In early 2024, David Heikka finished his first AI product: a form tool to qualify sales leads for high-ticket businesses. It launched, and nobody wanted it.
Over the following six months, he and his brother Felix tried every remedy they could think of: Facebook ads that wasted more than $800, a different landing page, a new name, and more features. After six months, the product closed. [1]
On August 19, 2024, they soft-launched Buildpad. This time it was not a form tool. It broke “starting a business” into stages and guided users through them, so they always knew what to do next. [2]
Breaking entrepreneurship into the next step
The homepage provides two example projects: a physics-learning app and a candle brand. The latter has specific conditions: sell through Shopify, a budget of about $4k, part-time, and no audience. The product’s AI research agents investigate demand on social media and the web, helping ideas develop around what people actually want. Documents, notes, and research share one canvas, alongside stage tasks and the live project websites shown in the examples. These are website examples, not verified customer cases.
Image: Browser capture from the official page[2] on 2026-11-18, retaining the original interface. The example is not a hands-on test from this research.
The product logic is simple. David recalled in the interview that they believed general-purpose AI kept forgetting projects, starting each conversation from zero. They wanted something that could learn and remember a project as it was built. He traced Buildpad’s beginnings to two judgments: memory, and the idea that users really wanted guidance. Entrepreneurship became a sequence of stages.
Why stages?
That judgment came from their recent stumble. After the first product failed, David compared it with a project that had succeeded earlier: its monthly revenue had reached $5,000, $10,000, and $30,000, then stalled at $30,000, with the whole operation depending on one developer. In July 2023, he enrolled in a course and learned programming for five months, building the AI form in early 2024. They tried ads, landing-page changes, renaming, and adding features. They added actions without changing their judgment about demand. The product was built, but nobody wanted it.
He later said in the interview that these mistakes were not unique to them. Most entrepreneurs, especially beginners, make them: failed products, no users, no money. Buildpad’s answer was guidance in stages, telling users in sequence what to validate next.
For the soft launch, they set a target of 20 active users in two weeks. They thought that was ambitious. They defined daily actions: 5 X posts and 50 replies, with new Reddit posts when X was working. Two weeks later, they had 100 users. A month later, they launched on Product Hunt to a good response, and the first paying customer arrived at 7 p.m. that evening.
About six months after launch, David said in the interview that revenue was then $5,800/month and forecast $7,000–$8,000/month for April 2025. Paying customers came from more than 40 countries, and the platform hosted over 8,000 founders at the time. These were statements and forecasts at the interview date, not current figures or profit. Costs were undisclosed.
From unlocking stages to credits
Buildpad now charges a subscription. Pro is $39/month with 200 credits, said by the company to cover about 5 projects, with unused credits rolling over for one month. Max is $85/month with 500 credits, about 12 projects, one month of rollover, and premium support. Both have a 3-day free trial with 10 credits a day; the subscription starts automatically when the trial ends. After cancellation, unused and rolled-over credits expire at the end of the current billing period. [3] · [4]
It was different at the interview date: 7 stages, the first two free, then a starter plan at $20/month to continue. That historical model cannot be applied to today’s credit plans. Both versions retain a free entry, but the mechanisms differ: the old version let users complete two free stages before deciding to buy; today offers a 3-day trial that automatically renews afterward, requiring users to cancel if they want to stop.
A name changed there and back
Another detour is worth recounting. In October 2025, they bought aicofounder.com and redesigned the product around that name. Nine months later, it became Buildpad again—the same product and team. The announcement said there were about 20,000 users at the redesign and 100,000+ when that announcement was published. [5]
A separate announcement published on February 7, 2026 announced Buildpad becoming aicofounder.com. Buildpad becomes aicofounder.com![6] The July 2026 retrospective records that same product’s journey to the new name and back.
They had recently tried renaming and adding features without saving a product; more than a year later, they changed the brand and visuals again for a new name. This was not necessarily repeating the same mistake. The rename also migrated published websites from .aicofounder.co to .buildpad.site with automatic redirects. Users might need to log in again on their first visit, while projects and data stayed in place. That was concrete migration work. But domains and branding seem to be precisely the part entrepreneurs find easiest to tackle: they have a clear moment of completion, whereas “should this step be taken at all?” does not.
The same work, sold per use elsewhere
Fonda does the same category of work with a very different billing model. It divides entrepreneurship into Discover, Validate, Launch, and Scale—14 steps in four stages—remembering every interview, decision, and pivot. Progress is saved automatically, pivots do not erase prior work, and each step gives just one next action. It sells one-time credit packs that never expire, without a subscription: Validate at $19.99/300 credits, Launch at $49.99/900, and Scale at $84.99/2000, with 40 signup credits. Before each step, it shows the cost and balance and requires confirmation. Failed generation returns credits automatically, and each step includes one free retry. [7] · [8]
Their credit units and per-step costs differ. The source material does not support directly calculating which is cheaper.
The real divergence is not price, but a judgment about how often people return. Buildpad charges monthly, betting that users will still be progressing in month two. Fonda charges by step, betting that people will stop midway and return later to buy the next step. If someone really matches the website example—part-time, about $4k in budget, no audience—it is hard to imagine them subscribing for twelve consecutive months. A more likely path might be getting stuck at demand validation, closing the page, and remembering it two months later. That is only a hypothesis drawn from the charging mechanisms, not user data.
If the product really moves people forward, a subscription makes sense: guidance is an ongoing relationship. The number that could substantiate that judgment is retention, and it has not been disclosed.
Buildpad sells “what to do next.” The business is most vulnerable where a subscription bills monthly but a person may stop at one step. In that month, the product has to prove it is worth keeping.
[1] www.fakemayo.com
[2] buildpad.io
[3] buildpad.io
[4] buildpad.io
[5] buildpad.io
[6] aicofounder.com
[7] fonda.co
[8] fonda.co