No cookies here. This site doesn't use trackers, analytics or ad pixels.

Works with Claude, GPT or any LLM · REST · MCP · LangChain · n8n

Send us the catalog. Save your model for the thinking. Don't waste time or tokens on grunt work. We already mapped the ways shoppers look for your products.

What it is: an API built on a map of the market from the shopper's side. We processed over a billion product listings and variants and linked them to search keywords and buyer questions. Send a product title and get JSON back with who buys it, when, where, what for, why and how, plus the search phrases, questions and categories that lead shoppers to it. Send up to 500 titles at once to group a catalog. 25 tokens per field, with free tokens every day.

Reach Dog tokens are our billing units, not model tokens.

The demo opens huggingface.co. No key and no signup needed.

Request
curl https://api.reach.dog/v1/title-matches \
  -H "Authorization: Bearer $REACHDOG_API_KEY" \
  -H "Idempotency-Key: $REACHDOG_REQUEST_ID" \
  -H "Content-Type: application/json" \
  -d '{"title": "Nordic Kayaks NK K1 Kayak",
       "parameters": ["synonyms", "use_cases",
         "search_volume", "buyer_questions"]}'
Response (excerpt)
{
  "synonyms": ["kayak boat"],
  "use_cases": ["Recreational paddling", "Lake exploration", ...],
  "buyer_questions": [
    {"text": "kayak for beginners", "parent_search_volume": 12516, ...}, ...
  ],
  "search_volume": [
    {"text": "kayaks for sale", "search_volume": 40500,
     "context": {"market": "US", "currency": "USD", "provider": "Google Ads",
       "measurement_period": "average of the last 12 months", ...}},
    {"text": "river kayak", "search_volume": 2400, ...}, ...
  ]
}

4 fields, 100 tokens. Values from recorded examples in the title parameter docs; the search volume rows were recorded live on 16 November 2026.

Who, when, where, what for, why and how, as JSON.

Your title says what the product is. Shoppers start from their own life: who it is for, the season or the occasion, where they are going, the job it does. Ask for these fields and you get that side of the product back as JSON, ready to drive copy, collections, ads or an assistant.

The shopper's questionFieldRecorded example: Nordic Kayaks NK K1 Kayak
Who buys it?
  • audiences
"kayakers", "paddlers"
When do they need it?
  • seasons
  • occasions
"Spring", "Summer", "lake outings", "river paddling"
Where do they use it?
  • where
"Lakes", "Rivers"
What do they use it for?
  • use_cases
"Recreational paddling", "Lake exploration"
Why do they want it?
  • why
"outdoor recreation", "water exploration"
How do they use it?
  • how
"paddle across lakes and rivers"
What is it, exactly?
  • product_type
  • attributes
  • missing_attributes
{"name": "Model", "value": "NK K1"}
How does it compare?
  • segment_attribute_distribution
  • comparable_listings
market research request
"inflatable" 172 listings, "lightweight" 15

Occasions cover the events, trips and holidays a product suits, and seasons can be empty for products with no season. How these are made: our model writes them from your title, guided by the tags stored on the closest product neighborhoods we match. Who, when, what for and categories have stored tags behind them; where, why and how are written from the title alone. They are model output, not measured counts, so check them before you publish them. Attributes are different: they come only from what you send, and nothing is inferred. "How does it compare" counts how the product type's listings split by modifier, color, material, season, use and occasion, so you can see which of your product's traits are common and which are rare. Examples are from the title parameter docs.

Shoppers never type your product title. Here's what they do instead.

They search a phrase, ask a question, shop for an occasion or browse a category. We took the market's catalog and worked out those routes, so one title gets you each of them.

What they type into search

The phrases shoppers search for products like yours, with search volume and cost per click.

  • search_phrases
  • long_tail_keywords
  • search_volume
  • cpc

The questions they ask

Buyer questions mapped to your product, the kind people put to search engines and AI assistants.

  • buyer_questions
  • intent_phrases

What else they call it

The plain product type and the other names shoppers use for it.

  • product_type
  • synonyms

Where they are in the buying journey

Each keyword placed at its stage of buying, with the reason.

  • keyword_intent_stages

Where stores file it

Shopping categories, a Google product category path, collections and filters.

  • categories
  • category_paths
  • collection_suggestions
  • facets

What they buy with it, or instead

Products bought alongside it and the alternatives a shopper compares it with.

  • related_product_types
  • alternative_product_types

How they say it in other markets

Suggested search terms for another language or region, for a native speaker to review.

  • localized_terms

What your listing does not say yet

The facts shoppers look for that your title leaves out.

  • missing_attributes
  • variant_dimensions

How each field is made: search volume and cost per click come from keyword data, and each figure names its market, period and provider when they were recorded; treat a figure without one as unverified. Keywords and buyer questions are picked from our stored data. Intent phrases, category paths and translations are suggestions, and the docs label every one. Buyer questions are modeled phrasings, not search logs.

Good. Keep using it. Just don't make it do this part.

Your LLM is great at reasoning over data, but it can't measure how many people search a phrase. Get the market data from Reach Dog, then let your model do the thinking: fewer passes, fewer re-prompts and market data in place of guesses.

