+45 XP

Rebranding frameworks & methodology

The new identity is signed off. Now a narrower question needs an answer: on what date does the old name stop appearing on invoices, and what happens to the customer who searches for it eighteen months later? That part of a rebrand gets budgeted last and breaks first. Design takes weeks. Migration takes quarters, sometimes years, and it is where the accumulated preference the foundations lesson describes either transfers or leaks away.

Burberry's 2018 identity change (Riccardo Tisci arriving as chief creative officer, a sans-serif wordmark and monogram from Peter Saville replacing the equestrian knight in use since 1901) was executed cleanly and looked decisive. Five years later, under Daniel Lee, the knight came back. Saville's mark was not the error. The error was swapping the most recognisable assets the company owned without first pricing what each one was worth, and reversals get paid for twice: once to leave, once to return.

Core concept: the equity inventory

Before deciding what changes, list what you own, line by line, and score every line twice: how many people recognise it, and whether that recognition currently helps or hurts.

A usable inventory covers the name, the wordmark, colour and pattern trademarks, the domain and its inbound links, monthly branded search volume, tagline recall, packaging cues, the legal entity name on contracts and tenders, certifications and accreditations registered under the old name, part and SKU codes, physical signage, and app store listings with their review history.

Burberry's check is the clearest case of one asset scoring high on both counts at once. By the mid-2000s it carried recognition most brands would pay for and an association the company was actively trying to shed, so the answer was rationing rather than removal: check exposure was cut back to a small share of the range, then reintroduced deliberately once the association had cooled. An inventory forces that judgement asset by asset. A logo brief forces it once, globally, at the wrong altitude.

Key sub-concept 1: the brand audit as diagnostic tool

A brand audit has three components: internal perception (what employees and leadership believe the brand stands for), external perception (what customers and prospects actually think), and competitive position relative to the alternatives a buyer would genuinely consider.

Volkswagen took a 30% stake in Škoda in 1991 and full ownership by 2000. Well before that, the cars shared VW platforms and were competitive on quality, yet UK perception ran roughly a decade behind the product. The diagnostic finding that mattered: the liability sat almost entirely with people who had never owned one. Existing owners were satisfied. So the work Fallon produced from 2000, including "It's a Škoda. Honest.", was aimed at disbelief rather than awareness, which is a different creative brief and a different media plan.

The failure mode here is sampling your own customer base. Survey only current customers and you interview the people for whom the liability already failed to bite. Quote lapsed customers and never-considered prospects in at least equal number.

Key sub-concept 2: positioning architecture before visual identity

Most rebrands start with a logo brief. That is backwards. Define the target segment, the category you compete in, the single benefit you intend to own, and the reason a buyer should believe it. The sentence: for [target customer], [brand] is the [category] that delivers [benefit] because [proof].

Andersen Consulting had no time for a logo debate. An arbitration ruling in August 2000 required the Andersen name gone by 1 January 2001, about 147 days later. The name that emerged, Accenture, came from an internal employee suggestion scheme (accent on the future), and the launch strapline "Innovation delivered" stated the claim the firm was already building toward: work beyond audit-adjacent advisory, into technology and outsourcing. The architecture existed before the mark did, which is why the mark was arguable and the story was not.

Test every shortlisted name against trademark clearance in the classes and countries where you actually sell, plus a linguistic screen. Finding the conflict after launch means running the entire migration twice.

