LinkedIn industry codes are the fixed list of sector labels, each with an ID, that LinkedIn uses to place a company in one industry. You see the label on a company page, and advertisers and API users get the ID.
Key takeaways
- The list is closed. A company gets a label from LinkedIn's list, so you cannot invent a new one.
- There are two generations of the list. The current one is called Industry Codes V2, and the older labels still show up in real data.
- LinkedIn's IDs are not NAICS codes. Microsoft's own docs say not to confuse them.
- In LeadOcean you filter with the
industryfield and an exact label from a list of 534 values.
What it is
A LinkedIn industry code is a LinkedIn-defined label and numeric ID that assigns a company to a sector, arranged in a hierarchy of parent and child industries.
LinkedIn publishes the reference table as Industry Codes V2. It lists every valid code with its label, hierarchy path and description, split into Active and Inactive nodes (LinkedIn Industry Codes V2, September 2026).
The Active list has 20 parent groups, such as Education, Financial Services, Manufacturing, and "Technology, Information and Media". Each parent holds narrower children. The same page notes that LinkedIn cannot take requests to add or change industries right now.
How it works
You pick an industry, the platform matches companies carrying that label, and you get the people who work there. The label sits on the company, not on the person.
- A company page carries one industry label from LinkedIn's list.
- That label belongs to a parent group in the hierarchy, such as a software company under a technology group.
- Tools that read LinkedIn data store the label as text or as the ID.
- When you filter, the filter value must match the stored label exactly.
- People inherit the match through the company they work for.
Worked example, with placeholder data. Acme builds invoicing software and its LinkedIn page says "Software Development". A filter for "Software Development" returns Acme and every person at Acme.
A filter for "Accounting" does not return Acme, even though Acme sells to accountants. The label describes what the company is, not who it sells to. That is the most common mistake.
A parent group is wide on purpose. "Technology, Information and Media" covers software, internet platforms, media and more, so a filter on a parent label alone gives you a mixed list. Narrow to the child label that matches your buyer.
The older list matters too. LinkedIn marks retired codes as Inactive on the same reference page. Data scraped years ago can still carry old labels such as "Computer Software" or "Information Technology and Services". Current data carries "Software Development" and "IT Services and IT Consulting".
LinkedIn industry vs NAICS
LinkedIn industry codes and NAICS codes both classify companies, but they are two separate systems. LinkedIn states that its IDs are different from the IDs in NAICS-defined codes.
| LinkedIn industry code | NAICS code | |
|---|---|---|
| Owner | Statistics Canada, Mexico's INEGI and the US Office of Management and Budget | |
| Built for | Ad targeting and profile data | Economic statistics |
| Hierarchy | Parent and child industries | Nested numeric levels |
| Shape of a value | A label and a LinkedIn ID | A six-digit numeric code: two digits for the sector, then subsector, industry group, industry and national industry |
| Good for | Finding companies as they describe themselves | Matching to official or public-sector datasets |
| Count free in LeadOcean | Yes, with the industry filter | Not reliable. We do not publish counts that use it |
NAICS was built by the statistical agencies of Canada, Mexico and the United States, and its codes run to six digits (Statistics Canada NAICS 2022 introduction, October 2026). Pick LinkedIn industry when you target by how a company presents itself. Pick NAICS only when you must match an official dataset. Do not translate one into the other by eye: LinkedIn links a NAICS mapping table from its codes page, so use that.
When it matters
The label you choose changes the list you get. These are the cases where it pays to be exact.
Building a target list by sector
A broad label returns a broad list. "Software Development" and "Computer Software" are different values with different contents. If you filter on one, you miss companies stored under the other.
Check both, and check the neighbours too. A SaaS company can sit under "Software Development", "Technology; Information and Internet" or "IT Services and IT Consulting".
Excluding sectors
Exclusion is as useful as inclusion. If you sell to software buyers and do not want agencies, exclude "Staffing and Recruiting" and "Marketing Services". LeadOcean takes an excludeIndustry list for this.
Cleaning inherited data
CRM exports often carry labels from several eras and sources. Free text such as "SaaS" or "tech" will not match a closed list. Normalise your own column to the exact labels first, then filter.
Reading an industry from an agent
An AI agent that guesses a label will fail silently and return an empty list. Give it the exact value list to search. In LeadOcean that is a free lookup, covered in the Claude Code setup and the VS Code setup.
How LeadOcean handles it
LeadOcean stores company industry as the industry filter, a closed list of 534 values. The list holds both the current LinkedIn style labels and the older ones, so "Software Development" and "Computer Software" are both valid.
Punctuation differs from LinkedIn's page. A label that reads "Oil, Gas, and Mining" on LinkedIn is stored with semicolons as "Oil; Gas; and Mining". Copy the value from the lookup, not from a web page.
The lookup is free and spends no records. In MCP it is list_enum_values with the name industry and a search string. Over REST, GET /v1/enums returns the same sets (LeadOcean OpenAPI spec, September 2026).
Here is a free count. It sends count=true as a query parameter with limit 1, so no records are spent.
curl -s -X POST "https://api.leadocean.io/v1/people/search?count=true" \
-H "x-api-key: $LEADOCEAN_API_KEY" \
-H "Content-Type: application/json" \
-d '{"industry": ["Software Development"], "jobLevel": ["VP"], "country": ["US"], "limit": 1}'The response has meta.total and an empty data array. meta.total is capped at 100,000.
For a worked number, the MCP count_leads tool was run on 1 October 2026 with industry Software Development, jobLevel VP and country US. It returned 40,371 mailable people (verified, catch_all_valid or catch_all), re-checked on 2 October 2026. The count changes when the data refreshes monthly.
To go from a count to a list, use the same filters in POST /v1/exports, or in the app's Exports page with its filter builder and price preview. Email from a name is a separate step, covered in LinkedIn to email. Pro is $499 a month flat, and Free gives 1,000 one-off records (pricing, September 2026). See the pricing page for what is included, or the blog hub for related guides.
FAQ
How many LinkedIn industries are there?
LinkedIn's reference table lists Active and Inactive nodes, and the counts can change. Read the table directly for the current number. LeadOcean's own industry filter holds 534 values, which includes older labels.
Is a LinkedIn industry the same as a NAICS code?
No. LinkedIn says its industry IDs differ from NAICS-defined codes, and the two lists should not be confused. Treat them as separate systems.
Why does my industry filter return zero results?
The value probably does not match a stored label exactly. Look it up with list_enum_values, watch for semicolons in place of commas, and try the older and newer label for the same sector.
Should I filter by one industry or several?
Use several when two labels describe the same kind of company. A SaaS buyer list usually needs "Software Development" and "IT Services and IT Consulting" together. Count each one alone first, so you know which label carries the volume.
Can I filter people by industry?
Yes. The industry is a company property, so you filter people by the company they work for. Combine it with jobLevel, jobFunction and country to narrow the list.
Size any industry before you spend a record
Free to start. No credit card. 1,000 records to spend whenever you like.
Get your free API key →