85% of manufacturing buyers use vendor and supplier websites when they research a complex, high-value purchase, more than any other source, according to RH Blake and Industrial Equipment News's 2025 study of more than 250 manufacturing executives and managers. Only 23% go to a vendor's website first, though. 45% start with a search engine, so the site has to be found before it can be read.
This page collects the dated numbers on the industrial buyer journey: how much research happens before a supplier hears from the buyer, how many buyers want to do it alone, which sources they use, and what the website has to carry. It closes with a five-page test order for a manufacturing site. Most figures come from agencies, publishers and software vendors with an interest in the result, and each is labelled.
Where the manufacturing numbers come from
Most buyer journey surveys pool every B2B category, and the best known ones lean towards software. G2 and TrustRadius survey software buyers only, so we use them here as a contrast, not a benchmark. Our B2B SaaS Buyer Journey Statistics 2026 post covers that side in full.
Two surveys ask industrial buyers directly. RH Blake, a manufacturing marketing agency, and Industrial Equipment News, a trade publisher, surveyed more than 250 manufacturing executives and managers who decide on offerings with long sales cycles. GlobalSpec, which sells industrial advertising, runs the State of Marketing to Engineers survey with TREW Marketing and Elektor; its ninth edition in 2026 has more than 1,000 respondents worldwide. Both organisations benefit when buyers research online, and both publish only summary figures without margins of error.
Research time: how far buyers get before the first call
Industrial buyers do a lot of reading alone, but they make contact earlier than the software headlines suggest.
| Point of first contact | Share of buyers |
|---|---|
| Within the first 10% of the journey | 19% |
| Between 10% and 25% | 33% |
| Between 26% and 50% | 24% |
| Between 51% and 75% | 12% |
| More than 75% of the way through | 6% |
| Only after the final decision | 6% |
Source: RH Blake and Industrial Equipment News, B2B Manufacturing Buyer Journey Study, 2025, more than 250 manufacturing executives and managers
Three in four buyers reach out before the halfway point. For a capital purchase, that usually means the specification is still open when the first conversation happens, and the supplier who gets that call can still shape it.
The averages from the wider surveys sit later in the journey. GlobalSpec's 2026 engineers put the share completed online before speaking to sales at 62%, with younger buyers waiting longer: 66% for those aged 35 and under, 63% for 36 to 55 and 62% for 56 and over. 6sense, which sells buyer-intent software, found first contact at 61% of the journey across all categories, about 27 weeks in.
| Measure | Figure | Source and sample |
|---|---|---|
| Average B2B buying cycle, all categories | 10.1 months, down from 11.3 | 6sense, nearly 4,000 buyers, 2025 |
| Physical goods cycle against software | 2.8 months shorter | 6sense, 2025 |
| Physical goods cycle against services | 4.8 months shorter | 6sense, 2025 |
| Point of first contact, all categories | 61% of the journey, down from 69% | 6sense, 2025 |
| Share of the process done online before sales, technical buyers | 62% | GlobalSpec, 2026, n = 709 |
| Contact a vendor before the halfway point, manufacturing | 76% | RH Blake and IEN, 2025 |
Sources: 6sense 2025 B2B Buyer Experience Report; GlobalSpec, TREW Marketing and Elektor, State of Marketing to Engineers 2026; RH Blake and IEN, 2025
6sense's sample is 14% manufacturing companies and 26% physical goods purchases, with a median spend of $200,000 to $300,000. It does not publish a cycle length for manufacturing alone, only the comparison. The direction is still clear: a component or machine purchase moves faster than a software rollout, and the research window before contact is shorter.
Self-serve preference: research alone, then ask an engineer
The cross-industry surveys say buyers want less of the salesperson. The industrial surveys say they want the right one.
| Measure | Figure | Source and sample |
|---|---|---|
| B2B buyers preferring a rep-free experience | 67%, against 61% a year earlier | Gartner, 646 buyers, August to September 2025 |
| Activities done via self-service only, reps only, or both | 1.8, 2.3 and 3.0 on average | Gartner, 632 buyers, August to September 2024 |
| Engineers who prefer to avoid salespeople entirely | 10% | GlobalSpec, 2026, n = 708 |
| First contact to ask about pricing or inventory | 26% | GlobalSpec, 2026 |
| First contact to validate information found online | 23% | GlobalSpec, 2026 |
| First contact about the technical complexity of a solution | 21% | GlobalSpec, 2026 |
| First contact to see a demo | 8% | GlobalSpec, 2026 |
| Software buyers who want sales only in later stages | Nearly two in three, up 17 points | G2, 1,169 buyers, April 2025 |
Sources: Gartner sales survey, published June 2025, via MarketScreener; Gartner survey reported by Digital Commerce 360, March 2026; GlobalSpec, TREW Marketing and Elektor, 2026; G2, May 2025
Gartner's headline is large, but its activity counts show the limit. Buyers who say they prefer no rep still complete more steps with a rep, or with a rep and digital tools together, than through self-service alone. In the industrial data the reasons for contact are specific: price and stock, a check of what the buyer found online, and a technical question. Only one engineer in twelve calls to see a demo.
Gartner also found 69% of buyers see inconsistencies between a supplier's website and what its sellers say. On a manufacturing site the usual gaps are a lead time quoted on the phone that the site never mentions, a datasheet older than the product, and an application claim the sales engineer will not repeat.
