> Enrich.chat gives AI agents verified work emails and mobiles. To use it from an agent, read https://enrich.chat/agents.md . This is the Markdown version of https://enrich.chat/blog/technographic-data/

# What is technographic data? Examples, sources and how accurate it is

Technographic data lists the tools a company uses. It is inferred from outside signals, not confirmed by the company, so here is how each source works, where it fails and how to get it.

By Enrich.chat Editorial. Updated October 15, 2026.

**Short answer:** Technographic data is a list of the software, hardware and cloud services a company uses, such as its CRM, payment processor or CDN. Vendors infer it from public traces like website code, DNS records and job posts, so a detection is evidence, not proof. The [2024 Web Almanac](https://almanac.httparchive.org/en/2024/cms) detected a CMS on 49% of mobile sites but only 8% of the top 1,000.

## Key takeaways

- Technographic data lists the software and hardware a company uses. Vendors infer it from website code, HTTP headers, DNS records and job posts, and some add partner data or surveys.
- Detection is inferred, not confirmed. The 2024 Web Almanac found a CMS on 49% of mobile sites but only 8% of the top 1,000, and a 2024 USENIX study found Wappalyzer named the right CMS on 60% to 91% of live test sites.
- Of 7,628 open fingerprints, 1,011 (13%) list an implied technology, so a tool can be reported without ever being seen.
- Prices run from free to custom quotes: Wappalyzer Pro is $250 a month and BuiltWith Basic $295, while ZoomInfo and HG Insights publish no prices (checked October 15, 2026).
- Enrich.chat's /v1/company/tech returns a company's technologies with a category for 5 credits, and nothing is used when no match is found.

## What is technographic data?

Technographic data is a list of the software, hardware and cloud services a company uses, such as its CRM, email platform, payment processor and CDN. [6sense](https://6sense.com/blog/technographics/) describes the word as a blend of technology and demographics. Where firmographic data says what a company is, technographic data says what it runs. Our guide to [firmographic data](https://enrich.chat/blog/firmographic-data/) covers the other half.

The word had an earlier life. Forrester's Consumer Technographics® study asks households how they adopt and use technology, and its [2005 benchmark](https://www.forrester.com/press-newsroom/forrester-research-the-state-of-consumer-technology-adoption/) surveyed 68,664 North American households. In B2B, the term now means company technology data. This guide covers that meaning.

A technographic record is keyed on the company's web domain. It names each technology and gives it a category, such as CRM or analytics. Some vendors add dates, a confidence level or a spend estimate. You will also see it called tech stack data, or [install data](https://hginsights.com/blog/why-low-cost-technographic-data-can-undermine-gtm-performance/). Sales and marketing teams use it to see which accounts fit a product, which rival tool an account runs and what to say in a first message.

One point shapes everything below: nobody confirms this data with the company. Every vendor infers it from outside signals. The sections that follow show what those signals are, where they fail and how to read a result without trusting it too much.

## What are examples of technographic data?

Examples are the tools a company runs, grouped by category. The table shows common categories with real products that the open [Wappalyzer-format fingerprints](https://github.com/enthec/webappanalyzer) can detect.

| Category | Example technologies | What the record tells you |
| --- | --- | --- |
| CRM | Salesforce | The sales system your integration or message has to fit |
| Marketing automation | HubSpot, Marketo | That the team runs forms and campaigns on a platform you may connect to |
| Payment processors | Stripe | How the company takes payments online |
| Email | Google Workspace, Microsoft 365 | Which mail platform the company runs |
| CDN | Cloudflare | Which network sits in front of the website |
| Live chat | Intercom, Zendesk | The chat or help desk tool on the site |
| Ecommerce | Shopify | The store platform behind the checkout |
| Authentication | Okta | The identity tool behind logins |

The open list shows how wide the field is. In October 2026 it held 7,628 technology definitions in 109 categories, which we counted in its [technology files](https://github.com/enthec/webappanalyzer/tree/main/src/technologies) and [categories file](https://github.com/enthec/webappanalyzer/blob/main/src/categories.json). The most common tags are ecommerce (803 definitions), marketing automation (548), CMS (500), analytics (430), live chat (378) and CRM (314). None of the 109 category names is ERP, HR, payroll or business intelligence software. That gap matters in the sections on collection and accuracy.

