Six SaaS rebrands that worked — what changed, and what carried over

Most “rebrands that worked” posts are a before-and-after gallery with the word success underneath. They rarely say what problem the rebrand was called in to solve, and they never say what it cost to move every surface the old brand lived on.

So a definition first, because it decides what is in this post. A rebrand here counts as having worked when three things are true: the company said out loud which problem it was solving, the new brand addressed that specific problem, and the migration shipped without breaking the business underneath it. That is a deliberately modest bar. Public revenue-against-logo charts do not exist, and anyone drawing one is selling something — brand effects arrive mixed in with the product launch, the funding round and the market that happened the same quarter.

Every fact below comes from the company’s own announcement, linked. Where the outcome is not knowable from public sources, this post says so instead of guessing. If you want current identities scored rather than rebrand events, that is the companion post: 12 SaaS brand identity examples.

1. ZenPayroll → Gusto (2015): the name had stopped describing the product

The problem. The name named one feature. The company was about to stop being a payroll company.

What they did. On 17 September 2015 the founders announced the new name and, in the same post, launched health benefits and workers’ compensation: “Today, we are also launching two incredible new services: health benefits and workers’ comp.” The name change and the scope change were one announcement, not two.

What carried over. The product itself — “fully integrated with our original product, small business payroll” — and the positioning that had always been there, that people are the most important part of a business. The company had been founded in 2011 and by then served “tens of thousands of businesses across all 50 states.”

The lesson. A name that describes a feature has an expiry date, and the cheapest moment to change it is the day you ship the thing that makes it wrong. A rename on its own reads as marketing; a rename attached to a launch reads as news.

2. RealtimeBoard → Miro (2019): three months, fifty people, five offices

The problem. A long descriptive name that was hard to say, hard to spell, and tied to a category the product had outgrown.

What they did. Announced on 6 March 2019, with the new name taken from the painter Joan Miró. What makes this one worth studying is that they also published how the work was run: roughly 50 people across five offices, an agency search that reviewed over 100 studios and shortlisted 18, then five weekly sprints with daily standups to produce the brand platform, logo, palette and visual system. Their own line on the pace: “We had five weeks to do developing our new brand platform, initial logo, color palette, and visual system. Most companies spend around a year on this project.”

What carried over. The product, the users, and the URL structure — this was a name and identity change, not a product migration.

The lesson. Speed came from the decision structure, not from cutting scope: leadership in the room for the brand sprint, feedback loops measured in hours rather than weeks. Most rebrands that take a year are not doing more work; they are waiting longer between decisions.

3. Mailchimp (2018): repositioning without losing the weird

The problem. The company had grown past email, and the experience had drifted. In their words, “with so many creative thinkers under one roof, over time we found that parts of our customer experience grew in different directions.”

What they did. On 27 September 2018 they updated “logo, wordmark, typeface, colors, and imagery like photography and illustrations,” and quietly restyled the name itself: MailChimp became Mailchimp, lowercase c.

What carried over. Freddie, the tone of voice, and deliberately, the strangeness: they set out “to retain all the weird, lovable elements that endeared our earliest customers to Mailchimp, while creating space for the brand to grow.” And for users, the thing that matters most in any rebrand of a live product — “everything inside our customers’ Mailchimp accounts will be largely the same.”

The lesson. The equity audit is the first step of a rebrand for a reason. Freddie was an asset; the inconsistency was the liability. Knowing which is which before the design starts is what separates a rebrand from a redesign of everything at once.

4. Slack (2019): the mark that failed in use, not in taste

The problem. This is the rare rebrand where the company published the mechanical reason, and it is the one designers argue about least once they read it. On 16 January 2019 Slack wrote that the original mark “was 11 different colors—and if placed on any color other than white, or at the wrong angle (instead of the precisely prescribed 18º rotation), or with the colors tweaked wrong, it looked terrible.” And: “Every app button looked different, and each one in turn was different from the logo.”

What they did. Worked with Michael Bierut and Pentagram alongside the in-house team on a mark with a simpler palette that “can scale easily, and work better, in many more places.” No name change, no product change.

What carried over. The octothorpe idea, the colour family, and the company’s own framing that a logo should be changed when “it’s not doing the job you want it to do.”

The lesson. This is the failure mode we test for on every identity: a mark that only works at one angle, on one background, at one size. The check costs nothing — render the symbol at 16 px as a favicon and drop it on your own product’s darkest screen. Slack’s old mark was loved and still failed that test; the brand audit checklist puts the same question in one line. Public reaction to the new mark was loud and mixed, which is worth remembering separately: a brand decision that solves a real constraint will still be unpopular for a week.

5. ZEIT → Vercel (2020): the promise that nothing breaks

The problem. A German word meaning “time” was not saying what the platform did, and the company was narrowing its focus to the deploy workflow.

What they did. On 21 April 2020, with a new name chosen to connect to “versatile, accelerate, and excel,” alongside a $21M Series A. The sentence that matters for anyone rebranding a live developer platform is the reassurance to existing customers: “If you’re an existing customer — you’ll find that none of your existing workflows, pricing plans, or projects have changed.”

