We rebranded Knock2. What the data says about rebrands

We moved knock2.ai off Webflow in late September, and a little over a week later rolled out a full rebrand on the new site: a new icon, a subtler logo, new type and color, and a new line, Move First. This is what we changed and why. It is also what I wish more rebrand posts did, which is separate what the research supports from what agencies quote, because the evidence that a rebrand drives revenue is thinner than most people admit.

Why we rebranded: Move First

Everything in B2B go-to-market comes down to timing. The buyer who read your pricing page this morning is in a conversation with someone by Friday. If you know who they are while they are still in that window, you can show up first, and in most deals showing up first matters more than showing up with the cleverest pitch.

That is what Knock2 does: it identifies the companies and people on your site and routes them to your reps, your CRM and your Slack while the visit is still fresh. "Move First" is that sentence compressed. It sits under the wordmark as part of the logo lockup, and it is the test we apply to everything we ship. The old brand did not say any of that. It was friendly, and it could have belonged to any software company.

What changed

The idea: something switching on and reporting what it found

We did not want a brand that decorates the product. We wanted one that feels like it: instruments, not illustration, and evidence, not adjectives. That produced three rules we design against.

• A black canvas with one green signal. Green means detected. It is a state, not a decoration, so it appears once per view as a status dot, a single word or one edge.
• Say the finding, do not dress it. Condensed caps at the top of a page, mono type for labels and data, nothing ornamental in between.
• Show the record. The identified-visitor card (name, title, page viewed, timestamp, ICP score) instead of stock imagery.

Icon and logo

The biggest visual change is the icon. The old one was a "Kn" with a door in the "n". The new one is the K2 inside corner brackets, the viewfinder motif that frames headlines across the site. Small sizes get a plainer version, because brackets do not survive at 16 pixels.

The logo change is much subtler. We kept the wordmark and dropped one detail: the small door-handle dot inside the "n". At the sizes people actually see it (a browser tab, a Slack avatar, an email header) it read as noise rather than a door. The same trimmed wordmark is now on the website, the dashboard and our emails.

Type and color

The old site ran on Thicccboi and DM Sans over a violet-tinted dark, with a green-to-blue gradient on headlines and buttons. The new system gives each typeface a job: Anton in condensed caps for the shout, DM Sans for anything you read in sentences, and IBM Plex Mono for labels and data.

Color got the same discipline: near-black instead of a tinted dark, and a single green, #4ADE80, that appears only when something has been detected. The gradient is demoted to an edge or an underline.

The website: we moved first, then rebranded

We did this in two steps on purpose. First we moved the site off Webflow with the look unchanged, same pages and same URLs, so the only thing that could go wrong was the move. A little over a week later the rebrand went out on top of it. That order is why the second step was manageable. A rebrand is mostly repetition, the same decision applied to every page, and the old site made that expensive: about a third of its pages were unpublished drafts, and eleven comparison pages were really one layout with different names pasted in. On our own code, that layout is one template and a data file, so changing the look means changing it once. The full story of the first step is in how we moved off Webflow.

Does a rebrand pay? What the research supports

I went through the usual citations looking for a controlled study showing that a B2B rebrand lifts revenue. I did not find one. Here is what the sources actually say.

• "Consistent branding means 23% more revenue" is an opinion poll. The 2016 Demand Metric report, sponsored by Lucidpress (a brand-management software vendor), asked 234 qualified respondents how much they estimated revenue would rise if their brand were always presented consistently. The answers ranged from 5% to 50%, with an average of 23%. That is an estimate of an opportunity, not measured revenue.
• Rebrands do not reliably pay off in the short run. Brand Finance's study of about 3,000 acquisitions found rebranded deals averaged -2.1% sector-adjusted returns after a year, against -0.4% for deals that kept the old brand. Rebranded outcomes were less spread out, but the study says it "does not identify whether rebrands deliver value in the long run." It is acquisitions, not B2B software, so use it as a caution, not a benchmark.
• Brand works on the buyers who are not buying yet. John Dawes of the Ehrenberg-Bass Institute, in a paper for LinkedIn's B2B Institute, reasons that if a category is bought about once every five years, only around 5% of buyers are in the market in a given quarter. He is explicit that "the 95% figure is not meant to be a precise rule" and is a heuristic. The point for a rebrand: it is a memory job, and a distinctive look is worth something only if enough future buyers see it often enough to remember it.
• Speed is where the numbers get harder. Portent's 30-day analysis of 14 B2B lead-gen sites found a site loading in 1 second converted about 3x better than one loading in 5 seconds. It is an agency's correlation, and it does not say which speed metric it used. Google's Renault case study (over 10 million visits, automotive leads) found a 1 second better Largest Contentful Paint went with 13% more conversions. Neither is a controlled test, and neither is B2B software, but the direction is consistent.

So the honest summary: a rebrand on its own is not a growth lever. One that makes you more recognizable, easier to explain, and faster to load for buyers who are not in market yet might be.

Should you rebrand? A quick test and a baseline

1. Does your brand misstate what you do today? If the product outgrew the story, fix the story.
2. Can you ship it everywhere at once? A half-finished rebrand is two brands. We shipped the new icon, logo, type and color across the site, dashboard and emails in one release for that reason.
3. Can you change the site cheaply afterward? This is where the platform matters.
4. Did you baseline first? Without a before, you cannot say what changed.

Before you launch, record branded search impressions and clicks (Search Console), demo and trial conversion from your top landing pages, real-user page speed on mobile and desktop, win rate and sales cycle for deals that touched the site, and pipeline by first-touch source. Use equal before and after windows, keep the counts behind every percentage, and write the result up as "what changed after launch" unless you ran an experiment. A rebrand also changes your traffic mix, so compare like with like and give it a full sales cycle.

If you want to see the thing the brand is built around, start with the plans and try it on your own site. For another look at how we think about what the product does with a visit, read why not every page visit is intent and why our agents live in Slack.