9 min read · 1,927 words
Marketing in a manufacturing company is rarely lost in the market. It is lost inside the building, in the budget review where a sales director asks what the last campaign delivered and nobody has an answer that survives arithmetic. Win that room first. Media, content and positioning are all downstream of it.
I have spent most of fourteen years marketing things that photograph badly. Tractors. Printers. Freight capacity. Machinery sold into markets where the buyer’s father bought the same brand and saw no reason to reconsider. At Ford, at CNH Industrial across Asia, the Middle East, Africa and Turkey, through freight and logistics, and now in steel, the pattern has repeated with very little variation.
The pattern is this. The company was built by engineers. The product is genuinely good. Sales sits close to the customer and is proud of it. And then marketing arrives with a vocabulary that sounds, to everyone else in the room, like decoration. I got this wrong for a long time before I got it right, and what changed was not my media plan.
Why does marketing in a manufacturing company get treated as overhead?
Because every other function in a product-led business can draw a straight line from its budget to a unit of output. Manufacturing produces units. Sales produces orders. Service produces uptime. Marketing produces a deck. That asymmetry is the whole problem, and industrial categories make it worse because the sales cycle is long enough to break the causal chain between spend and order.
The scale of what is coming makes this more than an internal irritation. Manufacturing contributes roughly 17% of India’s GDP, against a stated national target of 25% by 2035, and the manufacturing PMI stood at 55.4 in January 2026, according to IBEF. A large cohort of Indian industrial companies is about to discover that relationships and price alone will not hold a category. Most of them will hire a marketing head before they have decided what that person is for.
The four objections you will hear, and what each one actually means
Every objection to marketing spend in an engineering-led business is one of four things wearing a different coat. Naming the real objection is most of the work.
| What gets said | What it actually means | What answers it |
|---|---|---|
| “Our customers are technical. They don’t respond to marketing.” | They have only seen bad marketing, usually a brochure with a handshake photograph. | Evidence that technical buyers consume vendor content heavily, just not that kind. |
| “Sales brings in the revenue.” | Fear that marketing is claiming credit for orders sales fought for. | Shift the claim from attribution to inclusion: who gets considered at all. |
| “Put it into the trade show, or into R&D.” | A preference for line items that produce something you can stand next to. | A published split between demand capture and demand creation, with reasons. |
| “Brand is for consumer companies.” | Brand has been confused with aesthetics rather than memory. | Define brand as buyer memory under a specific set of buying situations. |
Notice that none of these are arguments about marketing. They are arguments about risk, credit and evidence. Answer them as marketing arguments and you lose.
What I tried first, and why it did not work
My first instinct was education. I explained funnels. I built a brand book with a tone-of-voice section. I presented an awareness tracker to a room that wanted to talk about order intake, and I watched a regional sales head do the mental arithmetic on what the tracker had cost per percentage point.
That failed because it asked people to accept a belief system before I had shown them a single testable claim. Engineers do not adopt frameworks on authority. They adopt them after the framework predicts something correctly.
My second attempt failed in the opposite direction. I over-corrected into pure lead generation, cut everything that could not be tied to an enquiry within the quarter, and reported cost per lead with great confidence. The pipeline did get cheaper. Then, about three quarters later, it got thinner, and the enquiries that did arrive were increasingly from buyers who had already decided on someone else and wanted a third quote.
Engineers are not anti-marketing, they are anti-unfalsifiable, and most marketing plans are deliberately written so that they cannot be proven wrong.
Are engineers actually hostile to marketing?
No, and the data is fairly blunt about it. In the 2025 State of Marketing to Engineers study from TREW Marketing and GlobalSpec, 73% of technical buyers said they rely on vendor websites and online technical publications as trusted sources, 91% subscribe to newsletters, and 64% listen to work-related podcasts. Roughly 60% of the buying process happens online before an engineer contacts a supplier at all.
The same study found 70% of engineers rarely or never use AI to evaluate vendors, and rated their trust in AI-generated content at 4.4 out of 10. That is a useful correction for anyone planning to fill a technical content calendar with generated material. This audience reads more vendor content than almost any other, and discards more of it, because they check things.
So the hostility is not to marketing. It is to claims that cannot be inspected. Which is, when you think about it, a reasonable position for someone whose own work fails visibly when a claim is wrong.
The reframe that worked: brand as a tolerance problem
What finally moved the argument was translating brand into the vocabulary the room already used: specification, tolerance, lead time and failure mode.
Tolerance. The Ehrenberg-Bass Institute’s work on the 95-5 rule, summarised by Professor John Dawes in a 2021 B2B report, found that only around 5% of business buyers are in-market in a given quarter and about 20% across a full year. Well-established brands typically reach 20–30% mental availability against the buying situations in their category. That is your operating band. Marketing to the 5% only is running a machine at the edge of tolerance and calling it efficiency.