### What fields does a technographic record hold?

Every record has a technology name and a category. Beyond that, vendors differ, and the extra fields tell you how far to trust a result.

- **BuiltWith.** Its [Domain API](https://api.builtwith.com/domain-api) returns a confidence score from 0 to 1, the dates a technology was first and last detected, and an average monthly spend estimate based on the technologies detected.
- **TheirStack.** It rates each signal [low, medium or high](https://theirstack.com/en/docs/data/technographic/how-we-source-tech-stack-data) and shows the first and last dates it found the signal.
- **HG Insights.** It labels each install [current, historical or inferred](https://hginsights.com/blog/why-low-cost-technographic-data-can-undermine-gtm-performance/) and adds estimated contract renewal windows.

**Figure 1: Anatomy of one technographic entry.** An illustrative entry for HubSpot with eight fields in three groups. The core fields are the technology name and its category. The trust signals are a confidence score from 0 to 1, first seen, last seen and install status, as documented by BuiltWith, TheirStack and HG Insights. The estimates are spend and renewal window, from BuiltWith and HG Insights. Enrich.chat returns the two core fields for one domain.

*An illustrative entry for one technology. The fields come from [BuiltWith's Domain API](https://api.builtwith.com/domain-api), [TheirStack's data docs](https://theirstack.com/en/docs/data/technographic/how-we-source-tech-stack-data) and [HG Insights' install labels](https://hginsights.com/blog/why-low-cost-technographic-data-can-undermine-gtm-performance/). The category follows the open fingerprint list.*

Enrich.chat's company tech lookup returns the two core fields for a domain. The last section shows the response.

## How is technographic data collected?

Vendors describe five main routes: scanning websites, reading DNS records, parsing job posts, using partner data and asking people. Most blend several, and none sees everything. A [2022 write-up of ZoomInfo's announcement](https://gzconsulting.org/2022/09/13/zoominfo-expands-its-technographics/) says its pipeline combines more than 20 types of source, including company websites, job postings and customer testimonials.

| Method | What it can see | Where it falls short | Who describes it |
| --- | --- | --- | --- |
| Website scanning | Scripts, tags, cookies, headers and page elements: CMS, analytics, chat, payments, CDN | Front-end tools only, and nothing behind a login | [BuiltWith](https://builtwith.com/faq), [ZoomInfo](https://pipeline.zoominfo.com/sales/technographics) |
| DNS records | The mail provider from MX records, and some SaaS tools from TXT records | Only tools that touch DNS: 140 of the 7,628 open definitions use it | [Open fingerprints](https://github.com/enthec/webappanalyzer), [ZoomInfo](https://pipeline.zoominfo.com/sales/technographics) (email-domain fingerprinting) |
| Job posts | Back-office and data tools named in a role's requirements | Only when a company is hiring, and silent on tools it dropped | [TheirStack](https://theirstack.com/en/docs/data/technographic/how-we-source-tech-stack-data), [ZoomInfo](https://www.zoominfo.com/data-sources) |
| Partner or purchased data | Usage data bought from SaaS, PaaS and IaaS providers, or job posts captured by data partners | Buyers rarely learn whose data it is | [6sense](https://6sense.com/blog/technographics/), [ZoomInfo](https://www.zoominfo.com/data-sources) |
| Surveys and testimonials | Anything a respondent knows, internal tools included | Low response rates, and the answers are self-reported | [6sense](https://6sense.com/blog/technographics/), [ZoomInfo (2022)](https://gzconsulting.org/2022/09/13/zoominfo-expands-its-technographics/) |

### How does website scanning work?

Website scanning matches patterns against what a site shows publicly. The open fingerprint list [says](https://github.com/enthec/webappanalyzer) a long list of regular expressions identifies technologies in HTML, JavaScript variables and response headers, among other inputs. [BuiltWith says](https://builtwith.com/faq) every site gives off signals, and that it indexes the internet the way a search engine does, with no data from third parties. HTTP Archive [runs a fork of Wappalyzer](https://github.com/HTTPArchive/wappalyzer) in its monthly crawl.

