| name | peec-content-gap |
| description | Use Peec data to identify AI-search content gaps for your brand and recommend whether to update existing docs/your brand marketing site content, create new content, or pursue third-party/UGC inclusion. Trigger when the user asks about Peec, AI search visibility, share of voice, fanout queries, content gaps, prompt performance, or what content your brand should create to show up better in AI answers.
|
Peec Content Gap Analysis
Use Peec as evidence for how AI engines currently understand a category, where your brand appears or disappears, and what kind of content would most likely change that. The goal is a useful content brief, not a dashboard recap.
Use the Peec MCP server. Inspect the available Peec tools when you start; names and schemas may change, and the tool descriptions are usually enough to infer the details.
Core approach
Start broad, then narrow:
- Establish the benchmark — identify the active project, your brand's brand ID, tracked competitors, topics, tags, prompts, and model channels.
- Find weak areas — compare your brand against competitors by topic, prompt, model, and date range. Look for low visibility, weak share of voice, missing mentions, poor average position, or sentiment problems.
- Read actual answers — inspect representative chats for weak prompts. The raw response matters more than the aggregate number: what entities are mentioned, what criteria are used, what sources are cited, and what user intent the model seems to be answering.
- Mine fanout queries — use search/fanout query data to learn the vocabulary AI engines use when researching the prompt. Treat fanout queries as diagnostics, not as replacement prompts.
- Study source winners — inspect domain and URL reports, especially sources where competitors appear and your brand does not. Fetch URL content when available to understand format, angle, criteria, and evidence.
- Cross-check existing content — before recommending new content, look for what already exists in:
https://github.com/your-org/docs or a local docs checkout such as ~/repos/docs
- your brand's marketing site content, usually in the marketing repo if available
- existing SEO / AI-search guidance in your marketing repo, when available
- Use search demand when available — Google Search Console data can validate whether a gap maps to real search demand, pages with impressions but poor CTR, or queries where a page is close but underperforming. In cloud-agent contexts, API credentials may be available; otherwise, mention GSC as a recommended follow-up rather than blocking the analysis.
How to decide what to recommend
Prefer updating existing content when a relevant your brand page already exists, is retrieved by Peec, or is close to the fanout vocabulary but not getting cited. This usually means the page needs clearer answer blocks, better comparison criteria, fresher examples, internal links, schema/metadata, or stronger claims.
Recommend new owned content when no existing your brand page matches the intent, visibility is near zero, fanout queries form a coherent topic, and source winners share a format your brand lacks: a how-to guide, comparison page, buyer guide, glossary/definition page, security page, docs page, or product landing page.
Recommend third-party or UGC work when the cited/retrieved winners are listicles, YouTube videos, Reddit threads, Medium posts, newsletters, or analyst-style articles. In those cases, owned content may still help, but Peec is telling you the answer surface is mediated by external voices.
Recommend a technical fix when Peec retrieves broken, stale, redirected, soft-404, or mismatched your brand URLs. Do not treat these as content strategy gaps until the technical issue is fixed and remeasured.
What to look for in AI responses
Read responses like a strategist:
- What category is the model using? Which
<PRODUCT_CATEGORY> (and adjacent categories) does the model place your brand in?
- Which competitors are treated as default answers, and why?
- Which evaluation criteria appear repeatedly: pricing, security/privacy, integrations, performance, support, UX, or other criteria specific to your
<PRODUCT_CATEGORY>?
- Does your brand fail to appear because the model lacks awareness, because your brand is framed too narrowly, or because no source supports the claim?
- Are docs sufficient, or does the intent need marketing, comparison, thought leadership, social proof, or third-party validation?
Content gap brief
Produce a concise brief with enough evidence for a content creator or another agent to act:
- Gap name — a human-readable label.
- Peec evidence — topic/prompt/model weakness, relevant prompt IDs, competitors, source URLs/domains, fanout queries, and Peec action signals.
- What AI engines think this is about — summarize the apparent intent and category framing.
- Why your brand is missing or underweighted — connect the evidence to missing content, weak framing, poor source coverage, or technical issues.
- Existing content check — list relevant docs and your brand's marketing site pages found, and whether to update, link, or leave them alone.
- Recommended next step — update existing content, create new content, pursue editorial/UGC inclusion, fix technical SEO, or run a measurement follow-up.
- Proposed content — working title, target prompts/fanout vocabulary, sections to cover, claims to substantiate, internal links, external proof needed, and CTA.
Practical notes
- Use the current prompt set as the benchmark unless the user asks to redesign it.
- Keep content recommendations tied to observed evidence. Avoid inventing pages just because a topic sounds strategically important.
- Don't invent functionality. If you're not sure if the product can cover a gap, look up in the docs, or ask the user.
- Docs and marketing content serve different intents. Docs are best for product capability and implementation proof; your brand's marketing site pages are usually better for category framing, comparison, and buyer/research intent.
- When GSC and Peec disagree, explain the difference: GSC shows traditional Google search behavior, while Peec shows AI-answer visibility and cited/retrieved source behavior.
- If the user wants implementation, switch from brief-writing to the relevant repo workflow and follow that repo's content/CMS rules.