COMPANY AI API MCP READY

Owler Company Intelligence API

Most company APIs tell you what a company is. This one also tells you who it competes with, by name, with a follow-on profile link for every rival so a second pass maps the whole market. Alongside that: a numeric revenue estimate and a banded range, employee count and band, total funding, acquisitions, industry, SIC codes, headquarters address, and exchange and ticker for public companies. Pay per company record, MCP-ready for Claude and other agents.

MEDIAN LAGon demand

FIELDS35

TOTAL USERS11

MONTHLY ACTIVE4

TOTAL RUNS183

SUCCESS (30D)100.0%

RATINGno ratings yet

LAST MODIFIED2026-11-02

PUBLISHED2026-08

Input parameters

PARAMETER	TYPE	REQ	DEFAULT	DESCRIPTION
companyUrls	string[]	yes	—	One or more company profile URLs, for example https://www.owler.com/company/stripe. A bare company slug such as stripe is accepted as shorthand. Up to 1000 per run, collected in chunks of 20. Malformed entries are rejected locally with a warning instead of being sent upstream. Duplicates are removed before collection, so a list with repeats is not billed twice. One charge per company record returned; an input that yields no record is free. — drives your bill

Output schema

FIELD	TYPE	DESCRIPTION	NULLABLE
result_type	string	company for a collected record, error for an input that could not be collected. Filter on this before doing anything else.	no
companyId	string	Stable identifier for the company on the source directory. Example: 100441 for Stripe.	yes
companyName	string	Registered or commonly used name, usually with the legal suffix, for example Stripe, Inc.	yes
profileUrl	string	The profile page the record came from. Also present on error rows, so you can tell which input failed.	yes
website	string	The company's own website, including scheme and trailing slash.	yes
domain	string	Bare primary domain. The most reliable join key against a CRM or another dataset.	yes
description	string	One or two sentences of prose on what the company does.	yes
industry	string	First listed industry classification, for example Banking, Financial Services and Insurance.	yes
industries	string[]	Every industry classification on the profile. In practice this is usually a single-element array.	yes
founded	integer	Year founded, as a four-digit number.	yes
ownership	string	Private or Public. Reliable in testing, and the cleanest way to split a list before pulling filings for the public half.	yes
status	string	A source-side label. It came back as NEW for every company tested, including decades-old public ones, so do not read it as an operating status.	yes
city	string	Headquarters city.	yes
state	string	Headquarters state, province or region, spelled out rather than abbreviated.	yes
country	string	Headquarters country code. Not consistently ISO 3166-1 alpha-2: US companies come back as USA while others come back two-letter, so normalise before grouping.	yes
zipcode	string	Headquarters postal code, as a string so leading zeros survive.	yes
streetAddress	string	Headquarters street address.	yes
phoneNumber	string	Main listed phone number. Formatting varies by country and is not normalised.	yes
revenue	integer	Estimated annual revenue in US dollars as a number. An estimate, not a filed figure, and the field to use when you need arithmetic.	yes
estimatedAnnualRevenue	string	The same estimate as a banded range, for example $5B - 10B. The field to use when you need to group or segment.	yes
employeeCount	integer	Estimated headcount as a number.	yes
estimatedEmployees	string	Headcount as a banded range, for example 1,000 - 5,000. Note the thousands separator inside the string.	yes
totalFunding	integer	Total disclosed funding raised, in US dollars.	yes
totalAcquisitions	integer	Number of acquisitions the company has made.	yes
totalCompetitors	integer	The source's own competitor count. It does not equal the length of the competitors array; treat the two as separate figures.	yes
competitors	object[]	The named competitor set. Each entry has a name and a profileUrl you can feed straight back into companyUrls for a second pass. Around 30 names come back per company.	yes
ceoName	string	Chief executive, when the profile lists one. Frequently absent: it was empty on every company tested, so treat it as a bonus rather than a column you can rely on.	yes
followers	integer	How many people follow the company on the source directory. A rough attention proxy, nothing more.	yes
exchange	string	Stock exchange for public companies, for example NYSE. Absent for private companies.	yes
ticker	string	Ticker symbol for public companies. Pairs with exchange as the handoff key into market data.	yes
sicCode	string[]	Standard Industrial Classification codes listed for the company, for example 7372.	yes
summary	string	One-line plain-language recap of the record, so an agent can read a company without post-processing.	yes
error_message	string	On error rows only: why this input produced no company record.	yes
error_type	string	On error rows only: machine-readable category, such as CollectionError or MissingRequiredParameter.	yes
fetched_at	string	UTC ISO-8601 timestamp for the row. Run the same list on a schedule and this turns the dataset into a size and funding history.	no

