The distribution layer creators keep ignoring: what a ChatGPT plugin tracker tells us about the next discovery war
Most creators I know are still optimizing for the feed. They obsess over hook retention on TikTok, thumbnail CTR on YouTube, and whether the algorithm is favoring carousels or Reels this quarter. Meanwhile, a quieter discovery surface has been growing underneath all of it: conversational search. When someone asks ChatGPT “what’s a good tool for X,” the answer isn’t a ranked feed — it’s a shortlist of three to five names, chosen by mechanics nobody outside OpenAI fully understands. That’s the world Plugins Radar is trying to map, and while it’s pitched at plugin developers, the underlying lesson applies directly to anyone who publishes content for a living. If you’ve ever wondered why your brand shows up in some AI answers and vanishes from others, this launch is worth studying — not because you’ll buy it, but because it exposes how little visibility we have into the surfaces that increasingly sit between us and our audience.
What problem does Plugins Radar actually solve?
Let me frame this the way a social media operator would. Imagine you run a small SaaS brand. You’ve got a blog, a YouTube channel, a decent LinkedIn presence. You’ve spent two years building organic reach. Then one day you notice your signup volume is flat even though your content metrics look fine. You dig in and discover that a meaningful chunk of your target audience now starts their research inside ChatGPT, and your product simply doesn’t appear in the plugin directory results for the queries that matter to you. You have no dashboard for this. No Search Console equivalent. No way to see who’s beating you or why.
That’s the gap UI Bakery, the company behind Plugins Radar, is trying to fill. According to the maker’s launch post, the tool tracks ChatGPT’s plugin directory — category pages, search results, and individual listings — and surfaces patterns that explain why some plugins get picked and others get skipped. The team claims to have built reports for each of the 5,000+ plugins in the directory, covering search positions with a 7-day history, the plugins competing for the same searches, searches where competitors rank and a given plugin doesn’t, and a listing check against the leaders of its category. You can add your plugin and the rivals you watch, and the tool emails you when one passes you or edits its listing. Topic leaderboards are open to everyone, and there’s a free tier where you sign up and track a plugin to see its full report.
That’s the product. Now here’s why I think it matters beyond the plugin niche.
The pattern that should worry every content team
The maker shared a few observations from their directory tracking that read like a warning shot for anyone doing keyword or SEO work. Roughly three in four top-10 results have the search term in the plugin’s name or tagline. Among apps, that number is more than nine in ten. An app that mentions a term only in its description shows up for it in fewer than one in 100 searches. And a typical search returns only 10 to 20 plugins, so one missing word can keep you off the list entirely.
Read that last line again. Ten to twenty results. Not ten thousand. Not a scrollable feed with infinite pagination. A shortlist. This is the structural difference between conversational discovery and feed-based discovery, and it’s the thing I’d bet most creators haven’t internalized yet. On Instagram or TikTok, a mediocre post can still find an audience through the long tail of the algorithm — it just needs enough watch time or saves to keep getting distributed. On a conversational surface, there is no long tail. You’re either in the answer or you’re not.
How this differs from the tools you already use
If you’re a social media manager, your stack probably looks something like Buffer, Hootsuite, Later, or Metricool for scheduling and analytics, plus Canva or CapCut for asset production, and maybe a repurposing layer like Repurpose.io or Opus Clip to stretch one long video into ten shorts. None of those tools touch conversational discovery. They’re built for the feed era — publishing cadence, engagement rate, UTM tracking, the mechanics of algorithmic distribution inside a platform’s own app.
Plugins Radar sits in a different category entirely. It’s closer to what Ahrefs or Semrush do for traditional search, or what Appfigures and Sensor Tower do for app store optimization. It’s a rank tracker for a surface that doesn’t have an official rank tracker. The closest social-media analog I can think of is the early days of social listening tools — before Brandwatch and Sprout Social existed, brands had no systematic way to know what was being said about them. Same energy here. The directory is public, but the patterns inside it are opaque, and someone is now selling a flashlight.
Why TikTok creators should care more than LinkedIn ones
Here’s a nuance I want to flag, because it’s the kind of thing that separates useful analysis from generic launch coverage. The plugin directory mechanics described in the launch post — name and tagline weighting, short result sets, listing edits as competitive signals — map much more closely to how TikTok’s search and For You discovery work than to LinkedIn’s. TikTok has been pushing search hard for years, and its results are increasingly name-and-caption sensitive. If you’ve ever tested two versions of a TikTok caption and watched one get picked up by search while the other died, you’ve felt a version of this. LinkedIn, by contrast, still rewards dwell time and comment velocity more than exact-match keywords. So if you’re a TikTok-first creator, the “your term needs to be in the name or tagline” finding should feel familiar and actionable. If you’re LinkedIn-first, it’s a useful warning about a surface you probably don’t compete on yet — but might soon, as more B2B research migrates into AI assistants.
