Would you put an AI answer in front of your board if you couldn't tell which source supported its most important claim? On Tech Talks Daily, I'm joined by Julia Lecocq, who leads corporate sales in Europe and Asia-Pacific for AlphaSense. Julia works with teams making decisions about acquisitions, new markets, and competitive threats. We discuss why having access to thousands of documents can still leave someone with the hardest part of the job: working out which evidence applies to the decision in front of them.

Julia uses earnings call preparation to explain the difference between finding a document and putting it to work. A search might turn up an earnings report. An investor relations team also needs to know what executives have said before, whether outside evidence supports the proposed message, and what analysts might ask next. Julia also describes an unnamed M&A team using AlphaSense's SuperAnalyst to identify gaps in its strategy and compare possible acquisition or partnership targets. The output gave the team a starting point for further research. The decision about which targets to pursue still belonged to the people responsible for the deal.
We talk about what happens when a confident answer contains an outdated figure or rests on an assumption that nobody checks. Julia says a mistake early in a multi-step research task can carry through to the final recommendation. A citation gives an analyst somewhere to start. They still need to open the source, check when it was published, see whether it supports the exact claim, and look for evidence that points in a different direction. Julia also raises a point for managers: when AI takes on the initial research, teams need to agree who checks the work and how analysts will continue developing their judgment.
I joked that technology had been promising me spare time since I was a boy, yet I still haven't made it to the cinema on a Wednesday afternoon. Julia's answer was to look beyond hours saved and ask whether the team identified a risk earlier or spotted an opportunity it might have missed. I'd love to hear how you measure this in your own work. Has AI helped your team make a better decision, and what evidence persuaded you? Please share your thoughts.
Useful Links
[00:00:04] - [Speaker 0]
Would you act on an AI answer if you couldn't see where that information came from? Sure. It's easy to ask that question when the stakes are low. But try asking it when the answer is going into a bored paper and entering a new market or buying another company. Well, I'm your host, Neil c Hughes.
[00:00:24] - [Speaker 0]
Welcome again to Tech Talks Daily, where my guest today is going to be joining me from a company called Alfasense. She's gonna share what she is hearing from the people who have to defend decisions like that. And we'll also talk about how a team can prepare for an earnings call, how a research agent can help build a list of acquisition targets, and why checking the date on a source is sometimes as important as checking the source itself. We'll also get into what happens to the analyst job when the first draft arrives in minutes, not weeks or months, and how you can tell whether the time saved would lead to better decision making. We got a lot to talk about today.
[00:01:08] - [Speaker 0]
I got a fantastic guest, but let me officially introduce you to her now. So a massive warm welcome to the show today. Can you tell everyone listening a little about who you are and what you do?
[00:01:22] - [Speaker 1]
Yes. Of course. And before I do, I just wanna say great to be here. Thanks so much for having me. I'm actually a big fan of the show.
[00:01:29] - [Speaker 1]
You know, I'm someone that is still in my AI learning journey, and so I really appreciate your episodes, Neil, because I think you can take more complex concepts and make them more accessible. And so, yeah, I really appreciate being here. Hello to everyone listening. I'm Julia LeCoke, and I lead the corporate sales team in Europe and Asia Pacific for AlphaSense. I've spent my career in various commercial roles across market insights and technology companies, and I think the big theme if I like could characterize my career has really been in the clients that I work with.
[00:02:08] - [Speaker 1]
So usually corporates and large corporates, and also particular teams, so corporate strategy, m and a, competitive intelligence, market intelligence, and investor relations. I have a real passion for these clients because these are the clients that are making the big decisions for the company. What market to move into? How to respond to a competitive threat? What type of acquisition should we make.
[00:02:35] - [Speaker 1]
And so getting to sit alongside these clients and help them with those decisions has been just a real privilege. Just a little bit about AlphaSense, we're an AI powered orchestration and workflow tool that sits on top of a large content library focused on business content. We help clients conduct very complex end to end research workflows to enable them to do research in a matter of minutes. So just to make that a little bit more tangible for you, say you're in the corporate strategy team, you're looking at potentially entering a new market, and you need to understand the total addressable market. Through AlphaSense and our content library, we can pull together various data points, synthesize those data points, and deliver a full TAM analysis to you in a matter of five, ten minutes.
