In the last two months I have spoken to several entrepreneurs, product leaders and founders. Most of them pour significant amounts of money to build and give shape to their ideas but rarely validate them. This is why we have a high percentage of product failure.
In my latest Certified Scrum Product Owner Workshop in Person in NYC held on August 18-19 2025, one of my student teams came up with a new app with dynamic pricing using AI on parking spots in New York City. Great Idea!! I must say
They called it SpotCheck and the Slogan is "Shortcut to a sure spot" Product Discovery and discussions with each other in class lead them to believe that this is an idea they pursue and bring to life.
SpotCheck Product Discovery in Certified Scrum Product Owner Class in NYC on August 18-19 2025
Although creating an app using AI is trivial, unless AI has context of what you are trying to build, it's really hard for any tool to build anything meaningful. Introducing the concept of Context Engineering
Context Engineering:
If you are looking for a response from AI that goes beyond the basic answers you need to give it context. AI needs information about your particular request that goes beyond basic prompting. What are you trying to build?, are there any wireframes? , can you tell it a little bit about your customer segments? Are there customer interviews that can be fed to the AI?, etc.
Context Engineering is the discipline of designing and building dynamic systems that provides the right information and tools, in the right format, at the right time, to give a LLM everything it needs to accomplish a task.
This document is a great start to provide context but not enough to create a prototype for the product. As a Product Manager you need to think about what would your team need to create a prototype. They would need a Product Requirements Document PRD.
Product Requirements Document (PRD)
A Product Requirements Document (PRD) is
a comprehensive document that outlines the purpose, functionality, and behavior of a product or feature
. It serves as a blueprint for development teams, ensuring everyone understands what needs to be built and why. Think of it as the bridge between a product vision and its concrete implementation
To create a product requirements document (PRD) for spotcheck I turned to Claude - https://claude.ai/new and used the Claude Sonnet 4 Model and I used the below prompt
You are an expert AI Product Manager having worked on successful AI products in the various industries. I like your help in creating a PRD to develop a parking app called spotcheck that uses AI driven dynamic pricing to find parking in New York City. Here is an example article that describes the challenge, different sections and a format for PRDs that you can leverage
The attachment has the product discovery work conducted by the team that should be leveraged to create the PRD
Claude copied the sections from the PRD example and I found them to be inadequate. I continued my chat with Claude to refine the PRD it produced by adding initial User Stories
Can you refine the PRD to include additional user stories and detailed acceptance criteria so that UI wireframes, rules and workflow can be modeled
Additional Information (Resource Requirements, Risk Mitigation)
Actual Outcomes (placeholder for post-release)
Conclusion Section ✨
I then looked around in the App Store for parking apps and I found one that I really liked their User Interface Design. Of course you could a few others or create your own wireframes. These wireframes were just used for the AI to create context and then create its own.
Collage Created by Foxapple
I explored two AI apps Lovable - https://lovable.dev/ and Replit - https://replit.com and use it to develop the prototype. I found these two to be most appropriate to use if you are product manager.
Lovable.Dev
Lovable.dev is a cloud tool that can develop applications by chatting with AI. The landing page and the applications it creates is a good starting point. It is relevant for Product Managers more than developers to build working prototypes that can be further refined. I really like the way it integrates with Github.com to submit all code to a defined repository and then use it to continue development.
Lovable.dev
I started with the following prompt and attached wireframes:
Create a Parking App for NYC. The app is for apple iphones and ipads. The example wireframes from another parking app
The parking app requirements are
"Spot check Product Vision Slogan: Shortcut to a sure spot
Promises: Worry free booking platform, precise hyperlocal, and greater transparency to local parking laws Precautions: Finite amount of inventory, firm grace periods and users must adhere to local traffic laws Instructions: Download app, Make profile, book one time or recurring subscription
Who is it for: Individuals, Drivers and Commuters looking for parking, Other city residents, Parked Vehicle Owners, Carpool commuters, Drivers from Tristate area, City residents who avoid driving due to low parking availability, Visitors from out of town that want to use their cars
What are they looking for: Parking with Transparency and saves time, Find quick parking during morning commute to drop kids offs at school or college, Find defined parking, post parking space to others, Prioritized parking, Reserved spaces, Find and reserve hyper locally with loyalty and Pivot to Driving to save time Why will they buy: Save time, reduce commuter related frustrations
Outputs: App that provides users with platform to reserve parking
Outcomes: Decrease congestion, increase city revenue, less fines, increased safety, Parent have sense of relief knowing they wont get fined during dropoff, Reduce frustration, Reward participation, and reduce customer related anxiety and frustration
The team conducted a jobs to be done interview and found the below
Jobs to be done: Drop off kids, Save time and frustration, Get to work on time, Visit Family on Time
Pains: Late to work many times due to no available parking, Elevated anxiety due to no available parking, Extra expenses by using rideshare alternative, Miss out on family time, and have to wake up earlier
Gains: Less frustration when parking, on the spot parking location, cost cutting vs rideshare, Time block for parking, short term spot, and add minutes virtually
Customer Segmentation: 1. Local residents living in the Bronx, queens and Brooklyn, age range from 25 – 60 years of age who desire predictably and are in need of regularly parking spot overnight. They are recurring users 2. Gig workers aged 22- 40 years of age living in Manhattan or Brooklyn who have a time sensitive need for parking and they are high frequent users 3. Daily commuters who live in NJ, Bronx, Queens, or Brooklyn, they are on the lookout for efficiencies and dislike surprises. They have high usage.
