When our Innovation team, a blend of Product Managers and Data Engineers, met at our home office in Philadelphia, we had a clear target: to fast-track a prototype combining GenAI with our data platform Supernova in one day.
This project was more than just a technical exercise; it was a proof of concept to show how adding GenAI to our foundation of data could solve problems.
Let's dive into how we did it, from picking a problem to building a solution that could change the game.
Mapping Success
The key ingredient to our rapid prototyping was preparation. Without spending time planning, sketching out ideas, and aligning on requirements beforehand, spinning up a prototype in a single day would've been just wishful thinking.
It's like prepping for a marathon; you don't just wake up one day and run 20 miles. You train, plan your route, and maybe even do a few practice runs. That prep work was what really brought our rapid prototyping project across the finish line.
[TIP] Use GenAI to help with your planning, preparation, and helping to score requirements as nice-to-haves and must-haves.
The Challenge
We chose to focus on solving an SEO problem - finding quick wins for search terms in striking distance through keyword targeting and metadata optimization. Our problem wasn’t about reinventing the wheel; it was about making it spin faster and more efficiently with GenAI.
Laying the Groundwork
Another key ingredient: this day of rapid prototyping wasn't a standalone effort. It was built on a foundation laid over years—Supernova, our data platform, didn't happen overnight.
We had clean, enriched, and trusted data at our disposal, a vetted process from a mature SEO team, and a backlog of ideas we’d already been testing to continue to fine-tune that process. This deep reservoir of resources and knowledge was our springboard for the day's challenge.
Our problem to solve was also a specific one that allowed us to focus on a subset of opportunities (striking distance) instead of having to consider many different types of opportunities and create approaches for each type.
Dividing to Conquer
Dividing our forces was key. By breaking into specialized teams, we could tackle the project from all angles simultaneously. This method wasn't just about efficiency; it was about leveraging our collective expertise to cover more ground, faster.
We split up into the following groups:
- Workflow & UI: how a user would select what data should be used in recommendations
- Data transformation (input): how we would transform the data that should be sent to GenAI, based on the user selection
- AI: the specific prompts used to read the input and provide an output
- Data transformation (output): How the data should be returned to the user
Where Tech Meets Teamwork
An important part of our strategy was knowing when to leverage GenAI effectively. It wasn’t a blanket 'let AI handle it' approach; instead, we made deliberate choices about where AI could truly enhance our work, where another tool was the best solution, and where a human touch was needed.
We used data transformation techniques, not AI, to scale keyword research like prioritizing & mapping search terms to URLs, surfacing related synonyms and long-tail search terms, and to pull out the most important details to send to AI as inputs to inform its recommendations.
On the output side, we made sure to include several options for the AI’s recommendations along with the ability for users to tweak or edit the output. This wasn’t about sidelining human input for the sake of technology. It was about using GenAI as a tool where it made the most impact, enhancing our capabilities without compromising the quality or integrity of our product.
Fast Innovation, Real Results
While SEO was the prototype's focus, the real takeaway is how adaptable GenAI can be. It was a proof of concept for a new way of working and solving problems. And it's just the beginning. With GenAI, we're not just speeding up processes; we're opening the door to endless possibilities for innovation and improvement.
Building on Solid Ground
This experience underscored a crucial lesson: innovation isn't just about having groundbreaking ideas; it's also about building on what you've already created. Our rapid prototyping wasn't a magic trick but the next logical step in our ongoing journey of exploration and improvement, powered by the robust capabilities of Supernova and the transformative potential of AI.
Your Blueprint for a Prototype in a Day
Inspired by our journey? Ready to embark on your own day of rapid innovation? Here's a simple template to guide your team, harnessing the power of AI to accelerate your projects:
1. Set a Clear, Achievable Goal
Identify a Specific Problem: Choose a focus that's both impactful and manageable within a day.
2. Build on Existing Foundations
Leverage Your Current Assets: Evaluate what tools, processes, or data you already have that can be used or adapted for the project.
3. Divide and Conquer
Form Specialized Teams: Break your larger team into smaller groups, each with a specific focus area (e.g., data analysis, GenAI integration, UX/UI improvements). This approach allows for parallel progress and leverages diverse expertise.
4. Integrate GenAI Thoughtfully
Identify Opportunities for AI Enhancement: Look for ways GenAI can improve or streamline your processes and where it’s the right tool for the job.
5. Plan for Collaboration and Review
Facilitate Ongoing Communication: Ensure there are mechanisms for teams to share updates and integrate their work. Set aside time for review and refinement of the prototype.
6. Reflect on the Process
Document Learnings and Next Steps: After your Prototyping Day, gather feedback on what worked well and what could be improved. Plan how the prototype can be developed further or how the process could be refined for next time.
7. Share Your Success
Communicate Your Achievements: Share the outcomes of the day with the wider organization. Highlight the process, the product, and the potential implications for future projects.
Takeaways for Your Prototyping Day
A Prototyping Day is more than just a chance to build something new; it's an opportunity to galvanize your team, experiment with cutting-edge technologies like GenAI, and accelerate innovation. By starting with a solid foundation and maintaining a clear focus, your team can achieve remarkable results in just one day. Remember, the key to success lies not just in what you build but in how you build it—leveraging your existing assets, fostering collaboration, and embracing the potential of GenAI to redefine what's possible.