A few real fingerprints show the idea. A DNS TXT record containing `hubspot-domain-verification` points to [HubSpot](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/h.json). An MX record of `aspmx.l.google.com` points to [Google Workspace](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/g.json). A script loaded from `js.stripe.com` points to [Stripe](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/s.json), and a `Server: cloudflare` header points to [Cloudflare](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/c.json). Our [tech stack lookup guide](https://enrich.chat/tools/tech-stack-lookup/) walks through each signal type and shows which ones the fingerprints use most.

A bigger catalog is not the same as a more accurate one. BuiltWith says it tracks [127,814 web technologies](https://builtwith.com/plans) from data on 491.9 million domains, while the open list has 7,628 definitions. Counts like these show how much a vendor looks for, not how often it is right.

### How do job posts reveal technology?

A job post names the tools a team works with, including tools no website shows. [TheirStack](https://theirstack.com/en/docs/data/technographic/how-we-source-tech-stack-data) reads each post's title, description and URL and treats a mention as a hint that the company uses the tool. It scores the signal from the number of mentions, where they appear and how recent they are. It does not treat a missing mention as proof that a tool is absent. In ZoomInfo's own example, a role that requires [Salesforce administration](https://pipeline.zoominfo.com/sales/technographics) suggests the company runs Salesforce.

The limits are plain. The same ZoomInfo guide says passive sources lag actual adoption by months and that not every company posts roles publicly. It also says job postings do not tell you when a company retires a tool. This is why website scanning and job posts work best together: the first sees the front end, and the second hints at what sits behind a login.

## How accurate is technographic data?

No single figure answers this, because accuracy depends on the sample, the tool version and what each site chooses to show. Independent tests show website detection misses a lot, and the misses grow with company size. Every accuracy number from a vendor comes from the vendor.

### What do independent tests show?

The Web Almanac, run by HTTP Archive, applies a fork of Wappalyzer to millions of sites. In its [2024 CMS chapter](https://almanac.httparchive.org/en/2024/cms), a CMS was detected on 49% of mobile sites overall but only 8% of the top 1,000 (Figure 2). The authors say larger businesses often use custom CMSs they cannot readily identify and are more likely to hide them. They call it improbable that more than 90% of the top 1,000 go without a CMS and say it is far likelier those CMSs are missing from their data. The Web Almanac's [methodology page](https://almanac.httparchive.org/en/2024/methodology) adds that every detection tool, Wappalyzer included, has limits.

**Figure 2: Share of mobile sites with a CMS detected, by site rank.** A bar chart from the 2024 Web Almanac. A CMS was detected on 8% of mobile sites in the top 1,000, 17% in the top 10,000, 21% in the top 100,000, 27% in the top 1 million, 45% in the top 10 million and 49% of all sites. The authors call it improbable that over 90% of the top 1,000 use no CMS.

*The bigger the site, the less often a CMS is detected. Share of mobile sites by rank, June 2024 crawl, detected with HTTP Archive's fork of Wappalyzer. Source: [Web Almanac 2024, CMS chapter](https://almanac.httparchive.org/en/2024/cms), Figure 12.5.*

A peer-reviewed study found the same gap between the lab and the field. Researchers at Stony Brook University tested six fingerprinting tools, Wappalyzer among them, on lab installs. They then ran four of the tools on [726 live sites](https://www.usenix.org/system/files/usenixsecurity24-kondracki.pdf) known to run WordPress, Joomla, Drupal or MediaWiki. Across the tools, accuracy fell 20% to 80% against the lab results. Wappalyzer fell the least, yet it still lost double digits for three of the four CMSs.