Value to customer: Save money or time, Deliver an emotion, and solve a problem. The team will measure value by using sentiment surveys and monitoring user patterns and reviews
Value to Stakeholders: Increase customer retention by measuring active users
Value to Team: Learning Opportunity and Improve team effectiveness and well being which will be measured by engagement and happiness surveys and sentiment
Now, Later, Much Later Product Roadmap:
Now – Search for open spots as a landing page with map, highlight open spots, Active navigation to chosen spot
Now Product Goal: For local residents who want to get to work on time the app will display real time parking visibility and navigation
Later: Sign up, user creation and user history, Reserve a spot, time for reserved space, ability to pay.
Later Product Goal – For visitors who have a need to park near venues will get to confirm parking reservations nearby
Much Later: Real time Notifications, Virtually add time, and AI powered dynamic pricing.
Much Later Product Goal: For users who are looking for competitive pricing will get a parking spot at lower rates Wireframes from another parking app as an example"
App Generation
Lovable generated an app with two screen
First App Screen:
Frist App Screen
When I entered an address, the start date/time and end date/time and clicked on the submit button, it generated the next screen
Second screen
Second App Screen
The lack of a map was glaring and I chatted with Lovable "how can we incorporate the map of manhattan on this screen?"
Lovable answered with options
Map Integration Options for Manhattan
1. Mapbox GL JS (Recommended)
This is the most professional and feature-rich option for parking apps:
Benefits:
High-quality street-level maps of Manhattan
Custom styling to match your parking app theme
Real-time data and navigation capabilities
Interactive markers for parking spots
Smooth zoom and pan interactions
Perfect for mobile/responsive design
2. Google Maps (Alternative)
Familiar interface for users
Excellent street view integration
Strong location search capabilities
3. OpenStreetMap with Leaflet (Free option)
No API key required
Good customization options
Community-driven data
I asked Lovable to update the code with google maps
Lovable created an implementation plan and I asked it to implement the plan. The result
NYC Parking app with Google Maps Integration
The app is published live at https://spot-on-nyc.lovable.app/ you will need a Google API key to see the map. You can create one for free at Google Maps
Lovable is a decent tool and perhaps my usage is minima. I will continue to play with it more.
Pros: Quickly create beautiful apps, code can be written and visualized very quickly, great integration with Github and ability to publish and share with collaborators
Cons: Although I gave it the complete release plan with all the features, it created just the landing page and search page results. I had to tell it to integrate a map and future features like user profile, reservations and dynamic pricing.
In my next post I am going to use the same PRD and wireframes and try out Replit.com
Share with me what you think? Have you used Lovable before? What advice would you have for my readers?
Anil Jaising, CST®
On a mission to help Entrepreneurs and Product Leaders THRIVE, Unpack Product Innovation with AI Trainer, Product Consultant and International Speaker Follow me for real life case studies and learning videos.
Despite $30–40 billion in enterprise investment into AI 95% Get Zero Return. Here are 3 changes that actually work. A healthcare SVP called me last month. Let's call her Samantha Jones. Her company was about to blow their AI budget on a CRM feature that would fail. The plan sounded reasonable: mine CRM data to find and rate the best leads for marketing and sales. Sales reps were already using GenAI individually for their accounts, so why not build it natively into the CRM? Here's why it would...
Created with Nano Banana He Spent Two Years Building Strategic Skills. Then They Asked Him to 'Vibe Code.' A student of mine had prepared thoroughly. He attended my Scrum Alliance Certified Scrum Product Owner® and Advanced Certified Scrum Product Owner® Workshops and in both my workshops I highlight research about how important strategic skill are for a Product Owner https://conceptsandbeyond.com/key-skills-employers-look-for-in-product-managers-and-product-owners-in-2024/ He'd invested two...
Magic is created when a Strategy, Charity, Passion and AI join hands Strategy A key pillar in our strategy is for our products to create social impact in society. A satellite strategy cannot be complete unless your products are not bringing value at a social level. As a Product Manager if you are interested in maximizing the value for your customers, check out the value pyramid https://media.bain.com/elements-of-value/ Bain Consulting Value Pyramid Charity A charity event we look forward to...
