SaaS Media

49 Rebranding Statistics

49 figures on renaming a company: the changes that pay carry a real reason, and AI answers keep the old name for years.

Author:
Sharné McDonald
Contributors
Vlad Shvets
Date:
October 14, 2026

Bynder asked 1,002 marketers how long a rebrand takes, and the average answer was seven months.1 The old name takes longer to leave. In AI answers we collected on 13 October 2026 about 21 renamed companies and products, 21.7% of the answers naming the brand still used its retired name, nearly always as “formerly” beside the new one.2

Whether a rebrand is worth doing has numbers too, and they say more about reasons than taste. Here are the ones a founder or CMO needs before signing off on a new name, from the decision to what AI engines say afterwards. Planning that whole arc is what Brand Marketing at Empact Partners does, and I’ll point out where we come in.

Highlights

82% of marketers Bynder surveyed have worked on a rebrand, and 57% of them name a refreshed identity as the reason.1
The average rebrand takes seven months and touches 215 marketing assets.1
Mutual funds that renamed toward a fashionable style drew 28% more money than comparable funds within a year, with no better performance behind it.3
Tropicana’s redesigned carton came with a 20% sales drop in seven weeks, and the old carton was back inside two months.4
Meta’s Harris reputation score was 60.0 the year before its rename5 and 59.6 in 2024,6 still in the band Harris calls Poor.
Renames three to six years old still carried the old name in 42.3% of AI answers that named the brand. Past six years it was 1 of 94.2
Google AI Mode used the retired name in 36.5% of answers, ChatGPT in 6.3%.2

Read together, they say a rebrand earns its keep when it carries a change buyers can see, and that the new name needs years of help after launch before it stands on its own in the places buyers now ask.

Most Marketers Have Run a Rebrand, and Most Started With a New Look

Bynder, which sells software for managing brand assets, surveyed 1,002 marketers in 2023. Here’s what they said about the decision:

82% of marketers have worked on a rebrand at some point in their careers.1
57% say the purpose was to update the brand’s identity.1
45% were repositioning in the market.1
41% were following a change in their target audience.1
26% were addressing negative perceptions of the brand.1

The answers were multi-select, so most rebrands carry more than one reason, and the most popular is also the vaguest. “Update the identity” can mean the strategy changed and the look has to catch up. It can also mean the logo feels tired.

Horizontal bar chart of the reasons marketers give for a rebrand: update the brand identity 57%, reposition in the market 45%, reflect a change in target audience 41%, address negative perceptions 26%.
Share of marketers naming each reason as the purpose of their rebrand. Respondents could pick more than one.

Rebranding isn’t a rare decision. UpCity surveyed more than 600 business owners and marketers in the US and Canada in 2022 and found 51% had changed their branding strategy since the pandemic began. 48% handled branding in-house.7

The trigger the marketing surveys skip is the acquisition. Landor’s 2017 study of more than 2,300 acquisitions by S&P Global 100 companies found 74% rebranded the acquired company within seven years.8

Size decided a lot of it: 78% of deals under $99 million were rebranded, against 46% of deals over $5 billion, and 38% of mergers of equals ended in a brand transition.8 It’s 2017 data, so read it as how big acquirers behaved for a decade, not as this year’s rate.

Sharné McDonald
Senior GTM Consultant @ Empact Partners
Ask which of those reasons is yours before anyone briefs a designer. Repositioning and a new audience are business changes, and a new name can carry them to market. An identity that just feels dated is a design problem, and renaming the company to fix it spends the recognition you’ve built on something a refreshed look would solve.

A Rebrand Takes Seven Months and Touches 215 Assets

The same survey puts the average rebrand at seven months from first talks to rollout, and 8% of marketers have worked on one that ran for one to two years.1 The average one updates 215 marketing assets.1

Stat card showing that the average rebrand takes 7 months from first talks to rollout and updates 215 marketing assets.
Average length of a rebrand, from initial talks to rollout, and the average number of marketing assets updated, as marketers reported them.

Asked what was hardest, marketers named the work around the design more often than the design:

Updating marketing assets: 47%.1
Telling the audience: 42%.1
Creative alignment and budget: 36% each.1

I’d add one nobody listed, which is deciding how you’ll know it worked. Hanover Research found 94% of organizations measure their brand in some form,9 but we couldn’t find anyone who publishes how many measure a rebrand against a baseline set before launch. Without one, every result after launch is a story somebody tells.

The words and the coordination are where we spend most of our time on a rebrand. When SystemOne started its rebrand at the end of 2024, a design studio built the new identity and Empact Partners wrote everything around it.

We cut a long feature list into a dozen product and solution pages and ran a review-site campaign beside the launch, so the pages other people publish about the company would carry the new name too.

