Dataset
Web presence of 1,892 US AV integration companies
An entire sector of operating businesses publishes almost nothing. This is the measurement behind that claim, released so it can be checked.
What it shows
Method
Every company in the set was measured from the public web between 10 and 15 September 2026. Nothing was bought, scraped from a private source, or estimated. Four things were collected: the first capture in the public web archive, the sitemap at its conventional locations, the most recent dated content reachable from that sitemap, and the mail host from the MX record. Site level attributes came from a single request to the home page.
Coverage is uneven and the file says so rather than smoothing it over. Domain age resolved for 1,675 of 1,892 rows. Content dates resolved for 205, because most of these sites have no dated content to find. An early run reported far lower coverage than this: eight parallel workers were being throttled by the archive and the throttling was recorded as absence. The run was repeated with two workers and waiting periods, and coverage went from 24 percent to 88 percent. Not measured and does not exist are different findings, and the first version of this data confused them.
Response time is a single measurement per site, so treat it as an order of magnitude rather than a benchmark.
What is not in it
No company names, domains, email addresses, phone numbers or street addresses. State is the finest location given. Content dates are given as a month rather than a day, because a single publication date can identify a company on its own. Row order is shuffled.
Commercial scoring is also excluded. The original work ranked these companies for a sales process, and that ranking is a business asset rather than a measurement. What is published here is what the public web said on the day it was asked.
Columns
| Column | Meaning |
|---|---|
record_id | Row number. Order is shuffled, so it carries no meaning. |
state | US state or territory, two letter code. |
domain_age_years | Years between the first web archive capture and 2026. |
archive_first_year | Year of the first archive capture. |
has_sitemap | 1 if a sitemap was found at the conventional locations. |
sitemap_url_count | Number of URLs listed in that sitemap. |
has_blog | 1 if the sitemap contained a blog or news section. |
last_content_month | Month of the most recent dated content found. Month, not day. |
mail_provider | Mail host class from the MX record: google_workspace, microsoft_365, godaddy, zoho, other, none. |
site_reachable | 1 if the site answered a request. |
http_status | HTTP status of that request. |
response_ms | Time to first response, milliseconds, single measurement. |
page_weight_kb | Transferred size of the home page. |
has_viewport_meta | 1 if a viewport meta tag was present. |
has_open_graph | 1 if Open Graph tags were present. |
has_favicon | 1 if a favicon was served. |
copyright_year | Year printed in the footer, where one was printed. |
cms | Content management system detected, where one was detectable. |
Use and citation
Released under CC BY 4.0. Use it anywhere, including commercially, as long as it is credited.
Tural, A. (2026). Web presence of 1,892 US AV integration companies. Pixedi. https://alaintural.com/data/av-web-presence-2026/
Found a mistake in it? Tell me and the file gets corrected with a note saying what changed.