# When to rebrand: test what your buyers recognise before you change it

> Rebrand when the brand misdescribes the business, and decide how much to change by testing which cues buyers link to you and to nobody else. Keep and evolve the strong ones, change the weak ones freely, and plan the name, domain and listings in the right order. Your most committed customers react worst to change.

By LAIN, 29 Jul 2026. 11 min read. Category: Branding. Canonical: https://lainagency.com/resources/when-to-rebrand

Most rebrand advice starts with whether your brand feels dated. That is the wrong first question, because the people who feel it first are you. The better questions are whether the brand misdescribes the business, and which of its cues your buyers actually use to recognise you. The first tells you whether to change anything. The second tells you what you can safely change.

## Two questions decide it
1. **Mismatch.** Does the brand describe a business that no longer exists, promise something you do not deliver, or send you to the wrong customers? A name that no longer fits, a merger, a new market and a quality gap between the work and its packaging all count.
2. **Asset strength.** Which cues (name, logo, colour, shape, typeface, phrase, sound) do your buyers link to you, and to nobody else? Jenni Romaniuk of the Ehrenberg-Bass Institute frames this with two tests. **Fame** is how many buyers link the cue to you. **Uniqueness** is how exclusively it points to you rather than to competitors. Secondary summaries of her work give a rough working standard of 50 per cent on both, tested unprompted: show the cue with the name removed and ask who it belongs to. Treat 50 per cent as a starting rule of thumb, not a law.

![A grid of fame against uniqueness with four areas: make it known, keep it, free to change and shared.](https://lainagency.com/images/blog-art/when-to-rebrand-fig2.webp)

Put the two answers together:

| Mismatch | Asset strength | Move | Why |
|---|---|---|---|
| High | Weak | Full rebrand | There is little recognition to lose, so the name, mark and system can all change |
| High | Strong | Evolve | The mismatch is in what the brand says, and the recognition is in its form. Change the message, imagery and tone, and keep the strongest cues |
| Low | Weak | Fix the execution | Nothing misleads and nothing is recognised, so the problem is usually inconsistent use. Make a system and use it |
| Low | Strong | Leave it alone | Spend the effort on using the brand more consistently |

This is our reading of how to apply the grid, not a published model. It does explain why logo redesigns for fashion's sake are the riskiest move on the page: they change strong cues without fixing a mismatch.

Three of our case pages show what a set of assets looks like, though none of them is a rebrand. [Montreval](https://lainagency.com/works/montreval), a private golf club in Geneva, has a lion, a wordmark with a script beneath it, four colours, a strip of diamonds and lions and a knitwear pattern, all built so the club "looks like itself wherever people meet it". [Weyve](https://lainagency.com/works/weyve) takes its name from one continuous pen line and enlarges the same line into the ribbon behind everything. [Ordin](https://lainagency.com/works/ordin) keeps one citrine colour for data only, which is the kind of rule that makes a colour easier to own. Each of those would show up in an audit as a cue to protect.

## Test it, do not judge it
Your own opinion of your assets is a weak instrument. Brus, Hartnett, Faulkner and Driesener, all linked to Ehrenberg-Bass, compared marketers' predictions with consumer data on 405 brand elements from 50 brands in five categories (7,505 respondents, as reported in secondary summaries). Marketers' familiarity with their category led them to overrate fame and underrate uniqueness, and the authors' conclusion is that intuition is not a reliable substitute for research with consumers.

You do not need a large panel. The margin of error at 95 per cent confidence for a result near 50 per cent is about 1.96 times the square root of 0.25 divided by n. We calculated:

| People asked | Margin of error |
|---|---|
| 50 | 13.9 points |
| 100 | 9.8 points |
| 200 | 6.9 points |
| 400 | 4.9 points |

For a yes-or-no on a 50 per cent threshold, 96 people gives you roughly plus or minus 10 points. If you have a small customer list, the finite population correction lowers that: of 300 customers you need about 73, of 150 about 59, and of 1,000 about 88. Ask buyers, not colleagues or friends. Ask for the answer in their own words rather than from a list, because a list helps people guess. For a very small B2B list, ask every customer and record the answers.

## What research says about changing a logo
Two studies are worth knowing, with their limits.

- **Walsh, Winterich and Mittal (Journal of Product and Brand Management, 2010)** tested redesigns of the Adidas and New Balance logos with 632 undergraduates. When an angular logo was made rounder, strongly committed customers rated the redesign more negatively and weakly committed customers rated it more positively. The more the logo changed, the bigger the gap. The authors explain it as a threat to the customer's connection with the brand. A follow-up found culture moderated the result.
- **Müller, Kocher and Crettaz (Journal of Business Research, 2013)** ran an experiment with 385 respondents and found that redesigning a logo changes how modern the brand is judged to be, with the chain running through logo attitude to brand attitude and loyalty. We could not access the effect sizes.

