How Apurva Luty Built Optimly to Teach AI to Recommend the Right Brands

This story first appeared in Forum Ventures' Founder Stories Series on Substack.‍

▶ Watch Apurva's full founder story on YouTube

Marketing measurement has been broken for a long time. Apurva Luty spent 15 years watching it happen from inside some of the largest consumer platforms in the world — at Microsoft, Meta, and Discord — and saw the same pattern repeat across every team she worked with: you put enormous resources into understanding whether your brand is working, and you almost never get a clear answer.

Opinion surveys took weeks to come back. Social listening tools mostly measured noise. Brand trackers cost tens of millions of dollars to run. None of it gave teams the real, actionable signal they needed. "As a marketer in today's world, there are more surfaces, more things to do, more things to test," Apurva said. "And increasingly it feels like you're not getting very much in return. All the channels feel topped out. It's an incredibly frustrating feeling."

What changed her thinking wasn't a new analytics tool. It was watching ordinary people talk to ChatGPT. Apurva, a consumer researcher by training, noticed something the marketing world was still catching up to: people weren't searching anymore. They were having conversations. And she realized the brand that gets recommended in that conversation wins — whether it ever optimized for that outcome or not.

Forum Ventures-backed founder Apurva Luty built Optimly to close that gap. As of mid-2026, Optimly tracks nearly 30,000 brands across 6 AI surfaces and observes more than 145,000 automated agent visits every week, over 50,000 of them from named AI retrieval agents like ChatGPT and Claude.

How Apurva Luty identified AI as the next brand measurement channel

Apurva was still working at Discord when she started noticing a shift in how consumers were using AI tools. The pattern was specific: people weren't typing keywords. They were describing themselves and their situations in full — their lives, their constraints, their preferences — and asking for a recommendation. "People weren't just searching anymore. They were having conversations: I am a mama of three, I'm really busy, I really care about not buying things that have additives in them," she recalled.

For a consumer researcher, this was significant. "I just thought that was so fascinating," she said. "It was the first time I felt like, okay, this is a step change." The question she started asking was: which brands get recommended in these conversations, and why? And more importantly — could a brand actually influence that outcome?

She saw the analogy in 1992. Coca-Cola and Pepsi weren't the ones racing to build websites. It was the smaller, scrappier operators who saw an opening that everyone else was too comfortable to notice. "It was a great time for you," she said of that era, "if you were an enterprising individual trying to figure out how to break out in the next big thing." She believed AI-driven brand presence was exactly that kind of moment — and that the brands that moved early would be the ones AI agents learned to recommend.

What Optimly does — and why it matters for early-stage brands

Optimly gives early-stage founders and marketers a faster and easier way to measure how AI systems represent their brand — and act on that information. The platform is designed for the solo marketer who doesn't have a research team, a seven-figure measurement budget, or weeks to wait for an insight they're not sure will be useful.

Apurva's approach with clients who come to her overwhelmed isn't to hand them a checklist. "People want me to tell them the tactic, the exact thing they need to do," she said. "I usually turn the question back over to them. What's the story you want to tell? Who's your dream customer?" For her, the tactics only make sense once a brand knows who it is and who it's trying to reach. "I don't want to fish for them," she said. "I want to teach them to fish."

That philosophy is built into how Optimly works — not a set of one-size-fits-all recommendations, but a measurement and optimization system that helps founders understand their own position and what's actually driving the gap.

What the Optimly AI Brand Index tracks

The AI Brand Index grew out of something Apurva noticed in her own site traffic: AI agents were crawling pages that didn't exist, sending unexplained volume her way. "I remember looking at that data and thinking this must be a mistake," she said. "I could just ignore it, or I could go down this rabbit hole and try to figure out what's going on."

What she found became the foundation for the index. She drew the analogy to the 1950s television industry: "Back in the 50s, when people first started buying TVs, no one knew what shows people were actually watching. Nielsen built that instrumentation. Once you start to build that measurement out, you start to give brands and marketers the tools to make more informed decisions. Nobody's sharing the data that would actually help people do that right now, and we're setting out to do that."