*Share of sites where Wappalyzer named the right CMS, default scan. Source: Kondracki and Nikiforakis, USENIX Security 2024, Table 4.*

| CMS | Lab installs | Live sites | Live sites tested |
| --- | --- | --- | --- |
| WordPress | 100% | 67.5% | 231 |
| Drupal | 100% | 74.0% | 169 |
| Joomla | 100% | 60.2% | 231 |
| MediaWiki | 87.3% | 90.5% | 95 |

Read it for what it is. The study tests whether a tool can name a known CMS, not how well any vendor's database performs. Its authors also say they could not use BuiltWith's lists as ground truth, because BuiltWith itself relies on fingerprinting. That is the core problem with checking technographic data: the answer key is made the same way.

### What do vendors say about their own accuracy?

HG Insights, a vendor, says in its [August 2026 report](https://hginsights.com/wp-content/uploads/2026/08/HG-Insights-Insight-Report_-The-Hidden-Cost-of-a-50-Hit-Rate.pdf) that most technographic data clears a hit rate somewhere between 50 and 60 percent. The report also cites a consulting firm's blind sample of about 100 accounts that scored one provider at 87 to 88 percent. HG calls that a promising data point, not a guarantee, and says accuracy is almost never checked by a disinterested third party. Its advice holds whoever you buy from: ask for the false-positive rate as well as the hit rate, and test on accounts you already know.

ZoomInfo's [technographics guide](https://pipeline.zoominfo.com/sales/technographics) gives no accuracy figure. It says mature providers attach a confidence score to each data point and that you should treat a low-confidence point as a directional signal, not a fact. Wappalyzer's [lookup API](https://www.wappalyzer.com/docs/api/v2/lookup/) shows the trade-off in its settings. It drops low-confidence results by default, and turning that off returns more results with more false positives. It also returns only technologies verified in the last two months by default, and a longer window adds results that are more likely to be sites that no longer use the tool.

### Why is a detection inferred, not confirmed?

A detection shows that something on the public web points to a tool. It does not show that the company pays for the tool, runs it today or uses it for the job you care about. Five gaps explain most errors.

- **Seen is not paid for.** HubSpot's tracking code is available on [all products and plans](https://knowledge.hubspot.com/reports/install-the-hubspot-tracking-code), free accounts included, so a HubSpot script can sit on the site of a company that never bought a plan.
- **Implied is not seen.** A fingerprint can list technologies that a detected tool usually brings with it. The [WordPress fingerprint](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/w.json) implies PHP and MySQL. We counted 1,011 of the 7,628 open definitions, or 13%, with at least one implied technology.
- **Seen is not current.** HG Insights says a [historical install](https://hginsights.com/blog/why-low-cost-technographic-data-can-undermine-gtm-performance/) can be 14 months old with no re-check. ZoomInfo's guide says a tool adopted 18 months ago may already be replaced.
- **Most tools leave no trace.** ZoomInfo says website scanning reaches only front-end tools, not backend systems, internal tools or anything behind a login. [Zylo's 2026 index](https://zylo.com/blog/saas-statistics) puts the average company at 305 SaaS applications, from its own spend and usage data.
- **No detection is not "does not use".** TheirStack says a missing mention does not mean a tool is absent, and the same logic holds for every method here.

**Figure 3: Seen versus implied: how a tool can be reported without being seen.** Three things seen on a page, a script path containing wp-content, a meta tag reading generator WordPress and an X-Pingback response header, lead to a WordPress detection. The WordPress fingerprint then implies PHP and MySQL, which were never seen. Below, HG Insights labels each install as current, meaning confirmed running recently by direct detection, historical, meaning seen in the past and not re-checked, or inferred, meaning assumed from nearby signals and never directly observed.