What creators and social teams can borrow from this
Even if you never touch a ChatGPT plugin, there are three operational moves I’d steal from this launch.
First, treat shortlist surfaces as a distinct channel. When I scheduled 30 posts across 5 platforms last month for a client, I spent most of my time on cadence and creative. But the posts that actually drove signups weren’t the ones with the best engagement rate — they were the ones that appeared in a Google AI overview or got cited in a ChatGPT answer. That’s a different game. You don’t win it with volume. You win it with clarity: does your brand name, your tagline, your one-line description contain the exact words your buyer would type? If not, you’re invisible to the shortlist.
Second, watch competitor listing edits as a signal. The maker’s own framing — that a rival changing their tagline tells you what they’re about to go after — is a genuinely sharp insight, and it generalizes. When a competitor rewrites their LinkedIn headline, updates their YouTube channel description, or changes their X bio, that’s usually not cosmetic. It’s positioning. I’d bet most social teams never track this systematically. A simple shared doc with weekly screenshots of three competitors’ bios and taglines would give you more strategic signal than another engagement report.
Third, build your own 7-day rank history. Plugins Radar offers a 7-day search position history per plugin. You can approximate this for your own brand across platforms using a spreadsheet and a weekly manual check of the queries that matter to you. It’s tedious. It’s also the kind of thing that catches a slide before it becomes a crisis.
Where the math breaks
I want to be honest about the limits of the analogy, because I don’t want to oversell this. The patterns the maker describes come from the plugin directory specifically — a relatively small, curated, technically-gated surface. Five thousand plugins is a rounding error compared to the number of creators on TikTok or the number of pages indexed by Google. So the exact ratios (three in four, nine in ten, one in 100) almost certainly don’t transfer cleanly to broader conversational search. What transfers is the structural insight: shortlists reward exact-match clarity over long-tail volume. That’s the part I’d hold onto. The numbers are directional, not gospel, and the maker is describing their own tracking, not an independent audit.
Where my judgment says it falls short
A few things give me pause, and I’d rather name them than pretend this is a clean win.
The dependency risk is real. Plugins Radar tracks OpenAI’s directory. If OpenAI changes how the directory works — deprecates it, folds it into a different surface, or stops exposing the same search results — the tool’s core value proposition evaporates overnight. Anyone who lived through the Twitter API changes of 2023 knows how fast platform-dependent tools can become paperweights. I’d want to see the team diversify across multiple AI assistants before I’d build a workflow around it.
The audience is narrow. This is a tool for plugin developers and the teams that ship them. It is not, despite my framing above, a social media management tool. If you’re a creator with no software product, you can’t use it directly. You can only learn from it. I want to be clear about that distinction because launch coverage tends to blur it.
Pricing is not disclosed in the source. The launch post mentions a free tier — sign up and track a plugin to see its full report — but says nothing about paid plans, seat limits, or API access. That’s a real gap for anyone evaluating it for a team. I’d want to see the pricing page before recommending it to a client.
The alerting question is unresolved. The maker ends the post by asking the community what they’d want alerted on first: a competitor’s rank, their listing, or a new plugin in their searches. That’s a good question, and the fact that it’s still open tells me the product is early. Alert design is where tools like this live or die. Too many notifications and people mute them. Too few and they churn. I’d watch how this evolves before committing.
No independent verification. Every statistic in the launch post comes from the maker’s own tracking. That doesn’t make it wrong — it makes it unaudited. For a category this new, that’s expected, but it’s worth flagging. When I see claims like “fewer than one in 100 searches,” I want to know the sample size, the time window, and whether the methodology is reproducible. Not disclosed.
What I’d watch / test next
If you’re a creator or social operator reading this, here’s what I’d actually do this week — not because you need Plugins Radar, but because the underlying shift is real and you’re probably underprepared for it.
First, run a manual audit. Pick five queries your ideal customer would type into an AI assistant. Ask ChatGPT, Claude, and Perplexity each one. Screenshot the answers. Note whether your brand appears, who does appear, and what words those brands use in their names and taglines. Do this monthly.
Second, rewrite your one-line descriptions across every platform with exact-match clarity. If your LinkedIn tagline is clever but doesn’t contain the words people search, fix it. Cleverness is for the feed. Clarity is for the shortlist.
Third, set up a lightweight competitor watch. A shared doc, three competitors, weekly screenshots of their bios and taglines. You’ll spot positioning shifts weeks before they show up in a product launch.
Fourth, if you do ship a plugin or an AI-integrated product, sign up for the free tier of Plugins Radar and track one competitor. Treat it as a research experiment, not a workflow dependency, until the pricing and API story is clearer.
The bigger takeaway: discovery is fragmenting. The feed isn’t dying, but it’s no longer the only door. The creators who win the next three years will be the ones who learn to be legible on shortlists — human, algorithmic, and conversational — not just loud in the feed.
