About Outcome

I am obsessed with frameworks, mental models, and tools that help creators teach what they know and grow their businesses.

I saw the Gemini X Prize as a chance to explore one question:

How can creators use AI to help their audience actually apply what they teach?

A creator might publish “The Five Steps to...” but every reader is still left asking: How does this apply to me?

That is the idea behind Outcome.

It asks questions, learns about the person, then uses Gemini to combine:

  • Expert Context: the creator’s content, frameworks, and expertise.
  • Personal Context: the lead’s goals, situation, and answers.

The result can be a personal action plan, audit, score, recommendation, or report.

It turns “The five steps” into “Your next steps.”

The goal is to let creators provide the kind of useful, contextual guidance they would normally give one-to-one, but at scale.

What I learned

The biggest challenge was, and still is scope creep.

AI gives you so many new superpowers that it becomes very easy to build every idea you have... and I have a lot of ideas!

I fully embraced agentic coding while building Outcome.

I am a self-taught designer and programmer, and I originally learned to code because I wanted my ideas to become real.

With Gemini, Antigravity, and coding agents, I can spend more time creating useful ideas and testing them with real people, and less time fighting with React.

That comes with a challenge: making sure I still understand how things work, not just that they work.

This changes the speed of impact. Ideas that once took days or weeks to create and test can now be tried in hours.

That gives me room to experiment more, follow my gut, and throw away what does not work without being tied to the time already spent building it. The result is more freedom to focus on what is actually useful and meaningful.

Accomplishments

I started the contest late and only got Outcome live about a week before the deadline.

In that short time, a few hundred people signed up, Outcome was named Product of the Day on Product Hunt, and I had dozens of useful conversations with creators about how they could use it.

That feedback is already shaping the product, and early users are beginning to convert into paying customers.

What’s next

I am going to continue to work closely with users. I am creating more examples, templates, and AI tools that can do more of the work for the creator.

The larger goal is to keep expanding what creator content can do: not just inform an audience, but help each person take action and leave with a real outcome.

Built With

Share this project:

Updates

posted an update

Fun update, I was recommended by a DevPost email about the All Things Agentic Hackathon, and it was the perfect reason to do more AI, and make this startup super duper AI-Native

So I built a GTM Research Agent so I can test assumptions, and start growing based on real research vs. just my hunches - even if those hunches got me here, they may not get me to the next level, where I want to go - more users & revenue. Biggest challenge is copy that explains the product in the language ( and problems ) they are having.

https://devpost.com/software/gtm-partner-collaborator-agent

Really fun few days working on that

Full architecture diagram of the GTM Partner Collaborator Agent

Log in or sign up for Devpost to join the conversation.

posted an update

Post-launch report

Since submitting to the XPRIZE, I've been focused on two things.

First: getting inside my users' businesses.

Not asking what features they want, but understanding how their funnels actually work today. Where leads come from. What happens after someone opts in. What they're selling. And where Outcome fits.

Second: figuring out where AI should run the business itself.

I love building product. I could happily spend all day doing it. Which is exactly why I need to make sure the rest of the business doesn't depend on me doing everything manually.

What I've learned

The split I expected showed up pretty quickly.

My ICP understands Outcome almost immediately.

They already want to give their audience something useful before asking them to book, buy, or subscribe. Outcome gives them a way to do that personally for every lead.

The harder part is everyone who doesn't already think this way.

An Outcome Funnel doesn't fit neatly into an existing category.

An Outcome Funnel uses Gemini to create something different for each person, based on who they are and the creator's expertise.

So simply explaining it better isn't enough.

I need to show it.

That means more examples, better templates, and more education around what an Outcome Funnel looks like for a coach, consultant, creator, SaaS company, or whatever business is using it.

The problem I haven't solved yet: pricing

Pricing an AI-native product is surprisingly awkward.

I can charge for credits, requests, messages, tokens, generations - there are a dozen ways to do it.

The math isn't the difficult part. I know the COGS, marketing costs, affiliate commissions, and the margins I need.

The harder part is making the pricing make sense to the customer.

Nobody wants to think about tokens. Most people don't know what 1,000 credits means.

And they shouldn't have to understand what it costs me to generate a report any more than they need to understand what server it runs on.

They care about what they get.

So I'm trying to attach the price as closely as possible to the value delivered.

Right now I'm benchmarking how much AI usage I can reasonably include before limits are actually necessary.

That decision ends up affecting almost everything: the tiers, the limits, and what someone is really upgrading for.

Outcome launch stats: 3.56K visitors, 5.31K page views, 246 signups, 272 support conversations, 8 active subscriptions, and #1 Product of the Day on Product Hunt with 395 upvotes

Log in or sign up for Devpost to join the conversation.