01

We already did the market work

Your LLM can guess how people shop for a kayak. We looked: over a billion listings and variants grouped into 29.6M product neighborhoods, linked to 101.3M keywords and 70.5M buyer questions.

02

Numbers it cannot make up

Search volume and cost per click come from keyword data. Each figure names its market, period and provider when they were recorded; treat a figure without one as unverified. Attributes come only from the title and any description or attributes you send, and missing_attributes lists what the title leaves out.

03

The same shape every time

Every field has a fixed, documented JSON shape, so your code and your prompts can rely on it. No parsing free text, no re-prompting when the format drifts.

04

A file you keep

Results come back as JSON keyed by your own product IDs. Save it in your project and use it in every prompt after that. Replays with the same key are free.

  1. Send your titlesOne title, or a catalog of up to 500 in one call.
  2. Get a JSON file backThe ways shoppers look for a product, with the numbers behind it, or your catalog grouped.
  3. Hand it to your LLMClaude, GPT or Grok starts from market data and goes straight to the thinking.

68 fields. Ask only for the ones you need.

Every request lists the fields it wants, and you pay for those fields only. Here are some of them.

Title38 fields

POST /v1/title-matches

What a product is and how people search for it.

  • product_type
  • brand
  • attributes
  • missing_attributes
  • category_paths
  • collection_suggestions
  • facets
  • search_phrases
  • buyer_questions
  • search_volume
  • cpc
  • related_product_types

Research14 fields

POST /v1/market-research

The market around a product type.

  • demand_trend
  • geographic_demand
  • market_coverage
  • brand_presence
  • seller_concentration
  • comparable_listings

Content10 fields

POST /v1/content-generations

Content written from a result you already have.

  • content_brief
  • content_drafts
  • ad_topics
  • blog_topics
  • video_topics
  • first_test

Catalog

POST /v1/batches

Your whole catalog at once, up to 500 titles.

  • catalog_summary
  • catalog_groups
  • group_members
  • ungrouped_items
  • catalog_keyword_opportunities
  • catalog_content_ideas

10 ready-made workflows to start from.

Each recipe combines the right requests for a job. You load your titles, see the full request plan, and nothing is charged until you run it.

  1. 01

    Demand-space lenses

    Turn bare titles into overlapping product groups, each with the reason its products are in it.

  2. 02

    SEM ad campaign deck

    Keywords with volume and CPC, ad angles from stated facts, and a first test per product.

  3. 03

    Social and video launch kit

    A 30-day content calendar across channels, each slot tied to its product.

  4. 04

    Shopify and Algolia proposals

    Collection memberships and search records mapped to your IDs. Export only.

  5. 05

    Market whitespace memo

    Keyword demand set against how many listings already chase each term, with every line linked to its evidence.

  6. 06

    Shopping assistant grounding

    One evidence document per product, plus instructions so your assistant answers only from it.

  7. 07

    Cross-border expansion

    Suggested terms in each target language for a native speaker to review, next to English search demand by country. The two stay separate.

  8. 08

    Cart bundles and cross-sell

    Product pairs from your own catalog, with the reason for each pair.

  9. 09

    Supplier feed auditor

    Find missing fields and write the supplier questions. It never fills in values.

  10. 10

    Seasonal calendar and AEO

    Monthly demand next to weekly supply, plus the questions buyers ask, ready for a buying guide.

Use it from the tools you already have.

REST API

Call it from any language. Python and JavaScript quickstarts are ready to copy.

POST api.reach.dog/v1/title-matches

MCP server

Ask Claude Desktop or any MCP client to look products up for you.

npx --yes @reachdog/mcp@0.1.0

LangChain

Give your agent product tools with a budget and an approval step.

pip install langchain-reachdog==0.1.0

n8n

Audit a product feed from a Google Sheet, with an approval stop before it spends.

n8n.io/workflows/22575

Hugging Face

See real recorded requests and responses. No key, no signup, no tokens spent.

ReachDog/product-classification-api-demo

Pay per field. See the price before you spend.

25tokens

per field, per title. Pick only the fields you need.

1price

for a catalog output, whether you send 10 titles or 500.

0to start

Free tokens every day, so you can try it on your own titles.

Reach Dog tokens are our billing units, not model tokens. They are separate from what you pay Anthropic, OpenAI or any other model provider.

  • Catalog jobs show you a quote and wait for your OK before anything runs.
  • Repeating a request with the same key returns the saved answer at no charge.

Built from market data, and labeled by how each field is made.

1B+
Product listings and variants
29.6M
Product neighborhoods
70.5M
Buyer questions
101.3M
Search keywords

Source: Reach Dog's commerce map, built from product listings that online stores publish, which we collect and process, plus keyword data. Counts last reviewed November 2026. Keyword figures such as search volume and CPC name their market, period and provider when they were recorded.

Ask your AI to fact-check this page.

Everything here is in our public docs and code. Point Claude, GPT or Grok at these and ask whether Reach Dog returns who, when, where, why and how for a product, what each field is based on, and whether it fits your project.

We did the grunt work. Send us your titles.

Create a key and use your free daily tokens. Want to look first? The Hugging Face demo needs no signup.

Run a store? Reach Dog for merchants