Every growth-stage company has a strip of client logos on its homepage, usually grayscale, usually in a size order that flatters whichever name is most recognizable. It’s the single most copied piece of B2B web design in the country, and it proves almost nothing. A logo on a wall means a deal closed once. It says nothing about whether that company is still a customer, whether they expanded their contract, or whether they’d say yes again today.
Renewal is the harder claim to fake, which is exactly why it’s the more valuable one to publish - and almost nobody does.
Why the logo wall wins anyway
Logos are cheap to get and cheap to display: one signed contract, one email asking permission to use the mark, done. Renewal proof takes longer to earn and more discipline to collect - a founder actually has to track it, and be willing to publish the number even in a quarter where it’s not flattering. Most teams optimize for what’s easy to produce before a launch date, not what’s true a year later. So the wall fills up with acquisition and stays silent on retention.
The gap shows up first in due diligence and second in recruiting. An investor doing reference calls or a candidate cold-messaging existing customers will find the retention story either way - the only question is whether the company told it first, on its own terms, or got asked about it cold.
What publishing retention actually looks like
It doesn’t require a dashboard. A single customer who’s been paying for three years, named and quoted on why they stayed rather than why they signed, does more work than twenty grayscale logos. So does an honest renewal-rate range stated once a year, even an unimpressive one, because founders who publish an average number read as more credible than founders who publish none.
The instinct to hide a mediocre renewal number is understandable and backwards. Silence on retention reads to a sophisticated buyer as “worse than they’re telling us,” which is usually a harsher assumption than the truth. The logo wall will always be the easier thing to build. It just isn’t the thing anyone serious is actually trying to verify.