Failure mode. Bain & Company and Google surveyed 1,208 buyers of software, hardware, logistics and industrial equipment, and reported in Harvard Business Review in 2022 that 90% of buyers ultimately choose a vendor that was on their shortlist at the very start of the process. If you are not on the day-one list, you are not competing. You are quoting to lose, and providing a price anchor for the firm that wins.
Load. Forrester’s State of Business Buying, 2026 puts a typical decision at 13 internal stakeholders plus nine external influencers, with procurement acting as a decision-maker in 53% of buying cycles. Twenty-two people, and a field sales team of any realistic size covers perhaps four of them. The other eighteen form their view from whatever exists in public.
Lead time. This is the argument that lands hardest in a factory. Nobody commissions tooling in the week they need the parts. Brand has a lead time for exactly the same reason, and the invoice for skipping it arrives two years later as a longer sales cycle. I have written separately about why awareness is the wrong thing to track in B2B and about the five layers of brand architecture that decide which name carries that memory.
What an engineer will actually fund
Three things, in my experience, and they are the same three things any engineering proposal contains.
- A falsifiable hypothesis. Not “increase brand awareness”. Something closer to: “if we publish against the six buying situations where we currently have under 15% mental availability, our share of first-round enquiries in that segment rises within four quarters.”
- A named instrument. How the claim will be measured, decided before the money is spent, not reverse-engineered from whatever moved.
- A kill condition. The number at which you personally recommend stopping. Nothing buys credibility in an engineering-led business faster than volunteering the condition under which you would shut your own programme down.
The split itself has published guidance. Les Binet and Peter Field’s analysis for the LinkedIn B2B Institute, drawing on the IPA Databank, put the optimal B2B budget at roughly 46% brand and 54% activation, a ratio LinkedIn now summarises as an even 50/50 split. The same study found that 10 points of extra share of voice above share of market correlates with around 0.7% market share growth a year, and that emotional campaigns averaged 1.4 business effects against 0.2 for rational ones. The same work noted that only 4% of B2B marketers measure anything beyond six months, which is roughly the point at which the brand effects begin. I have set out how to convert that into an actual number in a post on how much a B2B company should spend on brand.
The buying behaviour has moved faster than the org chart
The internal argument is getting easier for a reason that has nothing to do with marketing’s persuasiveness. Buyers changed.
Gartner, surveying 646 B2B buyers between August and September 2025, found 67% prefer a rep-free buying experience and 45% used AI during a recent purchase. McKinsey’s 2026 B2B Pulse, covering nearly 4,000 decision-makers across 13 countries, reported that 73% are now comfortable placing orders above $50,000 online, up from 59% in 2022, and that buyers use an average of ten channels across a purchase.
Read those two findings next to each other. Two-thirds of your buyers would rather not speak to your sales team, and they are willing to commit serious money without doing so. In a company whose entire commercial model assumes a relationship built over site visits, that is not a marketing problem. It is a coverage problem, and marketing is the only function structurally able to solve it. That is the argument to make, and it is considerably stronger than anything in a brand book.
What this means for you
If you are the first or only marketer in an engineering-led business, this is where I would start on Monday.
- Stop presenting awareness numbers to people who own order intake. Present consideration and shortlist inclusion, because that is the mechanism the Bain and Google data actually describes.
- Map the twenty-two people. List the internal stakeholders and external influencers on your last three lost deals, then mark which ones your sales team has ever spoken to. The gap is your brief.
- Publish your split. Put demand creation and demand capture on one slide with a percentage against each and a reason for the ratio. Ambiguity reads as evasion in a factory.
- Write one falsifiable claim per programme, with a kill condition. One page. It will do more for your standing than a year of good creative.
- Audit what a buyer can find without you. Sixty per cent of the process happens before contact. Search your own category as a buyer would and look honestly at what comes back.
- Choose the arithmetic willingly. If you do not bring the numbers, finance will bring theirs, and theirs will be about cost.
None of this makes the job comfortable. Marketing in a manufacturing company is a permanent audition, and I have stopped resenting that. A function that has to justify itself every quarter tends to end up with a sharper argument than one that never has to. The mistake is trying to end the audition rather than getting better at it. I have argued something similar about marketing inside the supply chain, where the same instincts apply.
So here is the question I would put to anyone running marketing in an engineering-led business: what is the one number you report every quarter that, if it moved the wrong way for two quarters running, would make you recommend cutting your own budget? If you cannot name it, that is the work. Tell me what yours is over on LinkedIn.
