Industrial marketing measurement, where the buyer is a committee and the order is a PDF

Industrial marketers are told, usually by other industrial marketers, that their sector is not measurable. The reasons given are always the same three: the cycle is too long, the buyer is a committee, and a large share of revenue goes through distributors who do not share data. All three are genuinely true, and none of them prevents the useful measurement, which is considerably narrower and more valuable than what people imagine they are being denied.

You cannot divide credit fairly across a two-year committee decision. You can absolutely find out which enquiries became orders, and most industrial businesses have never done that.

This article is about the second thing. It is written for manufacturers of capital equipment, components, industrial services and anything else sold to engineers and bought by procurement.

The measurement that is actually available

The reliable claim in an industrial business is that a specific order traces to a specific enquiry from a specific account. It is verifiable, defensible, and sufficient for the budget decisions that need making.

Notice how much smaller this is than what attribution vendors typically promise, and how much more useful it is than what most manufacturers currently have. It does not tell you what share of the decision the trade show deserved. It tells you that this three hundred thousand euro order came from an enquiry generated at that trade show, with both records available to inspect.

For a business allocating a budget across trade shows, technical content, trade press, distributor support and paid search, that is enough. The question being asked is which activities produce enquiries that become orders, and matched revenue answers it directly.

What it will not do is settle the internal argument about whether the specification the engineer downloaded eighteen months earlier deserves credit. That argument cannot be settled by any method available, and pursuing it is the main reason industrial measurement projects stall.

Account-level matching is not optional here

In industrial sales the person who enquires is frequently not the person who orders. Matching on individual identifiers alone will miss the majority of genuine connections and will make every early-stage channel look worthless.

The typical journey involves a design engineer researching a solution, a production manager evaluating it, and a procurement officer executing the purchase, often across many months and sometimes across different sites of the same group. Three individuals, one buying decision, and an enquiry record that names only the first.

Account matching connects them: normalise the company name, use the email domain, or use a company registration identifier where both systems hold one. Domain matching is the most reliable of the three in practice, because company names are entered inconsistently and registration numbers are rarely present in marketing systems.

Because account matching is looser than person matching, it needs a review step. A large manufacturer may have several unrelated projects running at once, and an enquiry from one division becoming an order from another is plausible but not certain. Treat these as a review queue rather than automatic revenue, and the method stays defensible.

The distributor problem, and what can be done about it

Where distributors make the sale, the manufacturer often cannot see the end customer. This is a genuine limit, and it is usually narrower than assumed, because many distributor relationships include reporting that nobody has asked for.

The first step is to find out what is already available. Distributor agreements frequently include sell-through reporting, registration of major projects, or deal registration schemes for larger opportunities. Those records contain end-customer information and are often held by sales operations rather than marketing, unexamined for this purpose.

Where deal registration exists, it is the highest-value data in the business for this question, because the distributor has already told you which end customer a project relates to. Matching registered deals against your own enquiry history closes the loop for exactly the large opportunities that matter most.

Where nothing is available, measure the part you control — qualified enquiries passed to each distributor, and their eventual reported outcome where visible — and state plainly that the remainder is unmeasured. An honest boundary is far more useful than an estimate nobody can check.

What to measure while everything is still open

With cycles measured in years, revenue evidence arrives too late to steer activity. Use requests for quotation, sample and specification requests, and qualified pipeline value by source, and validate each against eventual revenue.

The right-hand column is the discipline. Every one of these indicators can be inflated by ordinary behaviour without any increase in eventual revenue, and the only protection is to re-check the relationship against closed orders every couple of quarters.

Trade shows, which everybody argues about

Trade shows are the largest single line in many industrial marketing budgets and the least measured. Matched revenue can settle their contribution better than any other method available, provided badge scans reach the CRM.

The obstacle is almost always intake rather than analysis. Badge data sits in a spreadsheet from the organiser, arrives two weeks after the event, and is either loaded into the CRM in a form that cannot be matched or is not loaded at all. Fixing that one process step converts the most contested budget item into a measurable one.

When it is measured, the result is frequently surprising in both directions. Shows widely regarded as essential sometimes produce very little traceable revenue, and smaller regional events sometimes produce a great deal. Both findings are worth having before the next annual booking cycle.

Expect resistance, because trade show decisions carry relationship and status considerations that are not purely commercial. The measured figure does not have to be the only input; it just has to be in the room.

Technical content, judged properly

Datasheets, application notes and specification guides are consumed early, often anonymously, by people who will not sign the order. Judging them on last-click revenue will always understate them, and judging them on downloads will always overstate them.

The measure that works is whether accounts that engaged with technical content go on to enquire and to close at higher rates than accounts that did not. That is an account-level comparison rather than a per-sale attribution, and it is answerable from the same matched dataset.

It is also worth tracking specification inclusion where your sector has it: whether your product is named in a tender or a design specification is a lagging but decisive indicator, and it is directly influenced by technical content years earlier.

CloseRev matches at account level as well as person level, and holds lower-confidence matches in a review queue rather than counting them automatically. That distinction is what makes account matching usable in a long-cycle business.

Aftermarket and spares, the revenue nobody attributes

Spare parts, consumables and service contracts often carry higher margins than the original equipment and are almost never included in marketing attribution. Including them changes which acquisition channels look valuable.

A machine sold at thin margin that generates a decade of parts and service is a fundamentally different customer from an identical machine sold to an operator who sources spares elsewhere. Attribution built on the original order value treats them as the same, and will therefore rank the channels producing them identically.