Investors Reward a Rename That Arrives With a Real Change

Whether a rebrand worked depends on what you measure, and the longest record belongs to the stock market, because a listed company’s new name comes with a date and a share price. The studies that looked at ordinary renames found this:

Horsky and Swyngedouw studied 58 US companies that renamed between 1981 and 1985. Most saw performance improve, and the authors read the new name as a signal that real changes were coming, not as the cause.10
Kot studied Hong Kong renames from 1999 to 2008. Investors reacted well when the new name was announced with a merger, a restructuring or a new line of business, and not at all when it was for clarity or reputation. No rename showed a lasting effect.11

The pattern is a rename as news about the business. Two studies measured what a name can do with nothing behind it, and both read as warnings.

In 1998 and 1999, 95 companies that added a dotcom to their names earned abnormal returns of about 74 percent in the ten days around the announcement, while ordinary renames in the same study earned an insignificant two percent.12

Between 1994 and 2001, funds renamed toward a fashionable style drew 28% more money than comparable funds the next year. Performance didn’t improve, and the money came whether or not the holdings matched the name.3 By the authors’ account, the investors who bought in lost money afterwards.

A new name works as a signal that the business changed. The evidence rewards the change, and the name gets the credit.

The psychology runs the same way. A familiar name is a shortcut buyers use without thinking. Round and Roper’s interviews found people build their own associations for a brand name, good and bad, which the company never put there and loses when it renames.13 A rename asks buyers to give that shortcut up, and they’ll do it for a reason they can see.

Committed Customers Push Back on a Redesign

The clearest experiment on this is about a logo. Walsh, Winterich and Mittal showed 632 people redesigned logos for two shoe brands. The more the logo changed, the worse strongly committed customers rated the brand, while weakly committed customers liked it better.14

That split explains the famous design reversals better than taste does. Each of these changed something loyal customers used every day:

Tropicana, 2009: a redesigned Pure Premium carton reached shelves in January, and the line’s sales fell 20% in seven weeks, by Advertising Age’s count.4 A study on Nielsen grocery data put the cost at $27 million,15 and the old carton was back inside two months.
Gap, 2010: the new logo lasted a week before the company went back to the blue box it had used for more than 20 years.16
Consignia, 2001: the UK’s Post Office group spent £1.5m launching the name and about £1m dropping it 15 months later.17
Sharné McDonald
Senior GTM Consultant @ Empact Partners
The people who complain loudest about a redesign are your best customers, which is why the complaint is worth hearing. Tropicana’s regulars found their juice in a crowded aisle by that carton. Before you change what loyal buyers use to find you, show it to them, and keep something they’d recognize from across the room.

Consistency is the everyday version of the same problem. In Lucidpress’s 2021 survey, 77% of companies said they deal with off-brand content, and respondents estimated a consistent brand would lift revenue by 10–20%, which is their guess rather than a measurement.18

A New Name Doesn’t Wash a Bad Reputation

Harris Poll found in 2023 that 79% of Americans think companies will need to rebrand to stay relevant, and 75% think companies rebrand their products when they should be rebranding their ethics and values.19 The reputation scores back the second half.

Meta and X Stayed in the Poor Band

Facebook renamed its parent company Meta in October 2021. Its Harris reputation score was 60.0 the year before5 and 59.6 in 2024,6 both inside the band Harris labels Poor. Harris’s 2026 list still prints the company as Meta (Facebook).20

Twitter fell from 63.4 to 58.8 over the same years, a slide that showed up in the spring 2022 survey, before the takeover closed and well before the July 2023 rename to X.5 Wells Fargo, a company in the same band that kept its name, went from 63.0 to 63.4.5

Dumbbell chart of Harris reputation scores in 2021 and 2024: Meta, renamed from Facebook in 2021, 60.0 then 59.6; X, renamed from Twitter in 2023, 63.4 then 58.8; Wells Fargo, which kept its name, 63.0 then 63.4.
Harris Reputation Quotient for the 100 most visible US companies, 2021 against 2024. Wells Fargo kept its name and is shown for reference.

Brand Finance puts X’s brand value at $498 million in 2025, 91% below its 2022 peak, and awareness of the brand at 78%, down from 94%.21 Both measures span the takeover and the rename, so nobody can pull the two apart.

