Freelogo / AI product
An AI logo tool founders can trust
HostPapa owned the market position. The category had trained everyone to expect junk.
Product strategy
AI workflows
Design system
The Challenge
HostPapa
AI product, brand identity generation
Product strategy, UX and UI, AI workflows, design system
The obstacle was trust, and it belonged to the category rather than the product.
Freelogo helps founders and small teams create a brand identity in minutes, for the moment when you need one on day one and cannot yet afford a designer. It is owned by HostPapa and replaces LogoMaker, their product from an earlier era of the category.
They came to us to rebuild it, not repaint it.
The problem
The model was never the problem
Anything built in this category starts already suspected. AI branding tools had trained people to expect the same output twice: templated marks, generic type, something a competitor could generate on the same afternoon. Every tool in the category also made users choose between speed and control, usually by advertising simplicity and then delivering a prompt box that rewards people who already know what to type. A founder gets a fast result they cannot adjust, or a set of controls they do not have the vocabulary to use. Neither produces something anyone is proud to put on an invoice. This is the design problem at the centre of most AI products, and it is almost never a model problem. The interaction has to let a novice get somewhere good without learning anything, while letting an expert direct the output without hitting a ceiling, and it has to show enough of its working that a user believes the result was made for them rather than pulled from a shelf. Agent UX, progressive disclosure and trust design are the vocabulary for that, and they matter more to whether an AI product gets used than the quality of what sits underneath. So the flow asks what the business is and who it is for, in the founder's own words, and does the translation internally. Control is disclosed progressively: a founder in a hurry never meets a setting, a designer finds each one at the point they go looking. Underneath sits a system built to keep thousands of generated variations coherent.
The input
Intent, not prompt syntax
Nobody should have to learn how to phrase a prompt to get a usable logo. The flow asks what the business is and who it is for, in plain words, and translates that internally. The skill of writing prompts stops being the thing that separates a good result from a bad one.
The control
Speed and control, without choosing
Every tool in the category made users pick one. Progressive disclosure means both are available at once: a founder in a hurry gets a result in minutes and never meets a setting, while a designer finds shape, type and colour controls at exactly the point they start looking for them.
The system
Thousands of outputs, one set of rules
An AI product that generates infinite variety needs a design system more than a fixed one does. Components, states and rules keep thousands of generated variations coherent and export-ready, which is the difference between an AI product and a generator.
One senior team, no handovers between them
- Brand
- Identity systemProduct namingArt directionTone of voice
- AI product
- Generation flowPrompt and output designResult curationEdge case handling
- Product
- Editor interfaceDesign systemInteraction designPrototyping
- Website
- Marketing siteFront-end buildPerformanceAccessibility