*A tool can be reported without being seen. WordPress's open fingerprint implies PHP and MySQL, and HG Insights labels the same uncertainty on its installs. Sources: [enthec/webappanalyzer](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/w.json) and [HG Insights](https://hginsights.com/blog/why-low-cost-technographic-data-can-undermine-gtm-performance/).*

## How should you check a technographic result before you use it?

Treat a detection as a lead for a question, not a fact for a sentence in your email. The table sorts common kinds of evidence by how much weight they can carry.

| Evidence | Example | How much weight it carries | What to do |
| --- | --- | --- | --- |
| Seen on the live page | A script from js.stripe.com | Good evidence the page loads the tool | Fine for fit. It still does not prove a paid plan. |
| Seen in DNS | An MX record pointing to Google Workspace | Good for mail and for SaaS tools that ask for DNS records | Fine for the email stack |
| Implied by another tool | WordPress implies PHP | A guess about something never seen | Do not state it as fact |
| Last seen long ago | Last detected 14 months ago | The tool may be replaced | Refresh before you use it |
| Mentioned in a job post | A role asks for Marketo expertise | A hint of use, with a lag | Confirm another way |
| Nothing detected | No result | Says nothing either way | Never write "does not use" |

1. **Check the date.** Prefer detections seen in the last month or two. Wappalyzer's API returns [results verified in the last two months](https://www.wappalyzer.com/docs/api/v2/lookup/) by default.
2. **Test a sample you already know.** Pick 30 to 50 accounts whose stack you know from your customers or your CRM, run them through the vendor and count hits and false positives. HG Insights [recommends a blind sample](https://hginsights.com/wp-content/uploads/2026/08/HG-Insights-Insight-Report_-The-Hidden-Cost-of-a-50-Hit-Rate.pdf), and ZoomInfo's provider guide tells buyers to [request a sample pull against their own ICP](https://pipeline.zoominfo.com/sales/b2b-technographic-data-providers) before signing.
3. **Ask for both rates.** HG Insights says a vendor that cannot give a false-positive rate has not measured the part of accuracy that costs your team the most time.
4. **Combine sources.** A page scan plus job posts covers more than either alone. The guide to [waterfall enrichment](https://enrich.chat/blog/waterfall-enrichment/) shows how to combine providers, and the guide to [B2B data decay](https://enrich.chat/blog/b2b-data-decay/) explains why you should re-check old records.
5. **Say less than you know.** Write "teams on HubSpot often ask us about this", not "I saw you installed HubSpot last week". A wrong detail in a first message costs more trust than a vague one.

## How do sales and marketing teams use technographic data?

Teams use technographic data to pick accounts that fit, find rivals' customers, time outreach and personalize messages. ZoomInfo's guide splits the jobs of three data types: firmographics define the market and ICP, technographics rank accounts by technology fit and stack gaps, and intent times the outreach.

| Use | What you look for | What a vendor says about it |
| --- | --- | --- |
| Account fit | Accounts on the platform your product depends on, such as HubSpot for a HubSpot add-on | Technographics show [technology fit and stack gaps](https://pipeline.zoominfo.com/sales/technographics) |
| Integration readiness | The CRM and email tools in place, before you promise a sync | The open [category list](https://github.com/enthec/webappanalyzer/blob/main/src/categories.json) has CRM and email categories |
| Competitive displacement | A rival product, when it was adopted and when its contract may renew | [6sense](https://6sense.com/blog/technographics/) says good data shows adoption and expected renewal dates |
| Change alerts | Accounts that add or drop a tool | [BuiltWith's plans](https://builtwith.com/plans) include alerts when websites start or stop using a technology |
| Market sizing | How a technology's adoption grows or shrinks | [BuiltWith's plans](https://builtwith.com/plans) include 18+ years of technology trend history |

Treat renewal dates as estimates. ZoomInfo's guide says it predicts them from [adoption patterns and typical contract lengths](https://pipeline.zoominfo.com/sales/technographics), and HG Insights calls its own [renewal windows](https://hginsights.com/blog/why-low-cost-technographic-data-can-undermine-gtm-performance/) estimates.

Technographics find the account, not the person. After the stack check you still need the right contact, which is the job of [data enrichment](https://enrich.chat/blog/what-is-data-enrichment/). To turn technology fit into search filters, read how to write an [ideal customer profile](https://enrich.chat/blog/ideal-customer-profile/). For timing, see the guide to [intent data](https://enrich.chat/blog/intent-data/).

