The claim you have seen a hundred times
“Consistent branding increases revenue by 23%.”
You have read that line on agency home pages, on the second slide of a proposal, in LinkedIn posts about why your logo needs work, in roundups titled something like “brand consistency statistics every business owner should know.” Sometimes the figure is 33%. It usually arrives with no source, and when it has one, the link goes to another blog post that had none either.
We sell branding and graphic design, so take this from a party with an interest in the opposite: no study found that number. Not 23%, not 33%. There is no dataset anywhere of companies whose revenue rose by a measured percentage after they standardized their fonts.
Where it came from
The figure traces to a 2016 report called the State of Brand Consistency, published by Lucidpress, a software company now called Marq that sells brand template tools. The same company published a 2019 edition that raised the figure to 33% and described it as an increase over the 2016 report.
A measured economic effect does not gain ten percentage points in three years because a fresh survey went out, and that is how survey answers behave rather than how revenue behaves.
One admission, by our own standard: we are working from secondary accounts of these reports rather than the originals. A reader who wants to check should search for the Lucidpress or Marq State of Brand Consistency reports directly.
We want to be fair to Lucidpress. A vendor surveying its own market and publishing what it hears is normal, and the reports call themselves surveys. What went wrong came afterward, as agencies and content mills repeated the headline with the word “survey” quietly dropped, until a self-reported opinion hardened into something people cite as a law of nature.
The relatives of the claim have thinner provenance still. “Consistent visual branding increases recognition by about 80%” and “a consistent brand identity produces 33% higher recall” have no locatable original study at all. Follow either one backward and you land in a loop of agency blogs citing each other.
Why a marketer survey is not a revenue study
The 2019 edition is a self-reported survey of marketers, published by a company selling brand template software. No sampling frame was published, so there is no way to know who was invited, who chose to answer, or how those companies differ from yours. No control group existed. No revenue figure in the report was checked against a financial statement.
What the survey collected was what marketers believe about their own work, and people who spend their days enforcing brand consistency tend to think it pays. Anyone answering a question about the value of their own decisions would do the same.
There is also the direction of the arrow. A company with a disciplined brand system usually has discipline everywhere else: written processes, someone accountable for design, budget for both. It grows, and it keeps its logo files organized. A survey built this way cannot tell you which caused which.
What peer-reviewed research measured instead
Serious research on brand consistency exists, and it answers a narrower question than a pitch deck wants answered.
A 2025 study in the Journal of Brand Management built a measure called the Portfolio Branding Cohesion Metric and ran it across the UK packaged goods sector: 9 categories, 125 brands, 1,023 individual products. On average, brands used their own branding elements with only 65% consistency across their portfolios. Cohesion fell where companies ran sub-brands, where products carried private-label status, and where the brand system contained more visual elements.
Notice what that measured: how consistent brands are in practice, not what consistency earns. The authors made no such claim. Most brands are looser than they believe, and the more pieces a brand system contains, the looser the output gets.
What the packaging research found, and what it did not
Design does move buying behavior, and careful work shows it under specific conditions.
A February 2025 study in Behavioral Sciences used an orthogonal experimental design to build 14 packaging prototypes manipulating five visual elements, surveyed 490 tea-bag consumers, and modeled the results with structural equation modelling. Color, graphics, logo, and layout each significantly influenced purchase intention, with the effect running through brand experience. Typography did not produce the same effect.
Two limits belong in the same breath as that finding. This is one low-involvement product category, tea bags, where shoppers decide in seconds, and it is the Chinese market. Certain visual elements of packaging can shift purchase intention under those conditions. That is not a revenue percentage for your business.
The counterweight nobody in this industry quotes
Citing only the research that flatters design work would be the same trick as the 23% claim, dressed in better footnotes.
McKinsey’s June 2025 analysis of its US Packaging Survey, fielded in March 2025 among 1,000 US consumers as part of an 11-country round, found that US consumers ranked packaging appearance the least important of all packaging characteristics tested. Food safety and shelf life ranked highest. Environmental impact was rated extremely or very important by 44%, and even that sat 17 points below durability, which itself came only fifth. Shoppers care more about whether a product survives the trip home than about how it looks on a shelf.
The “72% of consumers say packaging design influences their purchase decisions” line deserves pinning down too. It comes from a May 2018 Ipsos poll conducted on behalf of the Paper and Packaging Board, an industry trade group: 2,002 US adults surveyed in April 2018, credibility interval plus or minus 2.5 percentage points, 72% saying packaging design often influences their purchase decisions and 81% saying it influences gift selection. Competent polling. Also eight years old, paid for by the paper and packaging industry, and you deserved both facts before anyone quoted it at you.
Four questions to ask any marketing statistic
No research training required, and you can ask these out loud in a meeting without sounding difficult.
Who published it?
Ask for the organization, the report title, and the year. A good answer sounds like “a 2025 paper in the Journal of Brand Management.” A bad one sounds like “that’s a well-known industry statistic.” If nobody in the room can name the publisher, the number comes out of the deck.
What were they selling?
Every sponsor has a preferred outcome. A software company surveying marketers about brand consistency has one, and so does a paper trade board polling consumers about packaging. Sponsorship does not void a study, but it belongs on the table where you can see it. A bad answer is the word “independent” with no name attached.
How many people were surveyed, and how were they chosen?
Sample size alone proves little. Selection is where surveys break. A group that opted into a vendor’s survey is a different animal from 1,000 consumers drawn to match the US population. A bad answer gives you a headcount and says nothing about who those people were or how they were reached.
Was the effect measured, or self-reported?
This one dissolves most brand statistics on contact. “Companies reported higher revenue growth” means somebody typed a number into a form. “Revenue was measured against a control group” means something else, and little marketing research clears that bar. A bad answer blurs the two with a soft verb like “saw” or “experienced.”
What to do about your own brand consistency
We have no revenue study for you and will not invent one. What we have is one measured finding, the Journal of Brand Management result above: 65% consistency on average, cohesion dropping as visual elements multiply. It describes the problem without putting a price on it.
The rest is reasoning, and we will label it as such. Recognition has a cost, and inconsistency raises it. When your truck, your invoice, your sign, and your Facebook page each show a slightly different version of your business, a customer has to do a small piece of work to connect them. Most will. Some will not, and you never hear from the ones who did not.
The other half you can check yourself. Count the hours your team loses recreating a flyer because nobody could find the logo file, or debating which blue is the right blue. A small brand system your staff can follow without asking beats an elaborate one they route around: one logo with a couple of approved variations, one color set with hex and print values written down, files organized so the newest person on your payroll can pull the right one.
Brand work sold on invented numbers is a good part of why owners half-expect a pitch to be decoration with a markup. So, the honest version: the 23% figure came from a vendor’s marketer survey in 2016, the peer-reviewed work measures something smaller and truer, and the quieter case still holds. Your business should be recognizable. Your team should not lose afternoons hunting for files.
So do the inventory yourself. Gather every place your brand appears in public (sign, vehicle, invoice, business card, website header, social profiles) and put them side by side on one screen. Most owners find three or four versions of themselves. Pick the one you would keep, write down its exact colors and the logo file name, and replace the others as each item comes up for reprint. That is a year of small decisions rather than a project.