Both are laboratory studies with students or panel respondents, so they show a direction, not your odds. The direction matters: a refresh can make a brand look more modern, and the people who pay the price are your most attached customers. Tell them first, and keep the cue they hold on to.

## Three reversals, read with care
- **Tropicana, 2009.** Industry data reported by secondary sources put unit sales of Pure Premium down 20 per cent from 1 January to 22 February 2009, which is 53 days, about seven and a half weeks. The company announced on 23 February that it would drop the new packaging. Cost estimates range from 30 to 50 million US dollars. This was packaging, not a logo, and the sources note that other factors could have played a part, although competitors saw double-digit unit gains in the same period.
- **Gap, 2010.** A new logo appeared around 6 October and the old blue box was back by 12 October, six days later (Branding Journal). One widely repeated cost figure comes from a blog estimate, so we left it out.
- **Twitter to X, 2023.** Brand Finance valued the brand at 5.7 billion US dollars in 2022, nearly 3.9 billion in 2023 and 673.3 million in 2024, and one analysis argues the rebrand removed distinctive cues such as the bird and the word "tweet". The change also coincided with a takeover, and brand valuations have no single method, so read it as a warning about cues with both high fame and high uniqueness, not as a measured cost.

The common thread is our interpretation, not a finding: in each case something buyers used to find the product was changed or retired.

## The costs that are not design
**A new name needs a new trademark clearance.** Search before you announce. In first-to-file countries such as Indonesia, under Law No. 20 of 2016, the first party to file generally has the stronger claim. Our [naming guide](https://lainagency.com/resources/how-to-choose-a-brand-name) covers the search. This is general information, not legal advice.

**Your Google Business Profile follows the real-world name.** Google's guidance says the profile name should reflect the name used on your storefront, website and stationery. Third-party guides say a minor change can be edited in place while a major one may be treated as a new business that needs a new profile, and advise updating your website and other listings first. We could not confirm that distinction on Google's own pages, so check the current rules, and check what happens to your reviews, before you rename.

**A domain change is a site move.** Google's Change of Address tool is for moving a domain or subdomain, checks that you own both and that 301 redirects exist, and Google says a request may take up to 180 days to complete. It does not cover changes to page addresses inside a domain. Google's site move guidance recommends keeping redirects for at least a year (we carried this over from an earlier check of that page and did not re-read it in this revision). If the name changes and the domain does not, you avoid the move. The [redesign SEO checklist](https://lainagency.com/resources/website-redesign-seo-checklist) shows how to map the addresses.

**The legal and administrative list is long** and depends on where you are: the registry, the tax authority, permits, bank accounts, insurers, contracts and payroll all carry the name. Check each against your jurisdiction.

## Order of operations
The order below is our sequencing, based on what depends on what.

| Step | Why it comes here |
|---|---|
| 1. Clear and file the name | Everything public depends on it. Do it before you announce |
| 2. Legal entity, bank, tax, insurers, permits | The authorities set the lead time, so start early |
| 3. Domain, redirect map and email | Live on launch day, kept for a year or more |
| 4. Website, then listings and profiles | Profiles are checked against the real-world name, so the site goes first |
| 5. Templates: email signature, proposals, invoices, social | Ready on launch day so the brand is easy to use correctly |
| 6. Physical: signs, vehicles, packaging | Long lead times. Order early and use up old stock |

Signage checklists from print and sign firms agree on the same approach: inventory every item, sort by priority, give each an owner. Custom signs have long lead times, so they go in first.

## Where it goes wrong
- **Rebranding to fix sales.** Cause: the brand is rarely the bottleneck. Detect it by asking five to ten lost prospects what they compared. If they name price, timing or proof, not the look, a rebrand will not help.
- **Boredom.** Cause: your team sees the brand every day and your buyers rarely do (our judgement). Detect it with the asset test. If the cues are strong, refresh how they are used and keep them.
- **Changing everything at once.** Cause: no one can tell which change caused a loss, and no cue is left for buyers to hold on to. Fix: keep the strongest cue constant.
- **Announcing before clearing.** Cause: the name is public before it is protected. Fix: step 1 comes before step 4.
- **Surprising your best customers.** Cause: the commitment effect above. Fix: preview the change with a few of them and explain the reason.

## What to do next
1. Write one sentence on the mismatch, or admit there is none.
2. Run the asset test with about 100 buyers, or all your customers if there are fewer.
3. Find your quadrant and choose the smallest move that fixes the mismatch.
4. Plan the order of work, name first.
5. Run the same test again after three to six months, and compare fame and uniqueness with the start.

