What do you like best about Visby AI?
AI search visibility matters to me: my buyers increasingly ask ChatGPT, Claude, and Gemini for vendor recommendations. So I tested several GEO tools before settling on Visby. Here's the honest version.
**What disappointed me about other tools**
Most "GEO/AI visibility" tools fall into two traps. Either they overwhelm you with 200+ recommendations of mixed quality (most of which are noise), or they sell snake oil — telling you to install plugins that serve content as Markdown, claiming this helps AI crawlers, when in reality ChatGPT and Claude have their own HTML parsers and don't care.
I lost four days down that rabbit hole with several competing tools before I tested it directly: asked ChatGPT and Claude how they actually fetch pages. The Markdown output was completely irrelevant. Pure waste of time.
**Why Visby is different**
Visby gives you fewer recommendations — and that's the point. The 15 or so tasks I received weren't filler. They were exactly the high-impact items that actually move the needle: missing or generic SERP descriptions, inconsistent titles, outdated H1s that no longer match the product. Stuff a human reviewer would catch.
The Kanban-style task board is genuinely useful. Each task has a clear suggestion, a success criterion, and — this is rare — automatic validation when you mark it done. The tool re-checks whether the change actually went live. For solo operators or teams where someone else implements changes, that's gold.
**Standout features**
**Funnel-stage prompt tracking.** Visby groups tracking prompts by Top/Middle/Bottom of Funnel. So you see whether you're visible when someone is researching ("what is X"), comparing options ("best X for Y"), or making a decision ("buy X for company"). I'd never seen this in another tool, and it immediately reshaped how I think about content gaps. Different funnel stages need different content — Visby makes that operational.
**Multi-LLM coverage.** ChatGPT, Claude, and Gemini all tracked. Many competitors charge significant extra for Claude or skip it entirely. For me Claude matters because it's where my brand currently shows strongest visibility — without that data, I'd be flying blind.
**Content Improver.** Recently added. Paste a URL, get a fully reworked version with rewritten title, meta description (with target keyword auto-detected), and headlines reformulated as questions where appropriate. The diff view is clear — you see exactly what changed and why. The output needs human review (more on this below), but as a starting point it's strong.
**GSC integration.** Pulls clicks/impressions/CTR into the same dashboard as AI visibility data. One context, not five.
**One important caveat — and how I handle it**
The Content Improver sometimes adds FAQ entries about features that aren't actually in my product – so I decided to include it in the next version, great idea. But the lesson is clear: review AI-generated content against actual product reality before publishing.
This isn't really a Visby weakness — it's the nature of LLM-generated content. Visby flags it less than other tools I've tried.
**Verdict**
Visby fills the exact niche I needed: not a Semrush replacement, not an Ahrefs replacement, but the missing layer that tells me how AI engines actually see my brand and gives me a manageable list of high-leverage fixes. After two weeks of daily use, I'd recommend it to any small business or independent consultant taking AI visibility seriously — and especially to anyone tired of tools that prioritize impressive-looking dashboards over actionable insight. Review collected by and hosted on G2.com.
What do you dislike about Visby AI?
I don't see real disadvantages: It has less features than some other GEO tools, but this helps to focus and it's better to get the 20% of tasks done that deliver 80% of the results than to get lost in 50% of stuff that doesn't really matter, or get stuck because of overwhelm, so doing less is good as long as it is the right thing.
What I am missing (but they build new features fast, include the great Content improver three days ago):
**What I'd love to see next**
1. **MCP connector** — so I can have my AI assistant pull Visby data directly during research/writing sessions. Already requested; team is responsive.
2. **WordPress write-back** — improving an article and then having to copy-paste body, meta title, and meta description manually risks breaking layout. A direct WordPress push via application password would close that loop. Already announced on their roadmap as “considering”.
3. **14-day refresh** instead of monthly — even though training data doesn't change daily, two weeks would feel more responsive without being wasteful.
4. **Matomo support** alongside Google Analytics — for users who run privacy-first analytics (EU/DACH market segment).
5. **Reddit/G2 social proof tracking** — partial coverage exists, but deeper integration would help track third-party reviews of my courses. Review collected by and hosted on G2.com.