In the fast paced world of startups, maximizing time, resources and budget are critical. As a busy founder, you may already be using AI to free up time on your calendar and using it to tackle admin tasks.
But what about the important stuff - like launching the right MVP? AI can be a powerful tool to help accelerate the different stages of MVP creation - from upfront discovery to post-launch optimization. Here are a few of our suggestions for how to effectively use AI.
MVP Strategy, Validation & Research
thoughtbot feels strongly that all MVP efforts should be backed by customer research, and using activities from a product design sprint to validate assumptions and mitigate risk go a long way in shaping the right roadmap.
AI can be another team member during this phase and is a wonderful resource to support ideation efforts. For example, it could help brainstorm customer personas by analyzing your inputs around an individual’s goals and challenges. For Example try dropping in “who are some potential user personas for a dog walking mobile application” into ChatGPT. Our Senior Designer, Kevin Kwon wrote this great blog post about how you can run smarter design sprints with an AI PDS Assistant and walks through how to boost each activity with AI.
AI can also help you in these upfront planning stages by supporting your competitor analysis. Crayon, Kompyte, or SimilarWeb can analyze your competitors’ offerings, strengths, weaknesses, and help to identify gaps in the market. These ideas can help you find focus and nail product market fit early on.
MVP Design
You have likely seen new AI capabilities pop up in your existing tools, like Figma and Sketch. For Figma specifically, some of our favorite AI tools include the AI-powered Auto Layout and Smart Resizing capabilities, and a simple but useful background removal plugin.
Fatima Burke, a Designer at thoughtbot, described AI as a “creative collaborator” that can enhance and streamline the creative process and highlighted AI’s accessibility to everyone and the low barrier to entry. Here is her blog summarizing specific tips for using AI to generate design ideas and streamline design workflows.
Mosas Amama, a Senior Designer, also shared his experience using AI as an accessibility testing partner and also a tool for making typography decisions.
Some tips from our designers when using AI:
- Maintain your unique voice and don’t let individuality get lost
- Remember AI is not foolproof, bias-free or inherently creative
MVP Development
Things like GitHub copilot, and ChatGPT can assist developers by suggesting code snippets, completing functions, or even writing full classes based on natural language descriptions or existing code context. Our team has found it helpful to utilize these suggestions when building boilerplate code, and that this can also cut down on debugging down the line.
However, when using AI to assist with coding it is important to do so with caution. AI is fairly good at writing code that works or does a certain thing well. However, there is a big difference between writing code that works and code that is resilient to change; which is a big determining factor in the quality of your code.
AI can write code that works, however it may not be good at writing code of good quality that is resilient to change. If there is one thing you can guarantee in the future of your MVP, it is that the code will need to change as your product evolves.
We suggest treating AI like a pair programmer. Avoid prompts that are large in scope. Instead, utilise your foundational knowledge, best practices and thirst for high quality code to provide AI with more directed prompts. This is how we use AI to improve how we code but maintain quality.
Go-To-Market
HubSpot, Copy.ai, or Phrasee can help generate effective marketing copy for your MVP itself but also supporting documentation, like pitch decks, landing pages, etc. It might not be perfect but sometimes it’s easier to tweak something than build it from scratch. They may also have suggestions for phrases that better match your intended tone and audience.
Speaking of websites and landing pages, tools like SurferSEO or Clearscope have some great capabilities that can allow you to optimize your website for search engines. AI can help identify keywords that will rank highly, so you have some suggestions for how to optimize content accordingly.
Post-Launch
Being able to respond to activity or lack of activity will be critical. Analytics tools like Google Analytics, Mixpanel, or Amplitude now have AI capabilities that can provide deeper insights into user behavior, conversion rates, and retention. AI can predict trends, suggest optimizations, and even identify potential issues in the user flow.
AI, if used correctly, is a transformative tool that can help you streamline your efforts, cut costs, and expand your thinking. When it comes to building an MVP, it’s not just about utilizing the newest technologies, it’s about creating products that solve real problems more effectively, and AI can be a big help in generating ideas and validating solutions.
There are definitely some wins when it comes to bringing in AI, even if it’s just about optimizing your time and expanding your ideas. In our mind, the key MVP strategy remains the same, make sure you nail product market fit through high quality customer research, and then build the simplest, most impactful first version which should include weighing your options when it comes to build vs. buy. If we can help you navigate this stage of growth, please reach out.
Also check out some of our AI resources and our AI in Focus livestream series which is focused on making AI actionable for product leaders.