Investors read scandal renames the same way. In Unsal’s study of US financial firms with records of misconduct, the market reacted badly when those firms changed their names.22

ValuJet’s Recovery Came With a New Airline, Not Just a New Name

The rename usually held up as one that worked is ValuJet’s. After the May 1996 crash, revenue fell from $367.8 million in 1995 to $211.5 million in 1997,23 the year the company spent $5.2 million renaming itself AirTran.24

By 1999 revenue was $523.5 million, but that includes AirTran Airways, the airline it merged with, and the same years brought business class and new routes.23

Load factor, the share of seats filled, tells the cleaner story: 68.8% before the crash, 52.9% in the rename year and 63.5% in 1999, recovering but still short of where it started.23

Line chart of ValuJet and AirTran load factor: 68.8% in 1995, 57.1% in 1996, 52.9% in 1997, 59.6% in 1998 and 63.5% in 1999.
Revenue passenger miles as a share of available seat miles, from the company’s annual filings. From 1998 the figures include the merged AirTran Airways.
Sharné McDonald
Senior GTM Consultant @ Empact Partners
A rename after a scandal works as the last visible step of a fix, never as the fix. ValuJet’s passengers came back to an airline with a merger partner, a new cabin and new routes, and the new name told them something had changed. Meta’s score sat still, and I think it’s because the new name arrived before anything the public could see had changed.

AI Answers Still Carry the Old Name, Mostly as “Formerly”

None of those studies can see where many buyers meet a brand first today, which is an AI answer. Nobody publishes what happens to the old name there, so we measured it.

On 13 October 2026 we asked ChatGPT and Google AI Mode 42 buyer questions, two for each of 21 companies and products renamed between 2013 and 2024, three times each, and counted the answers that named the brand and still used the old name.2

21.7% of answers that named the brand still used its retired name: 49 of 226.2
Three to six years ago: 42.3%, or 41 of 97 answers.2
Six or more years: 1 of 94.2
Under three years: 20%, or 7 of 35, all of them about Kit. Google’s Gemini (from Bard) and Microsoft’s Copilot (from Bing Chat) never appeared under their old names.2
As the label: only 2 of 226 answers used the old name as the product’s name, both of them about Pardot. Every other time it came as “formerly” or in brackets beside the new one.2
Pardot, renamed Marketing Cloud Account Engagement in 2022, still carried the name Pardot in 10 of 11 answers that mentioned the product, mostly as “formerly Pardot”, and supplies 10 of the 49.2

The engines differ by a long way. Google AI Mode used the old name in 36.5% of answers naming the brand and ChatGPT in 6.3%, and on renames three to six years old Google AI Mode did it 71.4% of the time.2

Bar chart of the share of AI answers naming a renamed brand that still used its old name, by years since the rename: under 3 years, ChatGPT 5.6% and Google AI Mode 35.3%; 3 to 6 years, 12.5% and 71.4%; 6 years or more, 0% and 2%.
Answers that named the brand and still used its retired name, by engine and by years since the rename. Each band holds different brands, so this is not one name ageing.

One day of answers across 21 different brands isn’t a timeline for any single name, and it isn’t a market rate. What it shows is that the old name doesn’t leave when the new site goes live, and that the engines mostly carry it as context rather than as a mistake. Your new name gets introduced with your old one attached.

The reason sits in what the engines read. Qvery is our sister company: it runs buyers’ questions through ChatGPT and Google AI Mode every day and records what they cite, 716,821 answers as of 19 September 2026.25 We own it, so weigh its findings accordingly, and each one below links to the research behind it:

On a set of young-SaaS questions in August 2026, 95.59% of ChatGPT answers came with sources, across 408 runs.26
Listicles were 45.8% of the citations Qvery could classify by type, and 8.62% of all ChatGPT citations against 2.50% of Google AI Mode’s.27
Reddit appeared in 43.17% of answers in the same August set.26
13 of 23 SaaS roundup pages cited on broad questions in September showed no readable update date.28

Our own run caught it at the page level: 13 of 126 Google AI Mode answers cited a page whose title or snippet still carried the old name, and two of those pages were the company’s own, Google’s documentation saying Data Studio and a Salesforce page saying “formerly Pardot”.2

This part is my reading, not a measurement. An engine introduces you the way the pages it fetched introduce you. A roundup written in 2021 says Integromat, so the answer says Make (formerly Integromat), and the bridge lasts as long as those pages do.

Being named on the pages engines read is the core of generative engine optimization, GEO for short, and a rebrand is the moment those pages all go out of date at once.

Put the Afterlife in the Launch Plan

Empact Partners is a go-to-market consultancy for B2B software companies, and Brand Marketing is the workstream that settles what a company says before anyone decides where it says it. We plan a rebrand with the afterlife in it, which adds four steps to the usual launch:

A baseline before launch. We run the questions your buyers ask through ChatGPT and Google AI Mode and record how the answers name you and which pages they cite.
The page list. Every roundup, thread, review profile and help article the engines cite that still carries the old name, your own docs first, because they’re yours to fix the morning you launch.
The launch window. Your own pages change on the day. Review profiles and roundup entries need outreach, which is the mention work our SEO and GEO workstreams already run.
A monthly re-measure until the new name stands in the answers without the bridge.
Sharné McDonald
Senior GTM Consultant @ Empact Partners
The cheapest afterlife work happens before announcement day. Your help docs, comparison pages and review profiles can carry the new name the morning it launches, and two of the pages Google AI Mode cited with an old name were the companies’ own. Everything else is outreach to people who owe you nothing, and that takes months, which is why I’d start it the week you announce.

