Anslut Semrush MCP till AI
Världens mest omfattande dataset om trafik, synlighet och marknaden, direkt i dina AI-assistenter.
Prova Semrush One gratis
Den ledande plattformen för både SEO-auktoritet och AI-synlighet.
Säkra din biljett nu
En dag. Riktiga strategier. Skapat för marknadsförare som agerar för att vinna.
13 oktober 2026London, Storbritannien

Workflow: Turn gaps into publish-ready plans

Classify the top 3 clusters from WF1 Step 2 into refresh (improve an existing page) or create (build a new page from scratch) to route the right action for each.

Add your data

Adding details ensures a relevant result
Save time by automatically applying details and getting ready-to-use prompts.

Prompt

Using the top 3 prioritized content gap clusters and their existing_coverage data: If prioritized content gaps already exist in this conversation, use the top 3 clusters from that output. If no prioritized content gaps are available, first identify and prioritize content gaps for {your-domain.com} in {country}: * If competitor domains are provided, use them directly, up to 5 domains. * If no competitors are provided, identify top organic competitors for {your-domain.com} using Semrush data for {country}. * Identify content topic clusters competitors cover where {your-domain.com} has low or no visibility. * Check whether {your-domain.com} appears in the top 50 results for each cluster's primary keyword. * Prioritize clusters by demand, keyword difficulty, existing coverage, and business fit. * Select the top 3 clusters. Classify each cluster: * refresh: existing_coverage = yes — {your-domain.com} already has a page on this topic * create: existing_coverage = no — no existing page; build from scratch Return ONE table: Columns: * cluster_name * action_type (refresh / create) * existing_page_url (from coverage check, or n/a) * priority_rank * why_prioritized (1 sentence) Limit: 3 rows.

Example Output (illustrative)

- A 3-row opportunity split table — action type (refresh / create), existing page URL if applicable, and a one-sentence rationale per cluster