At LAIN a branding project typically takes 4 to 6 weeks over four phases, and the asset test fits the first, Alignment. The finished system is documented in [brand guidelines](https://lainagency.com/resources/brand-guidelines-what-to-include) and a [logo package](https://lainagency.com/resources/logo-file-formats-guide). If you are weighing a rebrand and want an honest view of how much change you need, [tell us about the business](https://lainagency.com/contact).

## Frequently asked questions

### How do I know if my logo is worth keeping?

Test it rather than judging it. Show buyers the logo with the name removed and ask whose it is. The share who name you is its fame, and the share who name only you is its uniqueness. Secondary summaries of the Ehrenberg-Bass work suggest a rough working standard of 50 per cent on both. If it passes, keep it and evolve around it.

### Will a rebrand upset my existing customers?

It can upset the most attached ones. In a 2010 experiment with 632 students, strongly committed customers rated a rounder redesign of the Adidas and New Balance logos more negatively, while weakly committed customers rated it more positively, and bigger changes widened the gap. Preview changes with loyal customers and keep the cue they hold on to.

### How many people do I need to ask to test my brand assets?

About 96 people gives a margin of error of roughly 10 percentage points at 95 per cent confidence when the true answer is near 50 per cent. With a small customer list the correction helps: about 73 of 300 customers, or about 59 of 150. Ask real buyers and let them answer in their own words.

### Does changing my domain name hurt SEO?

A domain change is a site move, and it carries risk. Use permanent 301 redirects from every old page to its new equivalent and Google's Change of Address tool in Search Console, which Google says may take up to 180 days. Google's site move guidance is to keep the redirects for at least a year. Keeping the domain when you rename avoids the move.

### Can I rebrand without changing my business name?

Yes, and it is often the safer choice. An evolution changes imagery, tone, type and system while keeping the name and the strongest recognisable cues, so you avoid new trademark clearance and a domain move. Change the name when it is legally blocked, no longer describes you or has to change after a merger.

### What went wrong with the Tropicana and Gap rebrands?

Both were reversed within weeks. Reported industry data put Tropicana Pure Premium unit sales down 20 per cent between 1 January and 22 February 2009, and the company announced it would drop the new packaging on 23 February. Gap's new logo appeared in early October 2010 and the old one was back six days later. Both figures come from secondary sources.

## Sources and fact check

Checked on 28 Oct 2026.

- [Walsh, Winterich and Mittal: Do logo redesigns help or hurt your brand? The role of brand commitment (Journal of Product and Brand Management)](https://www.emeraldinsight.com/insight/content/doi/10.1108/10610421011033421/full/html)
- [Penn State News: Committed customers do not like logo redesigns](https://news.psu.edu/story/161843/2011/01/10/research/starbucks-effect-committed-customers-dont-logo-redesigns)
- [Muller, Kocher and Crettaz: The effects of visual rejuvenation through brand logos (Journal of Business Research, via IDEAS/RePEc)](https://ideas.repec.org/a/eee/jbrese/v66y2013i1p82-88.html)
- [Marketing Week: Marketers judgement of brand asset strength rarely accurate, study finds](https://www.marketingweek.com/marketers-judgement-strength-brand-assets-rarely-accurate/)
- [Marketing Architects: Your brand assets may be working against you](https://www.marketingarchitects.com/blog/v41-n3-your-brand-assets-may-be-working-against-you)
- [The Branding Journal: What to learn from Tropicana's packaging redesign failure](https://www.thebrandingjournal.com/2015/05/what-to-learn-from-tropicanas-packaging-redesign-failure/)
- [The Branding Journal: Learning from the Gap logo redesign fail](https://www.thebrandingjournal.com/2021/04/learnings-gap-logo-redesign-fail/)
- [WARC: X-rated, the negative impact of Twitter's rebrand](https://warc.com/newsandopinion/opinion/x-rated-the-negative-impact-of-twitters-rebrand/6417)
- [Empathy Lab: The distinctive asset dilemma, when brands should change or not](https://www.empathylab.com/en/news-and-insights/the-distinctive-asset-dilemma-when-brands-should-change-or-not)
- [Google Search Console Help: Change of Address tool](https://support.google.com/webmasters/answer/9370220)
- [Google Search Central: Site moves with URL changes](https://developers.google.com/search/docs/crawling-indexing/site-move-with-url-changes)
- [Google Business Profile Help: Guidelines for representing your business](https://support.google.com/business/answer/3038177)
- [AGIP: Trademarks in Indonesia, a complete overview](https://www.agip.com/Agip-Country-Service/en/254/T/2)