Why I Built Seeqer: From an Internal Tool to a Public Product
The problem
For the past 10+ years, my team and I have been building NEWTON, our flagship product. Like every other software company in 2025, we’ve watched the way people find products shift dramatically. Google is no longer the only front door. ChatGPT, Perplexity, Gemini, Claude – these are where prospects increasingly start their research. They ask a question, and an AI hands them a shortlist.
That shift created a new, uncomfortable question for us:
When someone asks an AI assistant about tools for Legal Entity Management, does NEWTON show up at all?
For a long time, we didn’t know. And not knowing is a bad place to be when your pipeline depends on it.
Looking for a tool – and not finding one I liked
Naturally, the first instinct was to save time and buy, not build. So I went shopping.
What I found fell into two camps:
- Enterprise “AEO/GEO” platforms with impressive dashboards, sales calls, annual contracts, and pricing that made sense only if you were a Fortune 500 brand monitoring fifty product lines.
- Quick-and-dirty scripts and Chrome extensions that ran a prompt once, screenshotted the result, and called it a day. No history, no scheduling, no comparison across engines, no team access.
Nothing in the middle felt right. I wanted something that:
- Ran the same set of queries across multiple AI engines on a schedule.
- Tracked our position over time, not just a snapshot.
- Detected which competitors kept showing up next to us (and which new ones were creeping in).
- Was cheap enough to run continuously on dozens of queries without flinching.
- Could be handed to a non-technical teammate without a training session.
- Give profound advice on how to improve our visibility.
That tool didn’t exist at a price I was willing to pay. So one weekend I started building it for ourselves.
Building it for NEWTON first
The first version of what would become Seeqer was embarrassingly small: a cron job, a Postgres table, and a handful of prompts hitting the OpenAI and Perplexity APIs. The output was a CSV.
But even that ugly first version answered questions we’d been guessing at for months:
- We weren’t being mentioned for two of the queries we most cared about.
- A competitor we considered less relevant was being recommended constantly by Gemini.
Within a few weeks, the internal tool had become part of how we made marketing decisions. It stopped being a side project and started being infrastructure.
“Wait – other people probably need this too”
Every time I described what we’d built to a friend, the reaction was the same: “How do I get that?”
That kept happening. So at some point the obvious question stopped being “should we turn this into a product?” and became “why haven’t we already?”
The thesis was simple:
- Every B2B company now has the same blind spot we had.
- The tools at the top of the market are overbuilt and overpriced for most of them.
- The tools at the bottom don’t actually solve the problem.
- There’s a real, useful product to be made in the middle – opinionated, focused, and affordable.
So we cleaned it up, gave it a name (Seeqer), built a proper UI around it, added scheduled scans, multi-engine tracking, competitor detection, team access, alerts, and an API – and turned it into a public product.
What Seeqer is today
In short: Seeqer is the tool I wish I could have bought a year ago.
You give it the queries your customers are likely to ask an AI assistant. Seeqer runs them across the major engines on a schedule, tracks your visibility and ranking over time, surfaces who you’re competing against in those answers, and tells you when something changes.
It’s the boring, useful kind of tool – the kind that just sits in the background and quietly answers a question that used to keep me up at night.
What’s next
We’re still actively shaping Seeqer based on how teams use it (including our own – NEWTON is still customer #1). If AI visibility is something you’ve been hand-waving about, I’d love for you to try it and tell me what’s missing.
You can take a look at seeqer.ai.