[00:03:26] - [Speaker 1]
In terms of scale, we work with 8,000 enterprise clients including 92% of the S and P 118 of the top 20 investment banks globally. And we recently announced that we had hit 700 in annual... 700,000,000, I should say, in annual recurring revenue with 40% year over year growth. So that's just a little bit about AlphaSense.
[00:03:51] - [Speaker 0]
Wow. There is so much to unpack there from you being a long time listener, first time caller, to some incredibly impressive stats there. That's incredible. There's a lot that I wanna talk with you about today because we are at a time where companies have access to more data, more research and market information than ever before. But why has access not automatically translated into better decisions until now?
[00:04:16] - [Speaker 0]
Tell me a little bit more about that journey and why things are changing.
[00:04:21] - [Speaker 1]
Yeah. I think we're all suffering a bit right now from information overload. What I hear consistently from clients is that more information doesn't necessarily translate into bit better intelligence. In fact, businesses have enormous access to volumes of research, financial disclosures, competitive announcements, industry developments, and it's all sitting in different silos around the organization. And typical LLMs have made that information easier to access.
[00:04:52] - [Speaker 1]
They can pull together all these disparate pieces and summarize it for clients. But I hear a couple of things from that experience that still require human touch and human work to help make better decisions. Number one, those summaries, they need to be vetted for accuracy. So clients, especially, you know, considering they're making those high stakes decisions we talked about, they have to make sure that every single data point that's in that summary is relevant, that it's timely, you know, it's from 2026, not 2016, that it's not made up or based on a hallucination. So that's one thing, and that can be very manual.
[00:05:34] - [Speaker 1]
And then the second piece is that they... Within these vast summaries, they still have to find the right signals that matter to them within the context that they're working on. So it's not just a matter of actioning this great body of research, but they have to sort through it and figure out what are the little nuggets of information that are gonna be... That are gonna make the difference in this decision.
[00:05:58] - [Speaker 0]
Now I was gonna say, well, if we dig a little bit deeper on that and some of the problems that you're seeing inside organizations, what is the... Maybe the practical difference between search information retrieval and that general intelligence that is inside every single organization. It's a big difference. It's it's easy to get confused when everyone's putting all this data into LLMs, etcetera, and what intelligence looks like inside that organization. What are you seeing here?
[00:06:24] - [Speaker 1]
For sure. I think the easiest way for me to explain this is to give you an example. So say you're part of an investor relations team. That's a that's a big type of team that we work with. Traditional search might enable you to find a competitor's earning report, whereas AI powered retrieval might be able to gather several earnings reports and other other types of investor relations documents.
[00:06:50] - [Speaker 1]
But genuine intelligence finds not only the reports, but also connects them with market trends, expert perspectives, and internal knowledge. So just to build on this use case a little bit, when... In working with investor relations teams, a big use case we have is in helping them prepare their CEOs and CFOs for earnings calls. And so with AlphaSense, you can pull all past earnings calls transcripts, track sentiment, track what was said, and then AlphaSense can help create scripts that your CEO and c f... CFO can use on the next earnings call to ensure that he or she is building a consistent story.
[00:07:36] - [Speaker 1]
And then we can also pull in expert insights or other data sources to support any sort of claims or assertions that isn't... That are in that script, and also predict what sort of questions are gonna come from the analyst community. You know, if you were to do a simple search, you'd get... You might be able to find all this information, but then really pulling it together would take quite a bit of time. And, you know, just to build on that a little bit, Neil, genuine intelligence is something that I'm really excited about.
[00:08:06] - [Speaker 1]
I think we're gonna continue to get better and better at it. You know, I know in your podcast, you've talked about 2026 being the year of the agent, and certainly for AlphaSense, that is true. In October, we actually launched superanalyst, which is our always on agent. And superanalyst represents the evolution from AI search towards more autonomous end to end research workflows that plan, execute, and monitor complex research tasks for users. Can I
[00:08:40] - [Speaker 0]
give you an example? Yeah. Please do.
[00:08:42] - [Speaker 1]
Okay. So we're working with... One of our clients is a large multinational company and working with the m and a team. Historically, their approach to identifying m and a targets has been fairly opportunistic, which actually is quite common, I would say, across clients. You know, the board hands down an idea that they need to go check out or a colleague recommends an m and a target or a friend of a friend comes into the picture.