Key sub-concept 3: migration sequencing, not a launch date

  • Phase 1, internal and legal readiness: clearance secured, entities renamed, domains and registry records controlled, employees briefed live rather than by email. Accenture compressed this across dozens of countries because arbitration set the clock; most companies have no such excuse and still leave it to the last fortnight.
  • Phase 2, endorsed dual-running: the old name endorses the new one ("formerly X", or "an X company"). Budget six to eighteen months, longer in tender-driven B2B where prequalification lists, framework agreements and supplier registers carry the old entity and cannot be edited on your schedule.
  • Phase 3, the switch: one narrative built around what changes for the customer (contracts, support numbers, login domains, warranty terms), with 301 redirects mapped page by page, dual email delivery on both domains, and purchase orders re-papered so that accounts payable systems at your enterprise customers do not reject invoices from an unrecognised vendor. This is the step that quietly stops cash.
  • Phase 4, decommission and governance: a named owner with authority to reject off-brand work from any department, plus a dated kill list for old stock, signage, templates, PDFs and third-party listings.

The Gap Logo Disaster

Watch on YouTube

Key sub-concept 4: the measurement framework

  • Brand health: awareness, consideration, preference, NPS, baselined before the switch and tracked quarterly after. Without the pre-switch baseline you cannot distinguish rebrand damage from a soft quarter.
  • Business metrics: revenue per segment, acquisition cost by channel, retention.
  • Transition telemetry, which most measurement plans omit: share of branded search landing on the new name versus the old (expect a tail measured in years), direct and organic traffic against the pre-switch baseline, percentage of physical and partner touchpoints converted against plan, invoice rejection and payment-delay rates, and inbound contacts asking who you are now.

Set the recovery threshold before launch. If searches for the new name have not overtaken the old after four quarters, you have a communications and spend problem, and a second redesign will not fix it.

Real-world cases

Case 1: Accenture (2001)

A legally forced rename, executed in roughly 147 days, with a launch spend reported at around $100 million. The old name held real equity, none of which could be kept, so the inventory question inverted: what transfers instead? Client relationships, partner contracts and

personnel did, which is why the sequencing put contract novation and client briefings ahead of advertising. The compressed timeline was survivable because the reason was undeniable internally. Voluntary rebrands rarely get that alignment for free and should not copy the timeline.

Case 2: Škoda under Volkswagen

The opposite decision. The name itself was the liability, and VW kept it. Recognition plus an existing dealer network was judged worth more than a clean sheet, so the recovery ran product first, then perception, over more than a decade rather than a campaign cycle. Škoda went on to place at or near the top of UK customer satisfaction surveys repeatedly through the 2000s, and now delivers close to a million cars a year. Keeping a damaged name is a legitimate methodology choice when the damage is a perception lag and the product problem is already solved.

How Dunkin' Rebranded to Drop 'Donuts'

Watch on YouTube

CMO action items

  • Produce the equity inventory as a spreadsheet with a recognition score and a liability score per asset, and make the keep/ration/retire decision explicit on every row before any agency sees a brief.
  • Write the positioning sentence yourself. Every brief comes after that document exists.
  • Build the migration calendar backwards from decommissioning, with named owners and go/no-go criteria per phase. Include the unglamorous lines: entity renaming, tender registers, certifications, redirect maps, AP re-onboarding at your top 50 accounts.
  • Baseline brand health and branded search volume at least one quarter before the switch, and publish the recovery threshold you will be held to.

Common mistakes that kill results

Mistake 1: changing identity without pricing the assets first.

This is how a five-year reversal happens, as at Burberry. The cost is not only the second design fee. It is the recognition you spent five years teaching the market to unlearn, then re-taught.

Mistake 2: treating the visual identity as the rebrand.

A logo and palette are outputs. If the positioning and the customer experience are unchanged, the new mark signals activity without delivering value, and it invites the question of what else the money could have bought.

Mistake 3: no decommissioning plan.

Old assets persist in warehouses, PDFs, partner sites and forecourt signage, and two identities in market read as instability rather than continuity. The physical estate is usually the longest and most expensive line in the migration budget, so it should be dated and funded first, not treated as tidy-up after launch.

Resources

What to do, from this lesson

These actions are compiled in the role's Playbook.

  • Align internal teams and run enablement sprint before any external rebrand launch
See the full action playbook →

Related articles

Recent articles from the blog that build on this lesson.