Search Intent Lite

Search intent classifier (lite)

Labels each keyword with its likely intent and the page type that usually matches it, so a blog post is not aimed at a "buy" query (or the reverse).

When to use

• "What's the intent of these keywords?", "should this be a product page or a blog post?", "map keywords to page types".

How to run

python3 scripts/intent.py keywords.txt
python3 scripts/intent.py keywords.csv --brands "acme,acmecorp" --locations "riyadh,dubai"
cat keywords.txt | python3 scripts/intent.py -

Input is one keyword per line, or a CSV with a keyword or query column. Pass your brand names so brand queries are labelled navigational.

Present the table and the totals. Point out any unclear or mixed rows: those need a look at the live SERP.

How it works (and its limits)

Rule-based modifiers ("how", "what" → informational; "best", "vs", "review", "software" → commercial; "buy", "price", "coupon" → transactional; brand + "login" → navigational; "near me" or a city → local). It reads the words, not the SERP. Google can treat a query differently from its wording, so the confidence column is honest about that.

Example

keyword	intent	page type
what is a crm	informational	guide / FAQ article
crm vs erp	commercial	comparison page
crm pricing	transactional	pricing page
hubspot login	navigational	brand / login page

Want more?

• Paid skill that maps keyword → page type and flags existing pages built in the wrong format: SERP Intent Mapper on Agensi
• Hosted version for large keyword lists, by API or MCP: SERP Intent Mapper on Apify
• Free MCP server with a classify_search_intent tool: cygnix-seo-mcp