[00:09:10] - [Speaker 1]
But in working with this client, we understood that he wanted to be a little bit more strategic in terms of the different m and a targets he was assessing. So we used super analysts to build a deliverable that looked at his company's strategy, analyzed the performance in the gaps in performance based on that strategy, looked for other companies to acquire or partner with to improve where the gaps are, and then provided a heat map of attractiveness in terms of all of those different potential targets. And his feedback was absolutely stellar. He said that this was something that he'd always wanted to do and like had built in his head, but just never had the time to actually create it himself. So that's the type of thing we can do, and I'm really excited to see where we go with this and and how we use superanalyst to work with clients.
[00:10:05] - [Speaker 0]
What a great example. I think that really brings to life what you're doing here. And of course, for every leader listening, as generative AI has become that first front door to enterprise knowledge, and now we're adding agents that interact with customers, employees, and indeed other AI agents on many occasions. How do teams prefer, preserve source attribution? Context.
[00:10:27] - [Speaker 0]
Again, that word context is a massive word this year, and also confidence rather than just accepting that very polished and believable answer at face value. If you could share around that particular challenge too?
[00:10:40] - [Speaker 1]
For sure. Like, my own experience as well as what I've seen from clients. So it's a good point. More... As we...
[00:10:46] - [Speaker 1]
As more and more people use these type of tools in a professional setting, we need to really pay attention to the context and to the data feeding these tools and guard against what, you know, I think people lovingly or fondly call AI slop in a way.
[00:11:01] - [Speaker 0]
Yeah.
[00:11:02] - [Speaker 1]
Many of us have had the experience of running an AI report, receiving a polished, confident answer, and then probing a little bit and realizing that it completely falls apart. And, you know, again, just to go back to the clients that we're working with and the types of decisions that we're working with, a key prerequisite or requirement for business executives is really to be able to trace all of the conclusions back to the sources when they're using AI tools from when it was published to its relevance to the data point that supports that claim, or are there any other contradictory data points as well? And something that I hear from clients a lot is that, you know, Julia, when we get a... Get an AI report and it's not factual, it actually takes my team more time to go back and redo the work versus just having done it manually the first time around. So it's really important that humans are in the loop.
[00:12:07] - [Speaker 1]
We're verifying and we're using tools that do provide these types of traceable elements so that we're really smart about how we use them to inform decisions.
[00:12:18] - [Speaker 0]
And listening to you now, I can almost feel people all around the world nodding in agreement, in unison there. Because I think in our personal lives, maybe in the workplace as well, those type of hallucinations you've mentioned now, I think we've all seen. But I'm curious, when have you seen hallucinations like this and maybe incomplete data or fragmented data create the greatest risk when people use AI for anything in the workplace from financial, strategic, or competitive research because it'd be great to just shine a light on the kind of risk that maybe you could unwittingly create if you haven't got that critical thinking or knowledge that not just to believe everything that it spits out.
[00:12:57] - [Speaker 1]
Yeah. And I think this is really important for people making decisions to fully grasp and understand. Hands down, the greatest risk is when a plausible but inaccurate answer becomes a foundation for a major business decision, whether that is entering a new market, acquiring a new company, making an investment. And I think the risk is gonna continue to get greater as AI takes on more complex multi step research. An error early on in the process can have major impact on the downstream and the output of what comes up.
[00:13:37] - [Speaker 1]
So it's just absolutely essential that organizations can scrutinize the AI findings and verify their evidence before they act on it. And in fact, you know, I feel really strongly that we should probably not only maintain standards, but increase our standards for verification, for citations as we move forward to just help us guard against the potential risk that can come when that doesn't happen. And when I talk to senior executives, this is really one of the biggest watch outs for them because the bad investment that they make or a bad strategy could completely derail their company. So it's something not to be taken lightly.
[00:14:20] - [Speaker 0]
Your agentic AI might not be secure even with real time data and proper guardrails, But Denodo make sure your business has every avenue covered. By placing all your data platforms under one AI data layer, your business can reach semantic consistency safely and securely. So get your agents on the same page by visiting denodo.com, and you can learn more about how to start trusting your agents to make business decisions. But now back to today's guest. Although we have highlighted some of the risks there, on a more positive note, what changes in the role of an analyst or knowledge worker when AI can now summarize thousands of documents in seconds and what would have took weeks or maybe months before.
