Blue Puddles
Blue
Puddles
Not blue oceans. Emerging micro-markets where demand is forming and you’re the only one who’s noticed yet.
Blue Puddles is a market positioning framework for identifying emerging sub-categories in your first-party data — and claiming them before your competitors know the category exists. It’s not a keyword strategy. It’s a market intelligence move that starts in your conversations, not your keyword planner.
What Is a Blue Puddle?
A Blue Puddle is an emerging micro-market inside a larger category — one that surfaces in your first-party conversation data before it appears in keyword volume tools or industry research.
It is a sub-category that’s forming in real conversations — sales calls, lost deals, client proposals — before your competitors have named it or built architecture around it. You catch it because you’re paying attention to what people are actually saying, not what the keyword planner is reporting.
On the page, it functions like long-tail SEO. But the insight that drives it doesn’t come from search data. It comes from patterns in your conversations — and recognizing when those patterns represent a real shift in what buyers want.
The Two-Condition Test
A Blue Puddle is only a Blue Puddle when both conditions are true simultaneously. One condition alone is not enough.
Demand without capability is just a market opportunity you can see but can’t serve. The machine will eventually find the gap between your claims and your proof, and it will hedge. You’ve actively injected doubt into the retrieval layer — which is harder to undo than never having claimed the puddle at all.
Capability without emerging demand is just a service you offer. Not a puddle. There’s no emerging sub-category to claim — only existing competition in an established lane.
The Core Insight: Conversations Precede the Data
Keyword tools measure what people are already searching — which means they measure demand that’s already visible, already competitive, and often already saturated. By the time a micro-market shows up in meaningful keyword volume, the window for positioning is closing.
First-party conversation data captures demand before it becomes formalized search behavior. People describe emerging problems in conversations long before they know how to search for the solution. The pattern shows up in calls first. Then in proposals. Then in lost deals. Then in search volume. Then in competitor positioning.
You want to move at the first signal, not the last.
Proof of Concept
The clearest example of Blue Puddles in practice: a high-growth influencer marketing agency where ads were generating clicks but not converting. Something had shifted in what buyers wanted. Rather than adjusting copy based on assumption, the diagnostic went to first-party data — 15 brands analyzed across 24 sales conversations, using AI to extract what the actual asks were in both 2024 lost deals and 2025 active pipeline.
| Signal | 2024 (Secondary Ask) | 2025 (Primary Ask) |
|---|---|---|
| Performance | Nice to have. Awareness was still the primary goal. | Non-negotiable. Brands wanted influencers to generate revenue, not impressions. |
| Proof | Case studies were viewed, not required. | Test campaigns, measurable outcomes, and validation were table stakes before engagement. |
| Content | Campaign-specific creative. | Reusable asset catalogs — creator content as an ongoing production engine, not a one-time campaign. |
The agency already had receipts for performance-driven influencer work — the proof existed, it just wasn’t visible or machine-readable. The move was targeted: a performance marketing navigation tab, a dedicated page, sub-pages for allow listing and related plays, and modular case studies on each performance page.
No competitors were positioning in that specific sub-category. The window stayed open for several months. That was the puddle — and the agency was ready to claim it because both conditions were true.
Why It Works for AI Retrieval
Broad category terms — influencer marketing, content marketing, digital strategy — are contested, commoditized, and hard for machines to use to differentiate one entity from another. When a query comes in about influencer marketing, the machine has thousands of signals pointing in every direction.
A Blue Puddle sub-category has far fewer signals. When you build architecture around it — dedicated pages, case studies, schema-marked content — and those signals point clearly to your entity, the machine can make a confident match. You’re not competing in a crowded retrieval result. You’re the clearest answer to a specific query.
| Broad Category | Blue Puddle Sub-Category |
|---|---|
| Influencer marketing | Performance-driven influencer marketing |
| Real estate platform | Diaspora property investment platform for emerging Caribbean markets |
| AI visibility consulting | Entity architecture for founder-led businesses hitting the referral ceiling |
| Content marketing | AI-generated content compliance strategy for regulated industries |
Finding and Claiming a Blue Puddle — 5 Steps
When You Don’t Have First-Party Data
New businesses don’t have sales calls to analyze. No lost deal records. No client conversations. That’s not a reason to skip puddle identification — it’s a reason to use Reddit instead.