The data is usually available and unexamined, because aftermarket revenue is recorded against the account rather than against a marketing source and nobody has connected the two. Once accounts are matched to their original enquiry, the entire aftermarket stream can be attributed to the acquisition source with no additional matching work at all.

This is one of the few places in industrial measurement where the effort is small and the finding is frequently dramatic. Businesses regularly discover that their most profitable acquisition channel is not the one producing the largest orders, and that discovery is invisible without the aftermarket view.

The tender and framework complication

Where sales run through public tenders or framework agreements, the enquiry that mattered may have occurred years before the tender and to a different organisation entirely — a consulting engineer, a specifier, or a framework body.

This is a genuinely hard case and it is worth naming rather than papering over. The purchasing organisation may have had no contact with your marketing at all, while the specification that made your product eligible was influenced by technical engagement with a third party two years earlier. No matching method will connect those records, because there is no shared identifier and often no shared organisation.

The practical response is to measure specifier engagement as its own programme with its own indicators — specification inclusions, technical consultations, standards participation — and to accept that it will appear in the attribution report as unmatched revenue. Reporting it separately prevents it from being cut on the grounds that it produces nothing traceable.

It is worth stating this explicitly in any report that goes to a board, because the alternative is that the most strategically important marketing activity in the business sits invisibly inside an unattributed bucket and loses its budget to something measurable and less valuable.

Sales and marketing, and the handover nobody documents

In most industrial businesses an enquiry becomes an opportunity through a handover that is undocumented, inconsistent, and where a substantial share of enquiries are simply never worked.

The symptom is familiar: marketing reports a healthy volume of enquiries, sales reports that the enquiries are poor, and the matched data shows that a large proportion never received a follow-up at all. In long-cycle businesses this is especially costly, because an unworked enquiry does not resurface — the buyer completes their evaluation with somebody who answered.

Matched data makes the handover visible for the first time by connecting enquiries to outcomes, including the outcome of nothing happening. The most useful single number to produce is the share of enquiries with no recorded contact within a defined period, broken down by source and by territory.

Present it without blame and it usually fixes itself quickly, because nobody defends the position that enquiries should go unanswered. Present it as evidence that sales is failing and it will be contested on methodology for six months.

The export that makes this possible

Export closed orders with the account identifier, the value, the order date and the contact, and export enquiries reaching back at least three years with the same fields. The date range is the most common point of failure.

Three years is a floor rather than a target in capital equipment. If your ninetieth-percentile cycle is two and a half years, a three-year enquiry export explains a single year of orders, and anything shorter will leave a large share of revenue unmatched for purely mechanical reasons.

Include the account identifier on both sides even where the person identifiers are good, because it is the field that makes committee buying tractable and it is trivial to add at export time.

What to do with the first result

Expect a lower match rate than a consumer business would see, expect trade shows and referrals to be larger than assumed, and expect a substantial unattributable share reflecting the invisible part of industrial buying.

A forty to sixty percent match rate is reasonable in this sector and is not a failure. The invisible portion genuinely is invisible: technical evaluation on third-party sites, peer conversations, standards committees, and the specification written by a consulting engineer you have never met.

The value in the first result is rarely the overall figure. It is the ranking, and specifically the two or three sources that turn out to be much larger or much smaller than the organisation believed. That finding usually pays for the exercise several times over in the first budget cycle.

There is a second-order benefit to running this analysis that is worth anticipating. The first matched report in an industrial business almost always uncovers data quality problems that have been quietly degrading the CRM for years: accounts duplicated across spellings, enquiries logged against a generic company record, contacts with no email. Those are worth fixing on their own merits, and the attribution exercise is usually the first thing that makes them visible and gives somebody a reason to care.

If you take one action from this article, make it the three-year enquiry export. It is a single request to whoever administers your CRM, it takes them under an hour, and it is the precondition for every other analysis described here. Most industrial businesses that believe their marketing is unmeasurable have simply never had that file in the same place as their order history.

Industrial marketing is not unmeasurable. It is unmeasured, because the export nobody has run needs to reach back three years and nobody has asked for it.

Questions people actually ask

Can industrial manufacturers measure marketing ROI?

Yes, by matching closed orders to enquiries at the account level rather than the individual level, over a window long enough to cover the buying cycle. What cannot be measured reliably is how credit should be divided among the many touches inside a committee decision.

How do you attribute a sale when several people at one company were involved?

Match on the account as well as the person. The engineer who downloaded a specification and the procurement manager who signed the order are different records, and only an account-level match connects them.

How do you measure marketing when distributors make the sale?

Ask distributors for closed order data where the relationship allows, and match against enquiries you generated and passed on. Where it does not, measure what you can control — qualified enquiries passed, and their conversion where visible — and be explicit that the remainder is unmeasured.

What should manufacturers measure while deals are still open?

Requests for quotation by source, sample and specification requests, and qualified opportunity value created by source. Each should be checked periodically against eventual closed revenue rather than assumed to predict it.

Is content marketing measurable in industrial sectors?

Its influence is measurable in aggregate and rarely attributable to individual sales, because technical content is typically consumed early, anonymously, and by people who are not the eventual signatory. Judge it on enquiry quality and specification inclusion rather than on last-click revenue.

How far back should an industrial lead export go?

Three years is a reasonable default and more is better. Capital equipment cycles routinely exceed two years, and an export shorter than the cycle will leave most revenue unattributable for reasons that have nothing to do with marketing.

Found this useful?

• Share on X
• Share on LinkedIn
• Share on Reddit
• Share on Email