[00:15:11] - [Speaker 0]
What kind of changes have you seen here?
[00:15:13] - [Speaker 1]
You know, we actually hosted a recruiting event in our London offices this Wednesday, and we had a very similar topic. So I... I'm happy to address this. No no doubt AI is changing the value of human expertise, but I think there's an opportunity for us to really lean in and do more high value work. So the way that I see it, you know, analysts have traditionally spent significant time gathering information, doing initial research, but now AI can take away a lot of that groundwork, which means that individuals can be elevated to more strategic work.
[00:15:52] - [Speaker 1]
And in fact, we see this at play a lot with our clients. Just as one example, Wendell, which is one of Europe's leading investment firms, they use AlphaSense to monitor their portfolios, spot high potential companies, make better investment decisions quicker. And we ran an analysis with them and found that we were actually saving their analysts about 50% in terms of time, and that 50% got allocated to much more high value tasks. So as AI is sort of leveling the playing field when it comes to the manual work that has to happen in a knowledge worker's day, there's opportunity for high performers to really shine because the types of skills that are needed are all around decision making. So judgment, creativity, execution, can you get things moving fairly quickly?
[00:16:47] - [Speaker 1]
And so I think it gives, yeah, everyone an opportunity to shine. I think in my role leading the corporate sales team, I spend a lot of time thinking about this actually from a leadership perspective as well. Can I go off book and give you a couple of thoughts there, Neil?
[00:17:03] - [Speaker 0]
Oh, please do. I'd love to hear. I really I really think it helps bring what we're talking about to life. So... Yeah.
[00:17:08] - [Speaker 0]
Yeah. Go for it.
[00:17:10] - [Speaker 1]
So a couple of things I think leaders really need to think about as we move into this new era. Number one, just we've got to have clarity and guidelines between human work and AI work, and when each is needed and how we're managing performance for both. That clarity is really really important in order to make sure that we still maintain accountability on the human. So just as the human manager is gonna take accountability for any work an analyst does that's on her team, In the same way, we need to take accountability for the work that the AI does for us. And then I think if we don't do that and we don't have clear guidelines and clear accountabilities, we run the risk of AI probably doing too much for us and getting a bit over our skis, and then we also run the risk of not continuing to build our own skills and develop ourselves professionally.
[00:18:05] - [Speaker 1]
So I think that's important. Another component is vision setting and purpose being more important than ever. So as we upscale our workers on, you know, no longer needing to do as much manual work, now we're we're focused on decision making, judgment, creativity. These are all more high value skills, and these workers are gonna want and be interested in a place where they can really connect with the purpose and the vision. So I think that's important.
[00:18:35] - [Speaker 1]
And then driving connection. So work is no longer a come come into work, do these manual things, leave, see you the next day. Work is very much come in, involve... Understand what's taking place with the help of AI, and work with colleagues to get to the right solution or the right decision. So driving connection between leader and the team, and then also in between teams is gonna be really important.
[00:19:02] - [Speaker 1]
And and then just finally, like, at the end of the day, everyone's gonna wanna feel like a human and wanna feel valued. So just making sure we continue to focus on that too.
[00:19:13] - [Speaker 0]
And since I was a small boy many decades ago, technology did promise to save us all more time, give us more time back in our lives. But I think for most people, that means doing more work rather than drinking cocktails on a beach or going to the cinema on a Wednesday afternoon. So I'm curious from what you're seeing out there, which signals tell a leader that an AI research tool is improving that decision quality rather than just helping people create more and more output?
[00:19:42] - [Speaker 1]
Yeah. So it's a great question. Definitely, the number of reports produced or the hours saved doesn't necessarily tell you if a decision was better. Instead, we should be thinking about whether teams are identifying risks earlier, uncovering new opportune... Sooner, making decisions with better confidence.
[00:20:04] - [Speaker 1]
A question I really like to ask clients is whether AI is revealing any insights that they may have otherwise missed along the way rather than just summarizing data.
[00:20:16] - [Speaker 0]
Yeah.
