How to find companies that use a technology, and check the signal before you trust it

This guide is for SDRs, marketers and agencies who sell something that integrates with, replaces or works alongside a specific technology, and need a list of companies that use it. It covers where each kind of technology leaves a public trace, how to check page source and DNS records yourself, what the lookup tools cost, how accurate the signals are, and a worked example from a fleet fuel card company. For the definition and the main data types, see technographics.

Where each kind of technology leaves a trace

The table explains why no single source finds everyone. A website scanner can't see the accounting system, and a job board can't see a chat widget. W3Techs, which surveys the technologies used on websites, notes that sites can hide most of their technologies if they choose to, so the absence of a signal is never proof that a company doesn't use something.

How to find companies that use a technology

1. Define what "uses it" means for your offer. Installed, paying for it, actively using it, a particular version or plan? A company that tried a tool two years ago isn't the same prospect as one running it daily.
2. Pick the signals that fit the technology. Use the table above. For a website tool, start with a lookup tool. For back-office software, start with job postings and the vendor's customer stories.
3. Build a candidate list. Pull companies from the lookup tool, job board searches for the product name, the vendor's public case studies and its partner or integration directory where customers are named.
4. Filter by your customer profile. Industry, size and location still decide whether a company is worth a call. A technology filter only narrows a list well when the technology is uncommon in that market.
5. Check a sample by hand. Open the page source, run a DNS lookup or read the job ad for 20 to 30 companies. If more than a few are wrong, adjust the source before you use the rest.
6. Grade and date every record. For example: strong (a case study, a current DNS record, a job ad from the last six months), weak (an old job ad, a single tag on one page). Record the date you saw it.
7. Use it in outreach as a question, not a claim. "Your dispatcher job ad mentions X. If that's still your system..." is safe. "I see you use X" is embarrassing when the company switched last spring.

How to check page source yourself

For a website tool, the fastest check takes a minute:

1. Open the company's website and view the page source (Ctrl+U on Windows, Cmd+Option+U in Chrome on a Mac).
2. Search the source (Ctrl+F or Cmd+F) for the vendor's name or domain. Vendors' own install guides show what their code looks like. HubSpot, for example, tells customers to paste its tracking code before the closing body tag of each page.
3. If nothing shows, open the browser's developer tools, reload the page and search the Network tab. Scripts loaded by a tag manager, or only after someone accepts cookies, often don't appear in the raw source.
4. Check one inner page too, such as the contact or pricing page. Some tools only load where they're used.

How to read DNS records

DNS records are public, and several of them name the services a company uses. On a Mac or Linux machine, dig MX example.com and dig TXT example.com return them; online DNS lookup tools do the same in a browser.

Two limits. A domain can have only one SPF record, so companies add includes over time and rarely remove them; an include proves permission to send, not current use. And verification records are often left in place after a service is cancelled, so treat them as weak evidence on their own.

Lookup tools and data sources

Prices were checked on each vendor's own page on 26 October 2026.

For technographic and firmographic data delivered into your own systems, Bebop Insights provides B2B data by API or bulk delivery.

How accurate technographic signals are

Two kinds of error matter, and they pull in opposite directions:

• False positives: a tag left on an old landing page after the contract ended, a verification record from a trial, an agency's test install, a job ad that lists a tool as "nice to have".
• False negatives: scripts loaded by a tag manager or only after cookie consent, tools used behind a login, a subsidiary on a different domain, back-office software that leaves no public trace at all.

The base rate matters more than most people expect. Illustrative arithmetic: you have 5,000 accounts in your market, and 4% of them (200) really use the technology. A detector finds 90% of real users (180) and wrongly flags 3% of the rest (3% of 4,800 is 144). Your list has 324 companies, and only 180 of them, 56%, actually use it. Run the same detector on a technology that 20% of the market uses (1,000 companies): it finds 900 and wrongly flags 120 of the other 4,000, so 88% of the list is right. The rarer the technology, and rare technologies are often the most valuable to target, the more hand-checking the list needs.

Worked example: a fuel card company finds telematics users

Illustrative example. The companies, the telematics platform and the numbers are fictional, to show the method.

Tollgate Fuel Card (tollgatefuel.example) has built an integration with a telematics platform called RouteMeter, so fuel purchases can be matched to vehicle location automatically. It wants Texas companies with 20 or more vehicles that run RouteMeter.

Telematics is back-office software. It doesn't appear on a company's website or in its DNS records, so a website scanner returns almost nothing useful. Tollgate's SDR used three other signals instead:

Filtering and grading. Checking company websites and listings for fleet size left 31 companies with 20 or more vehicles. She graded 24 as strong (a case study, or a job ad from the last six months naming RouteMeter as required) and 7 as weak (an older ad, or RouteMeter listed as "a plus").

Outreach. The 24 strong accounts got a message that led with the integration. The 7 weak ones got a question first:

Hi [first name],

Your recent ad for a [role] mentions [technology]. If that's still what your team uses for [job], one question: are fuel purchases matched to [the relevant data] by hand today?

We built a [technology] integration that does it automatically for fleets like yours. Worth 15 minutes to see if it fits?

[Your name]

The job ads turned out to be the best source by far: they named the technology, the team that used it and, often, the problem ("reconcile fuel receipts weekly"). Finding signals like these across thousands of accounts by hand is slow, which is why it pays to decide up front which signal you trust and how fresh it has to be.

Self-check

• You've written down what "uses the technology" means for your offer.
• You chose signals that can actually see this kind of technology.
• You checked 20 to 30 records by hand before using the list.
• Every record has a grade and the date it was seen.
• Your outreach asks about the technology instead of asserting it.

Common mistakes

• Using a website scanner for back-office software. It can't see an ERP or a telematics system. Use job ads and case studies.
• Trusting a single old signal. A tag from 2023 or a verification record from a trial says little about today.
• Ignoring the base rate. For a rare technology, even a good detector produces a list that's half wrong.
• Stating the technology as fact in the first line. If it's wrong, the email is over. Ask.
• Filtering on technology alone. A company that uses the right tool but is the wrong size or industry is still a bad lead.

Next step: turn your graded companies into a working list with how to build a prospect list.