Worked examples

One company — the smallest useful run, and the exact input behind the Owler Company Data API task

{
  "companyUrls": [
    "https://www.owler.com/company/stripe"
  ]
}

Batch, with slug shorthand — a bare slug is accepted, so a spreadsheet column of slugs needs no formatting pass

{
  "companyUrls": [
    "https://www.owler.com/company/figma",
    "salesforce",
    "stripe"
  ]
}

Competitor map, pass two — profileUrl values taken from a first run's competitors array, fed straight back in

{
  "companyUrls": [
    "https://www.owler.com/company/moneytor",
    "https://www.owler.com/company/blockrize",
    "https://www.owler.com/company/duespayment",
    "https://www.owler.com/company/bondaval"
  ]
}

Public companies only — exchange and ticker populate here and stay empty for private firms, so this is the handoff into market data

{
  "companyUrls": [
    "https://www.owler.com/company/salesforce",
    "https://www.owler.com/company/figma"
  ]
}

Monthly watch list — save this as a task, attach a monthly schedule, and fetched_at turns the dataset into a headcount history

{
  "companyUrls": [
    "https://www.owler.com/company/stripe",
    "https://www.owler.com/company/adyen",
    "https://www.owler.com/company/gocardless"
  ]
}

POWER-USER TIP

Map a market in two passes, not one — Run one seed company, collect the profileUrl of every entry in its competitors array, dedupe that list, and run it back through companyUrls. Pass two is where the map actually appears. Budget for it first: roughly 30 names come back per company, so one seed costs about 30 charged records on the second pass.

POWER-USER TIP

totalCompetitors and the competitors array disagree, and that is expected — Stripe returned totalCompetitors of 23 alongside 30 named entries. The count is the source's own figure and the array has its own ceiling. Use the array when you need names, use the count only as a rough signal, and never assert one from the other.

POWER-USER TIP

The competitor set is sharpest on mid-market companies — For a household name the returned set mixes real rivals with small, loosely related firms: Stripe's list included Morgan Stanley next to several companies nobody would call a payments competitor. On mid-market and niche companies it is far tighter. Eyeball the first run before you build a pipeline on it.

POWER-USER TIP

Only company rows cost money — One charge fires per row with result_type of company. Error rows, rejected URLs and empty runs cost nothing, so a list with dead profiles in it is safe to submit. A slug that does not exist comes back as an error row reading "The data service could not return a record for this company." rather than vanishing from the output. The run also reads its own budget cap first and collects only what it can pay for, rather than fetching records it would have to discard.

POWER-USER TIP

Two numbers for revenue, and they are for different jobs — revenue is an integer you can sum and sort; estimatedAnnualRevenue is a band you can group by. Both are estimates rather than filed figures, which is the whole point for private companies, so present them as ranges when the number reaches a customer.

POWER-USER TIP

Read the dataset through its saved views — Append ?view=overview to the dataset items URL for the firmographics table, or ?view=competitors for company, competitor count and the named set. Neither needs a transformation written by hand, and both export to CSV or Excel.

POWER-USER TIP

Join on domain, not on name — companyName carries legal suffixes that vary by record, for example Stripe, Inc. and Figma, Inc. The domain field is the bare host and is the field to match against a CRM.

POWER-USER TIP

The per-record price falls with your platform tier — $0.006 on the free tier, $0.0057 at bronze, $0.005415 at silver, and $0.005144 at gold and above. A 1,000-company run therefore lands between $5.14 and $6.00.

Coverage

fields per company row — including error fields

1,000

companies per run — hard cap, raised per account on request

competitor names per company — observed ceiling on the array

companies per collection chunk — pushed and charged before the next chunk

saved dataset views — overview and competitors

published example tasks — each one a saved, runnable input

Written about this source

Walkthroughs, worked examples and posts about this source.