[00:20:17] - [Speaker 1]
And it makes me think about a conversation that we had earlier this year. So each year, AlphaSense hosts a client event in New York called Alpha Summit. And this year, some of the discussion was around the fact that so many companies have AI strategies, but very few are able to tie that AI strategy to a direct increase in revenue or profit. And certainly, I think what I'm excited to see over the next year is how that develops, because I believe what we're gonna see is companies being much clearer and more prescriptive in how to use AI, what tools for when to make sure that they're getting a real benefit and getting better decisions out of the AI. So stay tuned.
[00:21:03] - [Speaker 1]
I think there's more to come on that one.
[00:21:04] - [Speaker 0]
Well, I do love a good teaser. Sounds like we are gonna have to get you back on next year as well and find out more about that. But on on a personal level, we are recording this at the end of twenty twenty six. We're starting to think about life in 2027. What what have you learned from all the customer conversations, maybe at that event in person, about everything from workflows where AI is creating immediate value, where human judgment remains impossible to automate?
[00:21:30] - [Speaker 0]
What are you taking away? What have you learned from all those conversations?
[00:21:34] - [Speaker 1]
I am an optimist at heart, so I do typically look on the bright side, but I I do see of... Within business, a real opportunity for businesses to leverage AI while still absolutely needing humans in the loop and humans to be the the the people that are deciding what's actually gonna get done. I'm really excited to see how AI develops and changes what we're doing, but fundamentally, business is about relationships and thoughtfulness and judgment, and we're gonna continue to need those for sure.
[00:22:12] - [Speaker 0]
And I think that is a powerful thought provoking moment to end on. But for leaders listening, maybe they wanna hear more about anything we've talked about today. Indeed, your AI solutions like SuperAnalyst, which is a always on AI agent, Alfasense, how that delivers finance and corporate professionals with real time high value information to automate workflows. There's so much I could talk about there. But for people listening, where should they go?
[00:22:39] - [Speaker 1]
First, you know, if you're interested in learning more about AlphaSense, please go to our website. It's alpha,alpha,-sense,sense,.com. And then please, I'd love to hear from listeners as well. You can find me on LinkedIn. Please send me your thoughts.
[00:22:57] - [Speaker 1]
I strongly believe that as AI becomes more embedded in business decision making, the organizations that succeed will be those that combine the speed of AI with trusted information and human touch. So keen to hear from listeners if they agree, disagree. Let's start the conversation.
[00:23:17] - [Speaker 0]
That's incredibly generous of you. AlphaSense is the AI platform behind the world's most important decisions everywhere from Wall Street to corporate boardrooms around the world. So I will include links to the website, to your LinkedIn. I encourage people to grasp this opportunity, keep this conversation going. We can all learn from each other on here, and more than anything, just thank you for talking about it in a language we can all understand and also with some real world examples.
[00:23:43] - [Speaker 0]
Always makes a a massive difference, but thank you again. Appreciate your time.
[00:23:47] - [Speaker 1]
Thanks, Neil.
[00:23:48] - [Speaker 0]
One of my favorite questions from Julian there was whether AI helped the team spot something it might otherwise have missed. It could be a risk that nobody had priced in, a competitor moving into a new market, or a piece of evidence that really makes you think twice about an acquisition. You could produce twice as many reports and still miss all three. And I think this is an incredibly practical way to start. Pick a decision that your team made recently and look back at the research behind it.
[00:24:19] - [Speaker 0]
Can you trace the most important claim to its source? And did anyone look at the evidence that pointed them that way? Who was responsible for weighing it? And I think these kind of questions are worth asking whether the research took ten minutes or ten days. A big thank you to Julia.
[00:24:38] - [Speaker 0]
Remember, you'll find AlphaSense in the show notes. Gonna be linked to their website, her LinkedIn, and I think we both encourage you to share your thoughts. What would show you that AI has improved a decision that your team has made? Mummy, you can always get me at tech talks network dot com. Send me an audio message, DM, browse through 4,000 interviews, or learn how you can work with me or meet me on the road at an event next year.
[00:25:06] - [Speaker 0]
But that's it for today. So I'll look forward to hearing from you, and I will return back in your podcast feed tomorrow. Bye for now.