## Where can you get technographic data, and what does it cost?

You can get it free from open fingerprints and public crawls, from BuiltWith or Wappalyzer for $250 to $995 a month, from TheirStack by credit, or by quote from ZoomInfo and HG Insights. All prices below were checked on October 15, 2026, and vendors change them often.

| Source | Price (checked October 15, 2026) | How it says it works | Best for |
| --- | --- | --- | --- |
| [BuiltWith](https://builtwith.com/plans) | Basic $295, Pro $495, Team $995 a month. Free single-site lookups | Indexes the web and tracks [signals](https://builtwith.com/faq); data on 491.9 million domains | Lists of sites by technology, with history. Its [Lists API](https://api.builtwith.com/lists-api) returns websites using a technology |
| [Wappalyzer](https://www.wappalyzer.com/pricing/) | Pro $250, Business $450, Enterprise from $850 a month. Free account with 50 lookups a month | Pattern matching on websites; cached database or live scan. The [lookup API](https://www.wappalyzer.com/docs/api/v2/lookup/) needs Business | Browser research and lead lists |
| [TheirStack](https://theirstack.com/en/pricing) | 1,000 company credits for $109, one-time. Unused credits roll over for 12 months | [Job postings](https://theirstack.com/en/docs/data/technographic/how-we-source-tech-stack-data) | Back-office tools and hiring signals |
| ZoomInfo | [No published prices](https://pipeline.zoominfo.com/sales/how-much-does-zoominfo-cost); most paid plans bill annually | Website source code and job postings via partners, per its [data sources page](https://www.zoominfo.com/data-sources) | Teams already on ZoomInfo. See [ZoomInfo pricing](https://enrich.chat/blog/zoominfo-pricing/) |
| [HG Insights](https://hginsights.com/pricing/) | No published prices; standard term is two years; no free trial | Install data labeled current, historical or inferred, plus IT spend | Large-company installs and spend |
| Open data | The [webappanalyzer](https://github.com/enthec/webappanalyzer) fingerprints are free under GPL-3.0. [HTTP Archive](https://httparchive.org/docs/guides/getting-started/) publishes its crawl as a public BigQuery dataset | You run the detection yourself, or query a monthly public crawl | Developers and researchers |
| Enrich.chat | 10 free lookups a day, then plans from $29 a month. A stack found uses 5 credits | One domain per request, from an agent or code | A company's stack, then its people |

If you also need contacts, compare the vendors in our lists of the [best sales intelligence tools](https://enrich.chat/blog/best-sales-intelligence-tools/) and [best B2B data providers](https://enrich.chat/blog/best-b2b-data-providers/).

## How do you get a company's tech stack from Enrich.chat?

Send the company's web domain to `GET /v1/company/tech`. Enrich.chat returns the technologies found for that domain, each with a category, or `found: false` and no charge. The free tier needs no key and no signup.

```text
Use www.enrich.chat to check the tech stack of each company website in my list. Flag the companies that use HubSpot or Salesforce, mark any company with no result as not found, and do not add any technology the API did not return.
```

Behind that prompt, your agent makes a plain GET request like this one:

```
GET https://enrich.chat/v1/company/tech?domain=acme.com

{
  "found": true,
  "domain": "acme.com",
  "company": "Acme Logistics",
  "technologies": [
    {"name": "Google Workspace", "category": "Email"},
    {"name": "HubSpot", "category": "CRM"}
  ],
  "credits_used": 5,
  "free_lookups_remaining": 9
}
```

The response lists names and categories only. Categories vary by vendor: the open fingerprint list files HubSpot under marketing automation, while the sample above shows it as CRM. Compare tools by name, not by category. The [agent docs](https://enrich.chat/agents.md) list every endpoint, and the same lookup is the `company_tech_stack` tool on the MCP server described on the [install page](https://enrich.chat/install/).

Credits are used only on a match. A stack found uses 5 credits, and a lookup with `found` false uses 0. Checking 1,000 company domains uses 5,000 credits if every domain matches. That fits the Growth plan at $29 a month. Each IP address also gets 10 free lookups a day with no signup, and the free [tech stack lookup tool](https://enrich.chat/tools/tech-stack-lookup/) gives 10 checks shared across the website tools. See [pricing](https://enrich.chat/pricing/) for every plan.