TASK Owler Company Data API: Revenue, Size, Competitors Ready-to-run example · the full record for a list of companies TASK Find Company Competitors with Revenue and Headcount Ready-to-run example · rivals returned with a size figure attached to each name TASK Look Up Company Revenue and Employee Counts Ready-to-run example · straight firmographics lookup for a batch TASK Track Competitor Revenue and Headcount Monthly Ready-to-run example · saved input plus a monthly schedule TASK Get Company Competitor Data in Claude via MCP Ready-to-run example · the Actor as a tool inside an agent TASK Enrich Company Records in an n8n Workflow Ready-to-run example · firmographics attached to a CRM list on a schedule TASK Map a Talent Market from Competitor Companies Ready-to-run example · follow competitor URLs to find rival employers

Code

curl

curl -X POST "https://api.apify.com/v2/acts/johnvc~owler-company-api/run-sync-get-dataset-items" \
  -H "Authorization: Bearer $APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"companyUrls": ["https://www.owler.com/company/stripe"]}'

Python

from apify_client import ApifyClient

client = ApifyClient("APIFY_TOKEN")
run = client.actor("johnvc/owler-company-api").call(
    run_input={"companyUrls": ["https://www.owler.com/company/stripe"]}
)
for item in client.dataset(run.default_dataset_id).iterate_items():
    if item.get("result_type") != "company":
        continue
    print(item["companyName"], item.get("estimatedAnnualRevenue"),
          item.get("employeeCount"), item.get("totalCompetitors"))

Python - competitor map in two passes

from apify_client import ApifyClient

client = ApifyClient("APIFY_TOKEN")
actor = client.actor("johnvc/owler-company-api")

seed = actor.call(run_input={"companyUrls": ["https://www.owler.com/company/stripe"]})
rivals = []
for item in client.dataset(seed.default_dataset_id).iterate_items():
    for c in item.get("competitors") or []:
        rivals.append(c["profileUrl"])

rivals = list(dict.fromkeys(rivals))          # dedupe, keep order
print(f"pass two will charge for up to {len(rivals)} records")

mapped = actor.call(run_input={"companyUrls": rivals})
for item in client.dataset(mapped.default_dataset_id).iterate_items():
    if item.get("result_type") == "company":
        print(item["companyName"], item.get("domain"), item.get("estimatedEmployees"))

MCP - add to Claude Code

claude mcp add --transport http owler \
  "https://mcp.apify.com/?tools=actors,docs,johnvc/owler-company-api"

MCP

Apify's hosted MCP server loads this source from the URL below as the tool johnvc--owler-company-api. Sign in to Apify when your client asks (OAuth), or send an Apify token as an Authorization: Bearer header.

https://mcp.apify.com/?tools=johnvc/owler-company-api

Changelog

2026-08-08 0.0.8 Rebuilt after a platform-wide dataset type fix; the MCP tool call was re-verified against the hosted server.

2026-08-07 Published on the Apify Store with seven example tasks.

2026-08-05 Pay-per-event pricing set at $0.0057 per company record, with the standard tier ladder down to $0.005144.

What people use it for

map a competitive set from a single company
estimate revenue for a private company
enrich a CRM account list with firmographics
track competitor headcount and funding over time
size sales territories by employee count
market mapping for recruiters
qualify inbound leads by company size
find the ticker and exchange for a company

More sources for Competitor and market monitoring, Lead sourcing and CRM enrichment, Hiring signals and talent intelligence →

Alternatives

Crunchbase Company API — better when you need the funding rounds and the investors behind them, round by roundours PitchBook Company API — private-market financials, valuation and deal history on the same accountsours LinkedIn Company API — a second, independent read on headcount when the estimate mattersours Startup Investor Database — use this instead when the question is who funds a segment rather than who competes in itours Indeed Company API — employer ratings, CEO approval and open-job counts from Indeed for the same companiesours

Owler's own subscription product — genuinely better if you want curated alerts, a news feed and analyst-maintained competitor lists rather than raw rowscompeting

Older Owler scrapers on the same marketplace — cheaper on paper, but no documented field list and a visibly lower run success rate on their store cardscompeting