Reddit is where people speak without a filter. They describe their actual problems, their actual frustrations, what they’ve tried, what failed, and what they wish existed. That language is gold — not just for finding puddles, but for writing the copy that makes your positioning land with buyers who recognize themselves in it.
The method: Install the Keywords Everywhere Chrome extension. Find 3–5 active subreddits in your category — ideally threads with 20+ comments. Use Keywords Everywhere’s one-click Reddit thread analyzer to synthesize what’s being said. Then run a second prompt asking it to lift out pain points and organize them by frequency of mention, most to least.
Read what’s showing up at the top of that list. That’s what the market is actually saying — not what competitors are positioning around, not what keyword tools are reporting, but what real buyers are frustrated about right now. If something is showing up repeatedly that nobody in your competitive set is addressing directly, that’s a puddle forming.
| Platform | How to Pull the Data | Best For |
|---|---|---|
| Install Keywords Everywhere. Open any thread. One-click analysis. No copying needed. | Raw, unfiltered buyer language. Pain points before they’ve been formalized into search behavior. | |
| YouTube | Keywords Everywhere works on YouTube pages too. One-click summary of comments and video content. | How-to frustrations, feature complaints, workaround culture around existing tools. |
| No one-click tool. Copy comments manually from a post. Paste into Claude or ChatGPT for analysis. | Decision-maker language. How buyers describe problems to peers and leadership. | |
| G2 / Capterra reviews | Copy and paste reviews directly. No tool needed. | Post-purchase frustrations — the gap between what was promised and what was delivered. |
Important: Reddit data is third-party signal. It tells you what the market is saying, not what your buyers specifically are saying. Use it to find puddle candidates, then validate with your own conversations before committing to the architecture. First-party data always outranks Reddit when you have it.
Working Answers to Common Questions
These aren’t theoretical questions — they come up in practice. Here are the working answers based on how Blue Puddles has been applied.
| Question | Working Answer |
|---|---|
| How many conversations constitute a meaningful sample? | For first-party data: 15–25 conversations is enough to see patterns emerge reliably. For Reddit: 3–5 threads with 20+ comments each gives you a working signal. The threshold isn’t about volume — it’s about whether the same language is appearing across multiple independent sources. When you start hearing the same phrase from people who don’t know each other, that’s a puddle. |
| Can a puddle be found from industry signals alone, without first-party data? | Yes — Reddit, YouTube comments, G2 reviews, and LinkedIn posts are all viable substitutes when first-party data doesn’t exist yet. The tradeoff is signal quality. Third-party sources tell you what the market is saying. First-party tells you what your buyers specifically are saying. Use third-party to find puddle candidates. Use first-party to confirm them before building architecture. |
| What’s the minimum architecture required to claim a puddle? | One dedicated page and one case study is the minimum. But the page alone isn’t enough — it needs to be connected. Internal links from related pages, a mention in your nav if the puddle is strategic enough, and schema markup that names the sub-category explicitly. The machine needs to find the signal from multiple angles, not just one URL. |
| How do you know when a puddle has matured — when the window is closing? | Three signals: competitors start using the sub-category language in their positioning, keyword tools start showing volume for the phrase, and you start seeing generic content appearing around the topic. At that point the puddle is becoming a lane. You’re not too late — but the easy window is closing. This is when you go deeper rather than broader: more case studies, more specificity, more schema. And you start looking for the next puddle forming underneath the one you just owned. |
| Does claiming puddles accelerate or complicate the MVKG? | Accelerates — when done correctly. A well-claimed puddle adds a new, specific anchor to the Specialization node of your knowledge graph, with its own corroborating content cluster. The machine has more to work with, and it’s more specific, which means more confident matching. The risk is claiming puddles that conflict with your existing entity signals. That’s when complexity enters. The rule: a puddle should extend your specialization, not contradict it. |
