The Search Summary route returns the size and shape of a market without returning any companies. You send the same query you would send to segmentation or supplier_discovery, and you get back how many companies match, where they are, how big they are, what they do, and who they sell to.

The summary works two ways: as a read on a market, and as a step before pulling a list:

• On its own, it answers how big a market is and what it looks like without returning any company data. How large the supplier base for a product is, how a segment splits across countries, company sizes, and industries, who those companies sell to: for market sizing, territory planning, or a feasibility check, the summary is the answer, and no list is needed. Because it never returns a company profile, it is a light way to quantify and explore a market on its own.
• Used before pulling a list, the same call lets you refine a supplier discovery or segmentation query: size the set, spot any off-target results, and tighten the filters, then pull the list once the shape is right.

Typical scenarios

Request rules

• Same request body as the Search Query request. See How to build a search query for composing the filter tree.
• module is required. Allowed values: segmentation, supplier_discovery. Any other module, including company_lookup and company_verification, is rejected with 400: those start from companies you already have, so an aggregation over them describes no market.
• The module defines the matched set. Segmentation and supplier discovery can match different companies for the same filters, so the summary is always computed for the module you send.

{
  "module": "segmentation",
  "query": {
    "and": [
      { "attribute": "company_business_model", "relation": "in", "value": ["Software-as-a-Service"] },
      { "attribute": "company_location", "relation": "in", "value": [{ "country": "DE" }, { "country": "AT" }, { "country": "CH" }], "strictness": 1 },
      { "attribute": "company_employee_count", "relation": "between", "value": [50, 500] },
      { "attribute": "company_estimated_revenue", "relation": "greater_than_or_equal", "value": 10000000 }
    ]
  }
}

Successful responses

A successful call returns 200 OK with a response in three parts: the true count, distributions (one value per company), and frequencies (several values per company).

{
  "total_count": 1842,
  "distributions": { "country": {…}, "region": {…}, "revenue": {…}, "naics_4": {…} },
  "frequencies":   { "business_model": {…}, "core_offerings": {…}, "target_markets": {…} }
}

Rules for every field

• total_count is the true number of companies matching the query for the module.
• All counts are company counts, computed over the full matched set.
• Sorting: top lists are sorted by count, highest first; ties are broken alphabetically by code or label, so the same query always returns the same order. Fixed buckets keep their fixed order.
• Entry keys: every entry has label and count. code is added only where a standard code exists: ISO 3166-1 for country, ISO 3166-2 for region, NAICS for industry.
• The response shape is fixed. Every field is always present. A field is null only when it does not apply (region on a multi-country set) or when aggregations are withheld (small or empty sets; see Edge cases).

Distributions

Each company falls into at most one bar of a distribution. Top-10 fields (country, region, NAICS) add up exactly to total_count, because other holds everything outside the top 10, including companies whose value is not known. Fixed-bucket fields (revenue) have no other; companies without a value are left out, so their buckets can add up to less than the total.

Country

Where the matched companies are headquartered.

• Based on: the company's HQ country.
• Format: top = up to 10 entries {code, label, count}, plus other. code is ISO 3166-1 alpha-2; label is the English country name.
• other: all remaining companies, including those with no known HQ country.
• Always present. A single-country set returns one entry and other: 0.

"country": {
  "top": [
    { "code": "US", "label": "United States", "count": 610 },
    { "code": "DE", "label": "Germany",       "count": 188 }
  ],
  "other": 212
}

Region

Where the companies sit inside one country, for territory planning.

• Based on: the company's HQ state or region.
• When it is returned: only when every matched company has its HQ in the same country. Otherwise region is null, whether the single country comes from a filter or the query just happens to land in one country.
• Format: top = up to 10 entries {code, label, count}, plus other. code is ISO 3166-2 (for example US-TX); label is the region name.

"region": {
  "top": [
    { "code": "US-TX", "label": "Texas",      "count": 214 },
    { "code": "US-CA", "label": "California", "count": 198 }
  ],
  "other": 668
}

Revenue

How big the companies are, by estimated annual revenue in USD.

• Format: buckets = always the same 5 entries {label, count}, in the order below, including buckets with count: 0.
• No other: companies without a revenue estimate are not counted.

"revenue": {
  "buckets": [
    { "label": "<$1M",      "count": 612 },
    { "label": "$1M–10M",   "count": 588 },
    { "label": "$10M–100M", "count": 301 },
    { "label": "$100M–1B",  "count": 92 },
    { "label": "$1B+",      "count": 21 }
  ]
}

NAICS (4-digit)

What industries the matched companies are in. This is the main way to spot off-target results: an unexpected industry in the top 10 usually means a filter is too broad.

• Based on: the company's primary NAICS code, cut to its 4-digit industry group. Only the primary code counts, so each company appears once.
• Why 4-digit: 6-digit codes split even small sets into many tiny bars; 2-digit codes are too coarse (most results would be "Manufacturing").
• Format: top = up to 10 entries {code, label, count}, plus other. label is the official NAICS title of the 4-digit group.

"naics_4": {
  "top": [
    { "code": "3399", "label": "Other Miscellaneous Manufacturing", "count": 402 },
    { "code": "3152", "label": "Cut and Sew Apparel Manufacturing", "count": 288 }
  ],
  "other": 181
}

Frequencies

A frequency answers "how many of the matched companies have this value". A company with several values is counted in several bars, so the bars can add up to more than total_count. A company that is both Manufacturing and Business-to-Business counts in both bars.

Rules for every frequency

• Format: top = up to 10 entries {label, count}, plus more_values.
• Counted per company, not per mention. A company that lists the same value three times counts once.
• No other. A company can have values inside and outside the top 10, so no single number describes "the rest" honestly.
• more_values is the number of distinct values beyond the top 10. It is a count of values, not of companies.

Business model

How the matched companies make money and whom they sell to.

• Based on: Veridion's business model taxonomy, 25 labels. A company can have several.
• label is the taxonomy label exactly as written, which is also the exact value the filter accepts, so it can be passed straight into the company_business_model filter (for example not_in: ["Retail"] to drop retailers from a supplier search).

"business_model": {
  "top": [
    { "label": "Business-to-Business", "count": 1402 },
    { "label": "Manufacturing",        "count": 1288 }
  ],
  "more_values": 4
}

Core offerings

What the matched companies actually make or sell. This is the closest check that the query is catching the right companies: "work boot" in a glove search is a sign the query is too broad. Based on the company's core offerings (free text).

"core_offerings": {
  "top": [
    { "label": "industrial glove", "count": 612 },
    { "label": "safety glove",     "count": 498 }
  ],
  "more_values": 1286
}

Target markets

Who the matched companies sell to: the industries and customer types they serve. Most useful for segmentation. Based on the company's target markets (free text).

"target_markets": {
  "top": [
    { "label": "construction", "count": 688 },
    { "label": "healthcare",   "count": 541 }
  ],
  "more_values": 412
}

Edge cases

The response shape never changes; only the values do.

Error responses

• 400 Bad Request, where module is missing, or is not segmentation or supplier_discovery.
• other standard HTTP error responses, detailed in the Response codes section.

Any error response follows a specific structure. Please see the Error response format section for more details.