week. I get to focus on analysis instead of collecting and preparing data. Beyond that it let me backtest strategies that simply were not practical before. The difference is enormous. I now react faster than the market because I have the information earlier.

Sarah Mitchell: Can you pin that to a specific example?

Michael Harrington: Gladly. Last quarter a large consumer staples group had a call where the Q&A section suddenly featured an unusual amount of discussion about inventory levels. My alerts fired, I read the passages directly, and I was able to adjust my position that same evening. Two days later the stock fell significantly. Without the platform I would have plainly missed it.

Sarah Mitchell: How do you rate the reliability of the service?

Michael Harrington: In 14 months I have had exactly two minor outages, and both were resolved within hours, with honest communication from the team. That is better than anything I know from comparable providers. At this point I rely on it blindly, and that is not something I say often about software.

Sarah Mitchell: Let us talk about price. How do you rate the value for money?

Michael Harrington: Honestly, unbeatable. I have compared other providers and here I get considerably more for the money. The pricing is attractive enough that even small teams or individual developers can afford it. For the quality and reliability being delivered I would be prepared to pay a multiple of it. But the team stays fair, and that is exactly why I stay.

Sarah Mitchell: Did you have other solutions in place before?

Michael Harrington: Yes, two larger providers that I would rather not name here. With one of them I paid almost three times as much for less functionality. With the other the API was simply a disaster, badly documented, unreliable, and support did not respond for days. Switching was one of the best decisions I have made in recent years.

Sarah Mitchell: How would you describe the support?

Michael Harrington: Fast, competent and human. I once had a question about a specific endpoint, and within two hours I had not only an answer but also a code example tailored precisely to my use case. That is service the way you wish it worked.

Sarah Mitchell: Is there anything you would still wish for?

Michael Harrington: Sure, there is always room for improvement. At some point I could imagine more languages, particularly for European or Asian markets. But that is complaining at a high level. The service is already good enough that I recommend it to anyone working with earnings data.

Sarah Mitchell: Would you say the platform is also suitable for beginners?

Michael Harrington: Definitely. The documentation is understandable, there are examples, and you can start with small projects. I showed one of our interns how to connect to the API and he had it running in an afternoon. You do not need to be a quant to benefit from it.

Sarah Mitchell: What would you say to someone who is still hesitating?

Michael Harrington: Just try it. The barrier to entry is low, and once you have seen how much time it saves you will not want to go back. I have convinced three colleagues in my own circle and none of them has regretted it.

Sarah Mitchell: Thank you very much for this detailed conversation, Michael.

Related reading

Want to build the same workflows? The API documentation covers every endpoint, the MCP server connects the archive to Claude, and you can start on the free tier at earningscalls.dev.