What carried over. Everything operational. The rebrand changed the name, the mark and the story; the account, the CLI habits and the deployments kept working while the surfaces moved over the following weeks.

The lesson. For a developer product, the rebrand announcement is also a support document. The first question every technical customer asks is “what do I have to change?”, and the answer needs to be in the first paragraph, not in a linked FAQ. The same applies at smaller scale: the 50-surface rebrand checklist exists because a rename touches package names, OAuth consent screens and API docs long after the marketing site is done.

6. Intercom → Fin (2026): the newest one, and the least settled

The problem, in the CEO’s framing. Brand baggage. Announcing the change on 12 May 2026, Eoghan McCabe argued that newer competitors win on perception rather than technology because “they have no baggage. They don’t need to convince anyone of their new position in the market.”

What they did. The company took the name of its AI agent product: “today, we’re changing the name of our company to Fin,” with all 1,400 employees now employed by Fin, while “the name Intercom will still live on as the name of our customer service software platform.”

What happened next. A month later, on 15 June 2026, Salesforce signed a definitive agreement to acquire Fin for approximately $3.6 billion, describing it as “formerly Intercom” with “an established global customer base of more than 30,000 companies.”

The honest read. It is too early to call this one either way, and the acquisition makes it permanently hard to separate the rebrand’s effect from everything else that was happening. It is in this list as the clearest current example of a product name eating the company name — a pattern worth knowing when one product becomes the business.

The pattern across all six

1. The problem was named before the work started. A feature-shaped name, a drifting experience, a mark that broke in use, a category shift. None of them started from “our brand feels tired.”
2. The rename rode alongside something real. A product launch, a funding round, a repositioning. A rebrand with no news attached is the one that reads as expensive redecoration.
3. Equity was decided explicitly. Freddie stayed. The octothorpe stayed. The payroll product stayed. In every case someone wrote down what must survive before anyone drew anything.
4. Existing customers were told what they had to do — usually nothing. That sentence was in the announcement, not in a follow-up email.
5. None of them published a revenue chart afterwards. The companies that did the work most carefully are the ones making the fewest claims about what it earned.

What none of them did

• Change the name and the product and the pricing in the same week.
• Ship the new identity as a PDF and hope the product caught up. (The mechanics of not doing this are in brand identity to design tokens.)
• Rebrand to escape a product problem. Every company on this list had something working underneath.
• Treat the launch as the finish. Slack’s app buttons, Mailchimp’s in-product surfaces and Vercel’s package names are all long-tail migration work that happens after the announcement.
• Ask their customers to approve the new logo.

If you are 20 people, not 2,000

The scaled-down version of everything above fits on one page:

• Write the problem in one sentence before contacting anyone. If you cannot, the answer is probably not a rebrand — the three signals are the test.
• Audit the equity. What do customers recognise? That list is short and it is the only thing you cannot rebuild.
• Attach it to a launch. Ship the new brand with the release, the round or the repositioning that made it necessary.
• Inventory the surfaces first. Ours runs to fifty; yours might be thirty. The rebrand checklist is the same list we work from.
• Say what does not change in the first paragraph of your announcement.
• Budget for the tail. The announcement is the middle of the project, not the end.

FAQ

What makes a rebrand successful? That it solved the problem it was called in for, and that the migration did not cost the business more than the problem did. Public proof of a revenue effect essentially never exists, because a rebrand ships alongside product launches, funding and market movement. Judge it on the stated problem, the equity that survived, and whether customers had to do anything they were not warned about.

Should a SaaS company change its name during a rebrand? Only when the name is the problem — it describes one feature, it is hard to say or spell, or it belongs to a category you have left. Gusto, Miro and Vercel all changed the name for one of those reasons. A name change multiplies the migration: domain, packages, OAuth screens, app stores, integration listings and every backlink you have earned. A new visual system on the same name is a fraction of the work.

How long does a SaaS rebrand take? Miro published three months end to end, with five weeks for the brand platform, logo, palette and visual system — and noted that most companies spend about a year. Our own engagements run the identity in weeks and the migration in a separate phase, because the surface inventory, not the design, is what sets the calendar.

What should stay the same in a rebrand? Whatever customers recognise and whatever they depend on. Mailchimp kept Freddie and the voice; Vercel kept every workflow, plan and project. Write the “must survive” list before the design starts, and treat anything not on it as replaceable.

Do rebrands hurt SEO? A visual rebrand on the same domain does not. A rename that moves the domain does, unless every URL that earned traffic in the last 90 days gets a one-to-one 301 and the old domain keeps redirecting for years. That single mechanism is the difference between a quarter of lost rankings and an unremarkable migration.

---

If a rebrand is on your shortlist, the SaaS rebrand guide covers when it is warranted and what it costs, and the rebrand page is how we run one — fixed scope, published price, the fifty-surface migration included. Not sure yet? Send a URL for a free 15-minute audit and we will tell you whether the brand is the problem or whether it is something cheaper to fix.