Most founders who build a competitor to an existing tool do it because they couldn't afford the original. That wasn't my situation. I was paying for Mention.com, using it every day, and it was mostly fine. The problem was everything it couldn't do that I kept pretending didn't matter.
The Honest Version of How This Started
I was running outbound research for a B2B SaaS client in 2022. The job was simple in theory: find companies and buyers who were already expressing pain, already talking about the problem our client solved, and surface them before a competitor did. Mention.com was my starting point. It tracked brand keywords, flagged news articles, caught some social posts. It did what it said on the box.
But "what it said on the box" turned out to be the wrong box. What I actually needed wasn't brand monitoring. It was signal extraction. There's a meaningful difference. Brand monitoring tells you when someone says your name. Signal extraction tells you when someone is about to need you, even if they've never heard of you. Mention.com was built for the first thing. Nobody I could find had built cleanly for the second.
So I started stitching together workarounds. A keyword tracker here, a scraper there, a prompt chain I rebuilt every few weeks when the underlying data changed. It worked until it didn't, and when it didn't it usually failed at the worst possible moment - during a pitch, or when a client asked for proof that the leads were actually qualified. That's when I stopped telling myself I was just using a combination of tools and admitted I was maintaining a fragile system that I'd invented because nothing better existed.
What I Actually Found When I Dug In
The first thing I measured was lead quality, specifically what percentage of "mentions" I flagged as relevant actually converted to a first meeting for our clients. Using the stitched-together approach, that number sat around 11 percent. That sounds low, and it is, but the industry baseline I kept hearing from other growth operators was somewhere between 8 and 15 percent for cold outbound built on keyword monitoring. So we were in range. We just weren't good.
The second thing I measured was researcher time. For one mid-market client, we were spending roughly 14 hours a week on sourcing and qualification before a single outreach message went out. A lot of that time was me or a contractor reading through noise - posts that matched a keyword but had nothing to do with buying intent, news articles about a competitor that weren't actionable, forum threads that were three years old and still getting indexed.
The third thing I noticed was something I didn't have a metric for at first. I kept seeing conversations happening in AI-generated answers - the kind of responses you get from ChatGPT or Perplexity when someone asks a question about a software category. Companies were being mentioned or not mentioned in those answers, and that was starting to shape buyer perception before any traditional search or social signal ever fired. Mention.com had nothing for this. Most tools did. This became what I later started calling AI visibility - whether a brand shows up accurately and favorably when a potential buyer asks an AI assistant about their problem space.
I built a prototype specifically to track AI visibility alongside traditional mentions. When I ran it against the client accounts we were already working, the results were uncomfortable. Two of our clients were being actively misrepresented in AI-generated answers. One was being omitted entirely from category conversations where they had a legitimate claim to be included. None of them knew. None of the monitoring tools they were using were even looking.
That prototype became the core of what MentionFox does now. If you're curious how the feature set compares directly to what Mention.com offers, I wrote up a more detailed breakdown on the MentionFox vs Mention page, including where Mention genuinely wins and where I think we do better.
What I Built Differently and Why
The architectural decision that shaped everything else was this: I decided MentorFox should be opinionated about what a "mention" is worth. Most monitoring tools treat all mentions as data points and leave the qualification to you. That's a reasonable design choice if your users are brand managers who want volume. It's the wrong choice if your users are B2B growth teams who need to prioritize.
So every signal in MentionFox gets scored. The scoring looks at context - not just whether a keyword appears, but what kind of post it's in, what the poster's role and company look like, whether the language suggests active evaluation or casual interest, and whether there's a visible trigger like a funding announcement, a job post, or a public complaint about a competitor. This is what moved our lead-to-meeting rate from 11 percent to consistently above 27 percent across the accounts we tested during the first six months.
The investor research angle came later, almost by accident. A few of our early users were VCs and angels who realized that tracking what founders and operators were saying publicly - before a pitch ever happened - gave them useful signal about how a team thought and communicated. We leaned into that. The same infrastructure that surfaces buyer intent also surfaces company-level narrative, which turns out to be something investors actually want.
I also made a deliberate decision not to chase enterprise scale in the first version. Mention.com and its larger competitors are trying to serve every use case for every company size. I chose to build something that works exceptionally well for B2B SaaS companies between 10 and 250 employees, where the growth team is small, research time is expensive, and the difference between a qualified lead and a wasted hour is real money.
What I'd Tell Someone Evaluating This Category
If you are getting genuine value from a monitoring tool right now, don't switch just because something newer exists. The right question to ask is whether your current tool is telling you things you can act on, or just things you can report. There's a real difference between a dashboard that goes into a weekly update email and a system that changes what your team does on Tuesday morning.
The specific things worth testing are: what percentage of alerts require human triage before they're usable, whether the tool sees anything in AI-generated content about your category, and whether it surfaces buying signals or just brand signals. If your current stack handles all three, you probably don't need to change anything. If it doesn't, that gap compounds over time in ways that are hard to see until you measure them directly.
If you want to see how MentionFox handles AI visibility tracking and lead scoring specifically, here's our pricing page with a breakdown of what's included at each tier. There's a free trial, and I'd rather you test it on real data than take my word for any of this.
If you found this useful, I write about solo-founder distribution, B2B SaaS, and what's actually working in the AI-search era over on my Substack (one post per week, no spam).
I'm building MentionFox - a B2B intelligence suite that combines brand mention tracking with AI-visibility (GEO) measurement, investor research, and outreach automation. There's a free tier and a 5-day trial of Pro at mentionfox.com/pricing.
Top comments (0)