A procurement manager does not search for “metal thing.” She searches for a shop that can stamp heat-treated steel to a drawing, document the alloy, hold the tolerance, carry the right certification and deliver close enough to avoid a freight migraine. Google can supply links. Thomas tries to supply a shortlist. That difference - between finding pages and finding a qualified producer - has kept an improbable New York information company alive since the McKinley administration.
Thomas began in 1898, when Harvey Mark Thomas published Hardware and Kindred Trades. The publication grew into the Thomas Register of American Manufacturers, the “Big Green Books” that occupied entire shelves and earned the trade nickname “the purchasing bible.” By its centennial in 1998, the register had swollen to 34 volumes and more than 60,000 pages, carrying information and advertising from 23,000 companies. It was gloriously useful and absurdly physical - an industrial internet with binding glue.
The job beneath the product
Paper failed first. The directory did not.
Thomas put the register online in 1995. In 2005, after what it described as a transition of more than a decade, the company announced that print would end. The 2006 edition was the hundredth and last. The explanation was refreshingly plain: the internet had become industry's primary information source. Keeping thousands of supplier records current on paper had turned from a feature into a tax.
This was the moment that changed the company's mind, but not its purpose. Thomas did not decide that industrial directories were passé. It decided that annual, alphabetized dead trees were a poor delivery system for living data. The proprietary register database became the foundation of Thomasnet. Search replaced thumb tabs; filters replaced penciled notes; supplier profiles could be updated without waiting for another press run.
The first thing to become obsolete was the container. The customer’s job survived intact.
That distinction is why this is more than a sepia business tale. Companies often describe digital transformation as a new website wrapped around the old organization. Thomas did the harder thing: it identified the durable customer need - reduce the risk and time in selecting an industrial supplier - and allowed the famous object that served that need to disappear.
What Thomas actually sells
Free search on one side, expensive attention on the other
For buyers, Thomas is a free vertical search and procurement workspace. A user can search by product, service, capability, certification or location; inspect company profiles and product catalogs; filter for quality credentials, Made in USA status or diverse ownership; save shortlists; export results; and send requests to several suppliers. Its newer Smart Search accepts detailed requirements in ordinary language rather than forcing a buyer to guess the official category first.
For suppliers, Thomas is a customer-acquisition channel. A machine shop or equipment maker can claim a basic profile, then pay for sponsored placement, verified status, richer content, request-for-quote features and intelligence about the companies viewing its page. Performance-Based Listings let an advertiser choose relevant categories, set a monthly budget and pay when an engaged visitor evaluates the profile. Thomas also operates an industrial marketing agency offering websites, SEO, paid media, content, video and CRM work.
A third line sits between software and agency work: product data. Thomas builds and manages online catalogs, ecommerce experiences and CAD or BIM libraries, then syndicates that information to distributors and other industrial platforms. The company says its marketing arm has built more than 6,000 industrial websites and improved search rankings for more than 3,000 firms. That expertise matters because a valve catalog is not a sneaker store. Engineers need dimensions, materials, drawings and configurations before anyone reaches for a credit card.
Xometry’s 2021 price: about $198.5 million cash and $101.5 million in stock.
Monthly plans begin in double digits; catalog platforms and custom services rise from there.
The $300 million complement
Xometry bought the moment before the quote
In December 2021, public manufacturing marketplace Xometry agreed to buy Thomas for $300 million on a cash-free, debt-free basis - roughly $198.5 million in cash and $101.5 million in stock. At the time, Thomas reported more than 1.3 million registered users, 500,000 commercial and industrial sellers, 45,000 diversity-certified suppliers and more than 20 million annual sourcing sessions. It also said 93 percent of Fortune 1000 companies used the platform.
The fit was almost diagrammatic. Thomas knew who was looking and what they wanted. Xometry could price and route many custom-manufacturing jobs to a production network. Thomas earned primarily advertising revenue; Xometry earned primarily from the spread on parts and assemblies sold through its marketplace. One owned discovery and industrial intent. The other wanted the transaction that followed.
Integration was not magic. Xometry warned investors about the cost and risk of combining the businesses, and its 2022 results did not produce the neat profitability ending forecast when the deal was announced. The parent posted a $76 million annual net loss and cut its workforce by 6 percent in early 2023 as part of an expense reduction. Supplier-services revenue also fell 17 percent year over year in the first quarter of 2024, even while its gross margin improved. A sensible acquisition thesis still has to survive sales cycles, product work and operating expenses.
Why it is different
Google has pages. Thomas has qualifiers.
Thomas competes with general search, GlobalSpec, IndustryNet, Kompass, Alibaba, MFG.com, distributor catalogs and old-fashioned referrals. Its defense is specificity. An industrial buyer does not merely need a company mentioning pressure vessels. The buyer may need an ASME-certified custom manufacturer, in a certain region, working in a certain material, ready for new quotes. Thomas structures those qualifiers and reviews information for its Verified suppliers.
The platform is especially useful when the cost of a bad match is high: aerospace parts, medical components, plant equipment, custom machining, electronics manufacturing or maintenance supplies that can stop a line. It also gives procurement teams filters for ownership diversity and domestic origin, requirements that generic search results rarely express cleanly.
There is an obvious tension. Suppliers pay for priority position, while buyers arrive expecting the best fit. Thomas must make promotion visible and keep underlying information credible; otherwise monetization eats the trust that created the audience. Its claimed and Verified tiers, supplier analysts and performance reporting are attempts to manage that conflict. The model is not neutral, but it can remain useful when commercial ranking does not obscure capability.
Steal this playbook
You cannot copy the century. You can copy the sequence.
Five moves worth borrowing
- Pick a search problem where wrong answers are costly.
- Learn the qualifiers experts use, then turn them into structured fields.
- Make the buyer workflow useful before charging the supply side.
- Capture intent at the moment a buyer narrows, compares and contacts.
- Add services where customers lack the skills to use the platform well.
The sharpest idea to borrow is Thomas's separation of job from format. Ask what customers would still need if your present interface vanished tomorrow. Preserve that asset - in this case, structured supplier knowledge and buyer trust - and be willing to discard everything around it. The second idea is to treat taxonomy as product development. Every useful filter is evidence that the company understands how an expert makes a decision.
The supplier-funded model works because a single qualified industrial lead can be worth tens of thousands of dollars. Thomas showcases customers that attribute six-figure opportunities or $1 million in sales to the network. That does not guarantee the same return for everyone; it explains why a manufacturer can rationally buy visibility while the buyer searches free.
Thomas now wraps a plain-language search box around more than 80,000 industrial categories. A company test reported that Smart Search drove over 15 percent more supplier evaluations than the legacy experience. The tempting story is that a modern interface rescued an old directory. The more accurate story runs the other way: the interface works because generations of industrial classification sit underneath it.
That is Thomas's place in the market. It is not quite a distributor, not merely a media company, and not only a software marketplace. It is an index of North American industrial capability with marketing machinery attached - a place where the search for an unfamiliar supplier becomes a measurable commercial event. The Big Green Books are gone. The thing they knew is still for sale.