### What does the Enrich.chat lookup not do?

- It checks one domain at a time, so it cannot list every company that uses a tool. For well-known examples, browse the [companies using directory](https://enrich.chat/companies-using/). For full lists, a reverse-lookup vendor such as BuiltWith fits better.
- It returns a plain list with a category for each technology. It has no first-seen dates, confidence score, spend estimate or renewal window.
- Like every source in this guide, its result is a signal to check, not a confirmed fact.

To go from a stack to a buyer, pair it with the [company enrichment API](https://enrich.chat/api/company-enrichment/), which also returns the company profile and the people there. The guide to [Enrich.chat for AI agents](https://enrich.chat/agents/) covers setup.

## Frequently asked questions

### What is an example of technographic data?

A record for acme.com might list HubSpot (marketing automation), Google Workspace (email), Stripe (payments) and Cloudflare (CDN). Each leaves a public trace. The open fingerprints match a [hubspot-domain-verification TXT record](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/h.json) for HubSpot, an [MX record of aspmx.l.google.com](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/g.json) for Google Workspace, and a [script from js.stripe.com](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/s.json) for Stripe.

### What is the difference between technographic and firmographic data?

Firmographic data describes what a company is: industry, employee count, revenue, location and ownership. Technographic data describes what it uses: CRM, cloud host, payment provider and so on. [ZoomInfo's guide](https://pipeline.zoominfo.com/sales/technographics) uses firmographics to define the market and ICP, technographics to rank accounts by technology fit, and intent to time outreach. Read our guide to [firmographic data](https://enrich.chat/blog/firmographic-data/) for the first half.

### How is technographic data collected?

Vendors use five main routes: scanning websites for scripts, headers and cookies, reading DNS records, parsing job posts, using partner or purchased data, and running surveys. [6sense lists](https://6sense.com/blog/technographics/) surveys, data mining and third-party purchasing. [ZoomInfo names](https://www.zoominfo.com/data-sources) website source code and job postings. Most vendors blend several routes, and none can see tools that leave no public trace.

### How accurate is technographic data?

There is no single figure. The [2024 Web Almanac](https://almanac.httparchive.org/en/2024/cms) detected a CMS on 49% of mobile sites but only 8% of the top 1,000, and a [2024 USENIX study](https://www.usenix.org/system/files/usenixsecurity24-kondracki.pdf) found Wappalyzer named the right CMS on 60% to 91% of live test sites. [HG Insights](https://hginsights.com/wp-content/uploads/2026/08/HG-Insights-Insight-Report_-The-Hidden-Cost-of-a-50-Hit-Rate.pdf) says most providers land at 50 to 60 percent, a vendor claim. Ask for hit and false-positive rates, and test on accounts you know.

### Is there free technographic data?

Yes, in limited forms. [BuiltWith](https://builtwith.com/plans) is free for single-site lookups, and [Wappalyzer's free account](https://www.wappalyzer.com/pricing/) includes 50 technology lookups a month. [HTTP Archive's crawl](https://httparchive.org/docs/reference/tables/pages/) is public on Google BigQuery with a technologies column, and the [webappanalyzer fingerprints](https://github.com/enthec/webappanalyzer) are open source. Enrich.chat gives each IP address 10 free lookups a day, including tech stacks.

### What is the difference between technographic data and intent data?

Technographic data lists the tools a company uses and changes slowly. Intent data records research behavior that suggests a company is in the market now. ZoomInfo's guide says intent is [not a technographic signal](https://pipeline.zoominfo.com/sales/technographics) in the strict sense. Use technographics for fit and intent for timing. Our guide to [intent data](https://enrich.chat/blog/intent-data/) covers the second half.

