What an industrial rebrand actually changes, and what it doesn’t

The new name is approved. The wordmark is signed off and everyone in the room likes it. Then somebody works out that there is a workshop sign visible from the expressway, a fleet of vans, a rack of helmets, a decade of as-built drawings, a set of certificates issued by four different bodies, and an email address printed on every business card handed out since the last general election. All of them say the old thing.

That gap, between deciding on a name and living in it, is where most industrial rebrands either get finished properly or quietly stall. And the stalled version is worse than not starting, because a company halfway through a rebrand reads as a company that is not paying attention.

So it is worth being precise about what an industrial rebrand actually buys you, what it definitively does not, and what it costs in the places nobody puts in the budget.

What an industrial rebrand actually changes

Three things. They are real, and they are narrower than most people selling rebrands will tell you.

Which conversations you get into. Almost every industrial purchase now starts with an assessment you are not present for. A specifier, a QS, a technical manager or a procurement officer has your name from somewhere and is deciding, in under a minute, whether to keep reading. Identity is the compression of everything you have not had a chance to say yet. It does not persuade anyone. It determines whether the persuading gets to happen.

What you can charge. Positioning is a price argument, not a taste argument. A firm that presents as a specialist in a defined scope can hold a rate that a firm presenting as general jobbing capability cannot, because the buyer has no other way to distinguish the two and defaults to price. This is the strongest commercial case for rebranding and the least often made, because it is uncomfortable to say out loud.

How a buyer reads you before they know anything else. Age reads as either heritage or neglect, and the difference is entirely in the execution. A wordmark from 1988 on a well-run site reads as forty years of continuous operation. The same wordmark on a site with a copyright line from 2015 reads as a company that may not still be trading.

That is the honest list. Everything else attributed to rebranding is either a consequence of the work that happened alongside it, usually new content and new photography, or wishful thinking.

What it will not fix

A rebrand does not win work that capability, price and availability did not already win. If you are losing on lead time, you will lose on lead time under a new name and a better typeface.

It will not fix a capacity problem. A main contractor who has been told you are fully committed until the second quarter is not re-evaluating that because your vans changed colour.

It will not fix a price sitting above the market with nothing attached to justify it. Positioning can justify a premium, but only when there is something behind it. A premium identity over commodity capability is a shorter path to being found out, not a longer one.

It will not fix a defect history. In sectors this small, the reputation travels through the same six project managers regardless of what is on the letterhead.

And it will not fix an empty evidence base, which is the most common underlying problem of all. Across 54 Singapore B2B sites we reviewed in August 2026, 19 published no project record, portfolio or case studies of any kind. A new identity applied to a site with no projects on it produces a better-looking site with no projects on it. The frame improves. The absence in the middle of the frame is unchanged, and it is the thing the buyer came to look at.

The useful test before committing budget: write down the last five enquiries you lost and why. If the honest reason is price, availability or scope on four of them, a rebrand is not the intervention. If the honest reason on three of them is that you never heard back at all, it might be.

The cost nobody puts in the budget

The legal part is trivial and it sets a misleading expectation. Changing a business name through ACRA costs S$15 with a review of up to three working days, stretching to 15 working days if the name has to go to a referral authority. A newly approved name can be reserved for up to 120 days while you get organised. Two weeks and the price of lunch, and then the actual project starts.

Vehicle livery. Costed per vehicle, and the invoice is the smaller half. The real cost is the day each vehicle is off the road, multiplied by the fleet, which is why livery should follow the servicing schedule rather than a launch date. Half-liveried fleets are common and they look exactly like what they are.

Signage. Gate, workshop, reception, yard, illuminated boxes, directory listings in a shared industrial building. Landlord approval and, for anything external and illuminated, a permit. This is routinely the longest lead item on the physical side and it is routinely started last.

PPE and workwear. Helmets, coveralls, hi-vis, jackets. Two things get missed. Embroidery has a minimum order and a lead time measured in weeks, and safety-critical items should never be relabelled with a sticker over the old mark. Budget replacement, not relabelling.

Drawings and technical documents. Title blocks in the CAD template. Method statements, risk assessments, inspection and test plans, O&M manuals, datasheets, calibration records, product nameplates and rating plates. Every one of these carries the company name, most of them are templates somebody built years ago, and there is usually no list of where they all live. Historic issued drawings stay in the old name, correctly, and that is fine. What has to change is every template producing new ones from launch day forward.

Certificates and registrations. This is the long pole and it is almost never scheduled properly. ISO certificates, bizSAFE, insurance certificates, licences, class approvals, and in construction the BCA registration itself. BCA’s Contractors Registration System carries about 50 types of workheads across five registration groups, and a firm holding four of them holds four separate things that now say the wrong name. Each certificate has to be reissued by the body that issued it, on that body’s schedule. Until it is, you are quoting under one name and certified under another, and a QS running prequalification will stop and ask. Start this before the launch, not after it.

Everything with the name printed on it. Invoices, purchase order templates, the ERP, delivery notes, bank accounts, GST registration, the signature block on two hundred outgoing emails a day.

There is no reliable percentage to give you here because it depends entirely on how physical your business is. The workable rule is to budget at least as much for application as for the identity design itself, and to expect that a company with a fleet, a yard and a certification burden will spend considerably more on the second half than the first.