Repeat purchases are moving online. For consumables and MRO supplies the self-serve preference is already behaviour. Channel Marketing Group and BigCommerce, an ecommerce platform vendor, surveyed 345 industrial supplies buyers and contractors: 65% had bought online in 2025, against 45% in 2018, and 58% of purchases happened outside traditional distributor relationships. 60% said buying online saves them time. That data covers electrical, HVAC and industrial supplies, not engineered equipment.
Website touchpoints: first source, routine source and proof
The website appears at three points in the journey, and the surveys measure each one separately.
| First source | Share of buyers |
|---|---|
| Search engine | 45% |
| Vendor or supplier website | 23% |
| Peer recommendations | 11% |
| Generative AI tools | 5% |
| Industry analyst reports | 4% |
| Conferences or trade shows | 4% |
| Technical trade publications | 4% |
Source: RH Blake and IEN, 2025; single choice, smaller sources not shown
The 23% who go straight to a vendor's site already know the vendor. Everyone else has to find it, mostly through search. GlobalSpec's engineers agree: 77% use a search engine more often than a generative AI platform, and only 8% use AI more.
| Source | Manufacturing managers, RH Blake 2025 | Engineers, GlobalSpec 2026 |
|---|---|---|
| Vendor or supplier websites | 85% | 74% |
| Technical trade publications | 54% | 76% online, 53% print |
| Conferences or trade shows | 46% | 27% |
| Sales or application engineers | 45% | 41% |
| YouTube | 19% | 30% |
| Generative AI tools | 13% | 21% |
| 12% | 11% |
Sources: RH Blake and IEN, 2025, chart values (the report's text gives 12% for AI and 11% for LinkedIn); GlobalSpec, TREW Marketing and Elektor, 2026, n = 783
The website is used by three to four buyers in five in both surveys, but it is not the only page that describes the supplier. Online technical publications edged past vendor sites for the first time in GlobalSpec's 2026 data, and 31% of engineers routinely use industry directory websites. Thomas, the largest of those directories, reports more than 1.5 million monthly sourcing sessions across more than 500,000 suppliers, with two-thirds of active buyers at small and midsize firms, according to its 2025 Annual Sourcing Activity Report as summarised by Distribution Strategy Group.
Generative AI is growing from a low base. 69% of GlobalSpec's engineers now use it somewhere in the purchasing process, and those who never use it to evaluate vendors fell from 42% to 31% in a year. Trust is low: 4.7 out of 10 in GlobalSpec's data and 4.5 out of 10 in RH Blake's. Forrester's 2026 survey of nearly 18,000 business buyers found 94% use AI during buying, and that buyers validate its output with peers and experts.
The website as proof. RH Blake asked which factors indicate a credible supplier. An informative, helpful and well-maintained website and references from past customers were the only two to score above 4 out of 5, ahead of case studies with well-known brands at 3.72 and thought leadership at 3.6. Buyers rated expertise in their specific application at 4.43 out of 5, the highest of seven technical credibility factors, and 64% always or often seek sources not sponsored by a vendor.
Who else reads the site before the order
The person researching is rarely the only one deciding. In RH Blake's study, 76.9% of buyers rate C-level executives as highly or extremely influential on the final decision: 42.6% extremely and 34.3% highly. Their top concerns when evaluating vendors are reliability (73%), total cost of ownership (52%) and support and warranty terms (45%). Brand reputation is chosen by 16%.
Forrester counts 13 internal stakeholders and nine external participants influencing an average purchase, with procurement acting as a decision-maker 53% of the time. Brand still matters at the tie-break: 70% of GlobalSpec's engineers are likely to choose the better-known brand when two solutions are technically similar, and 53% say familiarity influenced their most recent purchase.
The triggers explain why the journey starts. Asset ageing or replacement starts 59% of significant purchases, regulatory compliance 53% and capacity expansion 52%, while cost saving starts 39% (RH Blake, multiple choice). The first page a buyer needs describes the problem in those terms.
What this means for a manufacturing site: a five-page test order
A manufacturing site has to serve an engineer reading alone, a manager comparing three suppliers and an executive asking about risk, usually before anyone has called. We build marketing sites in Webflow on a four-week process, which is our bias; this is the order we would test in on a manufacturing account.
1. Application pages. Application expertise scored 4.43 out of 5, the top credibility factor. Build one page per named application with the problem, the specification range and a reference, and measure quote requests against the generic capabilities page.
2. A pricing and availability path. 26% of engineers first contact sales about price or stock, and 59% prefer email for that first contact. Add a short "price and lead time" request beside the full RFQ form and compare reply rates and quote volume.
3. Specification pages with downloads. 85% of engineers will share their details for technical content they value, and 15% will never fill in a form. Test ungated datasheets against gated CAD files, measured on qualified requests rather than downloads.
4. A reliability, support and lifecycle cost page. Reliability, total cost of ownership and support are the top three concerns of the buyers whose executives sign off. Write the page for that reader and track who forwards it.
5. A proof page built on references. Customer references score above 4 out of 5 and 64% of buyers look for independent sources. Test named references and links to trade press coverage against a logo wall.
Our marketing websites service page explains how we scope a manufacturing site around this journey.