As of August 2026, the AI Brand Index covers nearly 30,000 brands tracked across 6 AI surfaces, with more than 145,000 automated agent visits observed weekly — over 50,000 from named AI retrieval agents like ChatGPT and Claude.

How Apurva validated Optimly before leaving Discord

Apurva didn't leave Discord with a pitch deck and a plan. She left with a paying customer. Before she was willing to take the leap, she needed proof that there was a real business — not just an interesting idea. She started building small workflows with AI tools and doing consumer research that would have taken a full team weeks to complete on her own. Over time, she validated the approach enough that she was able to land her first paying customer before Optimly had anything resembling a finished product.

That first customer, combined with a conversation with Kevin Corliss at Forum Ventures, gave her the signal she needed. She felt "this was my opportunity to go try something new that I haven't done before in my career."

Why Apurva chose Forum Ventures over YC, Antler, and HustleFund

Apurva researched her options the same way she advises her clients to approach their brand presence: rigorously and with evidence. She used ChatGPT to investigate VCs and accelerators before applying anywhere, weighing Forum against Antler, HustleFund, YC, and a path with no outside capital at all.

What convinced her wasn't a pitch. It was Kevin Corliss's belief in go-to-market first, product second. "I wasn't actively looking for an investor," she said. "I just wanted to figure out, is there a business here. What sold me talking to Kevin was, 'we'll help you figure that out, but either way you'll leave the accelerator with an understanding of how to build a successful business, whether it's venture backed or not.'"

The decision took longer than it looks from the outside. She spent months weighing the idea, and when she asked people in her network for their honest take, most told her to skip it entirely. "The vast majority said don't do it. You're already an expert in this domain, you don't need an accelerator." But she chose Forum anyway — because Kevin's commitment to being in the weeds with her through the hardest part was exactly what a solo founder needed to hear.

What Forum Ventures provides solo founders in the early stage

Being a solo founder in the first year is hard. Forum had told Apurva that while she was looking for a co-founder, they'd be with her in the weeds. "They followed through," she said. "I don't think I would have built as much conviction as quickly as I did had that not been the kind of support I got."

What Forum's accelerator gave her was structure, peer access, and a go-to-market framework she could actually use. "I wasn't actively looking for an investor," she said. "I just wanted to figure out if there was a business here." Forum's answer was to help her work that out — before worrying about the product — and to stay with her through the process. What she remembers most wasn't the pitch coaching or investor introductions; it was the absence of pressure to raise to a spec she hadn't set herself.

How Apurva built conviction when competitors raised large rounds

The last year hasn't felt like a straight line up. Watching competitors announce large fundraising rounds triggered real doubt. "I remember being in tears," she said, describing the moment a rival announced its Series A. "I was like, oh my God, what am I doing. There's already two companies that have raised all this money. Who do I even think I am that I think I can enter this space."

That feeling followed her through fundraising, where more than half her early conversations ended with investors citing the same worry: the space looked too crowded, too well funded, too late to enter. What kept her going wasn't a plan to out-fund anyone. It was a belief that in a market this new, nobody — not even the best-resourced competitor — actually has an edge yet. "Even the smartest AI researchers don't really understand how these things work," she said. "You can have a bunch of money to throw at the problem, but if you don't know what the problem even looks like, you don't necessarily have that much of an edge anyway."

Apurva Luty's advice for founders on AI brand presence

For founders trying to figure out their AI visibility, Apurva doesn't start with tactics. She starts with clarity. "What's the story you want to tell? Who's your dream customer?" Until you can answer those questions cleanly, she says, optimizing for AI recommendations is premature — because AI agents are retrieving and recommending based on what's already out there about a brand, not based on what a founder thinks is important.

Her broader advice for early-stage founders mirrors how she built Optimly: sell before you build, get one paying customer before you worry about product, and choose your early investors based on whether they'll actually be in the weeds with you when things get hard. "The vast majority said don't do it," she recalls of the feedback she got before joining Forum. She's glad she didn't listen.

Optimly is a Forum Ventures portfolio company. Learn more about how Forum supports early-stage B2B SaaS founders at forumvc.com.