Why does a rebrand go wrong on email?

Because a rebrand almost always brings a new domain, and a new domain means DNS, and DNS is the one part of this project where the failure is silent.

When a website moves badly you find out within the hour, because people can see it. When email moves badly nobody sees anything. Enquiries sent to the old address bounce back to a sender who assumes you are simply not interested. Mail sent from the new domain lands in spam folders because the authentication records that vouch for it were never published. Both failures look, from inside the company, exactly like a quiet fortnight.

The specific mistake is treating it as a cutover. Somebody points the MX records at the new provider and deletes the old domain’s mail routing on the same afternoon, before mailboxes are migrated and before a single supplier has been told. Google’s own setup guidance is blunt about the mechanics: mail may not work correctly if you keep old or incorrect MX records, and it can take up to 72 hours for new MX records to be recognised. Seventy-two hours is a long time to be invisible to a buyer with a deadline.

What works instead is dual running, permanently.

Keep the old domain registered indefinitely. Not for a year, indefinitely. It is the cheapest line item in the whole project and it is printed on every business card, drawing and quotation you have issued for as long as you have been trading.

Keep it receiving mail forever, aliased into the new mailboxes, so that the person who saved your address in 2014 still reaches a human.

Publish SPF, DKIM and DMARC for the new domain before you send anything from it, then watch the reports for the first month. A new domain with no sending history is treated with suspicion by default, and that suspicion is what puts your first post-launch quotation in a junk folder.

And redirect the old website page by page rather than dumping everything on the new homepage. The old domain carries whatever search position you have accumulated, and a blanket redirect to the root throws most of it away.

This matters more in this sector than in most, because industrial domains are already fragile. In the same review, 2 of the 54 sites had a domain that did not resolve at all while still ranking in Google under the company name, and 6 answered on only one of the www and non-www forms of their own domain. Those faults were created by somebody making a DNS change and not checking the result. A rebrand is the moment a company makes more DNS changes than it has made in the previous ten years.

The half-finished rebrand, which is the common one

The version we see most often is not a bad rebrand. It is a rebrand that stopped at the top of the page.

The header carries the new mark. The homepage has been restyled. And then everything below the fold is the previous decade. Of the 23 sites in that review whose footers we examined closely, 19 carried an out-of-date copyright year, with stale years running from 2012 through to 2021. One firm holding a serious sector accreditation has a footer reading “Copyright © 2019” with the final word of “All rights reserved” missing its last letter, a typo that has now been live for seven years. Three sites still carried template placeholder content that had never been replaced, and four published their email address only as an image, so it could not be copied or clicked.

The footer is where an unfinished rebrand is always visible, and on an industrial site the footer is not decoration. It is where the legal entity name, the UEN, the registration numbers and the registered address sit. It is the block a QS reads when checking that the company named on the quotation is the company named on the certificate. A footer carrying a shortened trading name, an old legal name, a copyright year from before the rebrand and a credit line for whoever built the site four agencies ago is telling that reader something specific about how the business is run.

The same pattern shows up in the content. The identity is new and the photography is the same stock library. The identity is new and the newest project on the site predates it by three years. In the same review, 12 of the 54 sites were running an outdated or end-of-life CMS or plugin underneath, which means somebody was paying for a redesign while nobody was responsible for the thing it was sitting on.

A rebrand that stops at the logo produces a site that looks newer and reads exactly as old.

Sequencing it so the phone still rings

Reserve the name first, because the reservation window gives you a planning runway rather than a deadline.

Then do the inventory, which is the actual project and the part nobody quotes for. Every physical surface, every document template, every certificate and registration, every system with the name in it, every domain and mailbox. Most firms find between eighty and a couple of hundred items and are surprised by roughly half of them.

Design the identity against the hardest applications first, not the easiest. A van at speed on the PIE. A helmet read from across a site. An A4 title block reproduced in mono on a site printer. A favicon at sixteen pixels. An identity that survives those will survive a business card. The reverse is not true, and industrial brands fail in exactly this direction because they are designed on a screen and applied to steel.

Start the certificate reissues before the public launch. Run email dual-track from the beginning. Announce once, directly to customers, with the old name in the subject line so it is findable. Roll the physical estate on the maintenance schedule.

And put the evidence in before the launch rather than after. A new identity on a site with a real project record, current certifications stated with their scope, and a named person to call is a different commercial proposition from the same identity on an empty site. That was the whole decision behind the rebuild for Hong Hock Global, a water infrastructure contractor whose work is awarded almost entirely on prequalification: the certifications and the evidence moved to the front, because that is the order the buyer reads in. THK Engineering works the same way across five disciplines presented as one package rather than five services.

We handle brand identity and the website together rather than sequentially, because the site is where an identity gets tested and it is where most of the application errors surface first. Pricing is published rather than quoted on request, and identity work is scoped separately from the build because a lighter refresh of what you already have and a full identity with guidelines are genuinely different jobs.

The last thing worth saying is the thing most agencies will not. If your enquiries are healthy, your capacity is full and you are losing on price and programme, keep the name. Spend the money on the evidence instead. A rebrand is a good answer to a positioning problem and an expensive answer to an operational one, and the difference between the two is usually obvious once somebody writes down the last five losses.