My verdict, which could turn out wrong, is to judge a rebrand at month twelve rather than at launch, and on two things. Did the reason you rebranded show up in the business? And do the AI answers about your category name you without the “formerly”? If the first is no, the name won’t rescue it. If the second is no, the rename isn’t finished yet.

If you’re weighing a rebrand and want the reason and the afterlife planned before anyone opens a design file, book a call with me, and let’s see whether we could work together.

Sources

  1. Bynder, “Study reveals that marketing teams transform 251 assets over 7 months in the average rebrand”, press release, 2023. The headline says 251 assets; the release’s body and its spokesperson say 215.
  2. Empact Partners, retired names in AI answers: 252 answers from ChatGPT (DataForSEO LLM scraper, web search on) and Google AI Mode (DataForSEO AI Mode results) to 42 buyer questions about 21 public companies and products renamed between 2013 and 2024, three runs each, US English, collected and read 13 October 2026.
  3. Cooper, Gulen and Rau, “Changing Names with Style: Mutual Fund Name Changes and Their Effects on Fund Flows”, Journal of Finance, 2005.
  4. Advertising Age (Natalie Zmuda), “Tropicana Line’s Sales Plunge 20% Post-Rebranding”, article, 2009.
  5. The Harris Poll and Axios, “2022 Axios Harris Poll 100”, report, 2022.
  6. Axios and The Harris Poll, “The 2024 Axios Harris Poll 100 reputation rankings”, article, 2024.
  7. UpCity, “UpCity Survey Results: 51% Of Businesses Have Changed Their Branding Strategy Since COVID-19”, press release, 2022.
  8. Landor, “First quantitative M&A brand study of S&P Global 100 shows 74 percent rebrand post-acquisition”, press release, 2017.
  9. Hanover Research, “The State of Brand Measurement”, report, 2020.
  10. Horsky and Swyngedouw, “Does It Pay to Change Your Company’s Name? A Stock Market Perspective”, Marketing Science, 1987.
  11. Kot, “Corporate name changes: Price reactions and long-run performance”, Pacific-Basin Finance Journal, 2011.
  12. Cooper, Dimitrov and Rau, “A Rose.com by Any Other Name”, Journal of Finance, 2001.
  13. Round and Roper, “Exploring consumer brand name equity: Gaining insight through the investigation of response to name change”, European Journal of Marketing, 2012.
  14. Walsh, Winterich and Mittal, “Do logo redesigns help or hurt your brand? The role of brand commitment”, Journal of Product & Brand Management, 2010.
  15. Lee, Gao and Brown, “A study of the impact of package changes on orange juice demand”, Journal of Retailing and Consumer Services, 2010.
  16. Gap Inc., “Gap Listens to Customers and Will Keep Classic Blue Box Logo”, press release, 2010.
  17. The Guardian, “Consignia consigned to the bin”, article, 2002.
  18. Lucidpress (now Marq), “The state of brand consistency”, report summary, 2021.
  19. Stagwell and The Harris Poll, “Special Report: The 2023 Axios Harris Poll 100 Corporate Reputation Findings”, report, 2023.
  20. The Harris Poll, “The 2026 Axios Harris Poll 100”, article, 2026.
  21. Brand Finance, “X: Navigating decline and the fight to reclaim brand trust”, article, 2025.
  22. Unsal, “Reputational damage and name changes: Impact of financial crimes”, Corporate Governance: An International Review, 2022.
  23. AirTran Holdings, “Form 10-K for the fiscal year ended December 31, 1999”, SEC filing, 2000.
  24. AirTran Holdings, “Form 10-K for the fiscal year ended December 31, 1997”, SEC filing, 1998.
  25. Qvery, production database of AI answers captured from ChatGPT (since 28 October 2025) and Google AI Mode (since 18 December 2025), read 19 September 2026.
  26. Qvery, “First AI Recommendations for a Brand-New SaaS: What the Answers Had in Common”, article, 2026.
  27. Qvery, “Listicles Are the Most Cited Content Type in AI Search”, article, 2026.
  28. Qvery, “How To Get Your SaaS Into The Tech Media Roundups AI Engines Cite”, article, 2026.

Ready
To Connect?

Let's Partner