Author Archive

The chatbot makes tool selection based on what you’re asking for. It analyzes your question and automatically picks the right search tool. The AI automatically understands user intent and applies relevant filters working in tandem with semantic search. For example, asking “find restaurants in New York rated 4.5+” uses semantic understanding to identify “restaurants” from description, while simultaneously applying location and rating filter when searching through database.

🔍 Tool Selection Logic:

1. Search Listings (Included in Free)
– For listings/places/businesses/venues
– Examples: “find coffee shops”, “restaurants in New York”, “hotels under $100”
– Searches: Business directory listings

2. Search Products (Included in Pro)
– For products to BUY/SHOP (if WooCommerce active)
– Examples: “wireless headphones under $100”, “on-sale laptops”
– Searches: E-commerce products

3. Search Universal Content (Included in Pro)
– For general website content
– Searches: Blog posts, pages, documentation, policies, guides

4. Follow-up questions
– Answers from conversation context
– The chatbot remembers previous search results

Additional tools:

5. Listing Details (Included in Free)
AI asked about specific listing can call “get listing details” tool to extract following data from the listing:
a. Listing title and description
b. Category and Listing type
c. Location (address, city, region)
d. Amenities & features
e. Operating hours
f. Pricing & menu items
g. Contact info (phone, email, website)
h. FAQs (questions & answers)
j. Reviews

6. Product Details (Included in Pro)
AI asked about specific listing can call “get product details” tool to extract following data from the listing:
a. Product name and description
b. Price (regular + sale)
c. SKU
d. Categories
e. Attributes (color, size, etc.)
f. Stock status

What filters AI can apply when searching?

For Listings (search_listings):
– location – City, address, region (“New York”, “Downtown LA”)
– price_min / price_max – Price range filtering
– rating – Minimum star rating (1-5)
– date_start / date_end – Date filtering (both for rentals and events)

For Products (search_products):
– price_min / price_max – Price range
– in_stock – Only available products
– on_sale – Only discounted products
– category – Product category filter
– rating – Minimum rating (1-5)

With OpenAI’s $5 credit balance (that’s minimal amount), you can expect approximately following amount of conversations before needing to add more credits.

~ 3000 with GPT-5 Mini
~ 1000 with GPT-5.1

Note: These numbers are approximate.

AI search understands meaning, so it finds relevant results even when the exact words aren’t in your listings. That’s why it can match “romantic dinner spot” with fine dining restaurants or “dog grooming” with pet salons.

But it’s not magic — it works from the text in your listings. If a listing has almost no description, or uses very unusual wording, the AI may miss it. Also, in rare cases, it might return a result that feels related but isn’t exactly what you meant.

In short: it’s far better than keyword search, but like any search, it depends on good, descriptive content.

Yes! That’s available in Pro version.

✅ Runs directly on your WordPress website
✅ You control your OpenAI/Gemini API costs
✅ Your database stays on your server

Note: MySQL database which WordPress uses, lacks native vector search – all similarity calculations run in PHP so search speed depends on your server CPU and number of indexed items.

✅ Optimized for up to a few thousand items/pages/products
✅ No external database indexing. Fast setup. Predictable performance.

Further reading: Learn more about building an AI chatbot for WordPress with RAG and how site content retrieval improves AI support answers.

Follow this checklist:

1. ✅ Make sure your listings have AI embeddings generated.
2. ✅ Make sure your listings have keyword-rich descriptions.
3. ✅ Try the Query Expansion feature.
4. ✅ Lower the Minimum Match Percentage setting.
5. ✅ Clear cache in PurioChat → Developer & Debug Options → Clear Cache. Cached search results may prevent new results from appearing immediately.

AI search understands meaning, so it finds relevant results even when the exact words aren’t in your listings. That’s why it can match “romantic dinner spot” with fine dining restaurants or “dog grooming” with pet salons.

But it’s not magic — it works from the text in your listings. If a listing has almost no description, or uses very unusual wording, the AI may miss it. Also, in rare cases, it might return a result that feels related but isn’t exactly what the user meant.

In short: it’s far better than keyword search, but like any search, it depends on good, descriptive content.

This setting determines how closely a listing must match the search query to appear in results.

Setting range: 0–100
Default: 65

Impact on search results:

• Lower values, 40–60: More results, but some may be less relevant.
• Medium values, 65–75: Balanced relevance and quantity. Recommended.
• Higher values, 80–95: Fewer but highly relevant results.

Example:

• Query: Italian restaurant
• 50% threshold: returns pizza places, pasta shops, Mediterranean food.
• 80% threshold: only returns specifically Italian restaurants.

Query expansion uses AI to understand and broaden search terms for better results.

Example:

• Original query: coffee shop
• Expanded query might include: café, espresso bar, coffee house

Trade-offs:

• ✅ Enabled: better coverage and more results, but not always perfectly relevant. This depends on the minimum match settings.
• ❌ Disabled: faster processing and more literal searches.

After installing the plugin, adding your OpenAI, Gemini, or Mistral API key, and clicking Start Training, the plugin sends structured listing data to the selected embeddings model to create vector embeddings.

Processed listing data may include:

• Listing name
• Category
• Location
• Amenities and features
• Description
• FAQ
• Pricing or menu data
• User reviews, positive only, rating ≥ 3.0

When this happens:

• Manual regeneration via the admin panel
• Daily cron job: listeo_ai_update_embeddings for changed listings
• When a listing is created or updated, if auto-update is enabled

What happens when a user searches?

• Listings data is not sent again when the user searches.
• The user’s search query text is sent to the selected AI provider to generate an embedding vector.
• This vector is compared locally against stored embeddings. No additional listing data is sent during the search.