## Sources

1. [webappanalyzer: open Wappalyzer-format technology fingerprints (README)](https://github.com/enthec/webappanalyzer)
2. [webappanalyzer technology files](https://github.com/enthec/webappanalyzer/tree/main/src/technologies)
3. [webappanalyzer categories file](https://github.com/enthec/webappanalyzer/blob/main/src/categories.json)
4. [WordPress fingerprint, src/technologies/w.json](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/w.json)
5. [HubSpot fingerprint, src/technologies/h.json](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/h.json)
6. [Google Workspace fingerprint, src/technologies/g.json](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/g.json)
7. [Stripe and Salesforce fingerprints, src/technologies/s.json](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/s.json)
8. [Cloudflare fingerprint, src/technologies/c.json](https://github.com/enthec/webappanalyzer/blob/main/src/technologies/c.json)
9. [Web Almanac 2024: CMS](https://almanac.httparchive.org/en/2024/cms)
10. [Web Almanac 2024: Methodology](https://almanac.httparchive.org/en/2024/methodology)
11. [Smudged Fingerprints: Characterizing and Improving the Performance of Web Application Fingerprinting](https://www.usenix.org/system/files/usenixsecurity24-kondracki.pdf)
12. [HTTP Archive fork of Wappalyzer](https://github.com/HTTPArchive/wappalyzer)
13. [HTTP Archive pages table reference](https://httparchive.org/docs/reference/tables/pages/)
14. [HTTP Archive: getting started](https://httparchive.org/docs/guides/getting-started/)
15. [BuiltWith plans and pricing](https://builtwith.com/plans)
16. [BuiltWith FAQ](https://builtwith.com/faq)
17. [BuiltWith Domain API documentation](https://api.builtwith.com/domain-api)
18. [BuiltWith Lists API documentation](https://api.builtwith.com/lists-api)
19. [Wappalyzer plans and pricing](https://www.wappalyzer.com/pricing/)
20. [Wappalyzer technology lookup API](https://www.wappalyzer.com/docs/api/v2/lookup/)
21. [How TheirStack sources tech stack data](https://theirstack.com/en/docs/data/technographic/how-we-source-tech-stack-data)
22. [TheirStack pricing](https://theirstack.com/en/pricing)
23. [Why Low-Cost Technographic Data Can Undermine GTM Performance](https://hginsights.com/blog/why-low-cost-technographic-data-can-undermine-gtm-performance/)
24. [HG Insights insight report on technographic hit rates](https://hginsights.com/wp-content/uploads/2026/08/HG-Insights-Insight-Report_-The-Hidden-Cost-of-a-50-Hit-Rate.pdf)
25. [HG Insights pricing](https://hginsights.com/pricing/)
26. [ZoomInfo data sources](https://www.zoominfo.com/data-sources)
27. [Technographic Data: The Complete B2B Guide for 2026](https://pipeline.zoominfo.com/sales/technographics)
28. [Top 10 B2B Technographic Data Providers for 2026](https://pipeline.zoominfo.com/sales/b2b-technographic-data-providers)
29. [How much does ZoomInfo cost?](https://pipeline.zoominfo.com/sales/how-much-does-zoominfo-cost)
30. [ZoomInfo Expands its Technographics](https://gzconsulting.org/2022/09/13/zoominfo-expands-its-technographics/)
31. [Technographics: what they are and how to collect them](https://6sense.com/blog/technographics/)
32. [Forrester Research: The State Of Consumer Technology Adoption](https://www.forrester.com/press-newsroom/forrester-research-the-state-of-consumer-technology-adoption/)
33. [SaaS statistics](https://zylo.com/blog/saas-statistics)
34. [Install the HubSpot tracking code](https://knowledge.hubspot.com/reports/install-the-hubspot-tracking-code)

## Try this prompt

Check a list of accounts for a tool

```text
Use www.enrich.chat to check the tech stack of each company website in my list. Flag the companies that use HubSpot or Salesforce, mark any company with no result as not found, and do not add any technology the API did not return.
```

## Related

- [Firmographic vs technographic data](https://enrich.chat/blog/firmographic-data/)
- [Tech stack lookup](https://enrich.chat/tools/tech-stack-lookup/)
- [What is intent data?](https://enrich.chat/blog/intent-data/)