First, I’d clarify which part of Google Maps we want to improve. People use it to navigate, discover places, and understand a location through information, photographs, and reviews. Each of those activities serves a different need.
I’d narrow the objective, choose a user segment, map their journey, and then prioritise the problems and solutions.
1. Clarify the scope and objective
I’d ask:
Are we improving navigation, place discovery, or the overall experience?
Are we focusing on mobile, web, or both?
Is the goal engagement, acquisition, or revenue?
Is there a particular geography or audience?
What are the timeline and budget constraints?
For this case, I’ll focus on the Android and iOS apps in India, with six months and no fixed budget constraint. The objective is to improve daily and monthly active usage.
I’d distinguish DAU and MAU from the DAU/MAU ratio. The first two measure active audience size; the ratio indicates how frequently monthly users return.
I wouldn’t optimise for session duration alone. Someone taking longer to find a destination could indicate a worse experience. As an external PM candidate, I’d also make my assumptions explicit rather than imply access to internal data.
2. Choose a user segment
I’d consider three starting groups:
Segment | Typical use case | Potential opportunity |
|---|---|---|
School-age children and their families | Explore places, learn geography, and participate in family outings | Build a useful introduction to maps |
Domestic and international travellers | Plan trips and navigate unfamiliar places | Reduce uncertainty and improve discovery |
Working professionals | Commute and complete everyday errands | Reduce repeated effort and improve convenience |
These groups overlap. A working professional can also be a traveller, so I’d classify people by their primary use case for research rather than call the groups strictly mutually exclusive.
I’d compare audience size, unmet need, current usage, and likely frequency of return. I wouldn’t assume travellers use Maps only monthly or that professionals have little growth potential.
For this exercise, I’d choose school-age children exploring places with a parent or guardian. I’d narrow the initial pilot to approximately ages 10–14, subject to research and appropriate account arrangements.
My hypothesis is that curiosity about places could create a useful reason for families to return. I’d estimate the reachable audience through households with suitable device access and interest in supervised use, rather than treating all schoolchildren as the market.
3. Understand the journey
I’d use acquisition, activation, retention, referral, and revenue to examine the experience:
Stage | Question I’d investigate | Opportunity |
|---|---|---|
Acquisition | How does a child first encounter Maps? | Parent-led discovery, school activities, or family outings |
Activation | What makes the first experience useful? | Explore a familiar landmark or save a place for a family visit |
Retention | What would bring the family back? | Saved interests, new places to explore, and unfinished activities |
Referral | What would they want to share? | A place or outing idea shared privately with family |
Revenue | How does this support the business? | Evaluate long-term family utility without making monetisation a requirement for the pilot |
I wouldn’t rely on paid reviews, referral incentives, or a presumed contributor revenue-sharing model to encourage children’s participation.
4. Identify and prioritise the needs
I see five possible needs:
Need | User impact | Contribution to the objective | Priority |
|---|---|---|---|
Learn about landmarks and places encountered during travel | High | Creates a reason to explore and return | First |
Remember interesting places for a future visit | High | Connects discovery with repeat family use | First |
Build navigation confidence with adult support | Medium–high | Supports useful participation in trips | Next |
Find and give feedback on places suitable for children | Medium | Helps discovery, but requires careful content design | Later |
Travel independently to tuition or social events | Potentially high | Requires much more than an interface change | Outside the initial pilot |
I’d validate these priorities with children and parents. My first focus would be curiosity and remembering places, because both can be addressed within a supervised experience.
For example, a child sees a fort during a family drive, wants to know its history, and asks to visit it another weekend. Maps could help the family connect those moments.
5. Explore solutions
I’d consider four approaches.
Contextual Maps: Let families save a place with a typed note or an explicitly recorded voice note, then optionally set a reminder. This could support revisiting a destination or returning to a geography topic before an exam. I’d use deliberate saving rather than automatically recording conversations.
Voice-based travel buddy: Let a child ask simple questions or save a place through a clear, parent-enabled interaction. Recommendations would need defined suitability criteria and review. I wouldn’t label a place “certified child-friendly” without a credible certification process.
Kids Mode: Provide a simplified, supervised experience with selected places, easy-to-understand information, and a private family wishlist. Children could help follow a route while an adult manages the journey.
Maps for Education: Connect selected geography or history topics with map exploration. An AI assistant could eventually help interpret a lesson or homework question, but uploading schoolwork, answer accuracy, and curriculum alignment would need further evaluation.
6. Choose the first solution
I’d use RICE to compare reach, impact, confidence, and effort. Without reliable reach estimates or engineering sizing, I’d use a directional comparison before calculating a numerical score.
Solution | Expected reach | Impact on selected needs | Confidence before testing | Effort |
|---|---|---|---|---|
Contextual saving and reminders | Medium–high | High | Medium | Medium–high |
Voice travel buddy | Medium | Medium | Low–medium | High |
Supervised Kids Mode | High within the target group | High | Medium | Medium for a narrow pilot |
Maps for Education | Medium | Medium–high | Low–medium | Very high for broad curriculum coverage |
I’d choose a narrowly scoped Kids Mode for the first six months.
It would include curated exploration, simple place cards, and a private family wishlist. This directly addresses the problems we prioritised. A kids mode focused only on independent navigation would not.
After validating the pilot, I’d consider richer contextual notes and reminders over the following 6–18 months. That would be a later roadmap decision, not part of the six-month commitment.
7. Consider trade-offs and delivery
Parents need control over setup and sharing, while children need an experience they can understand and enjoy. I’d make those roles clear and keep saved places private by default.
I’d also avoid suggesting that Maps can guarantee a route is safe for a child to travel alone. The initial experience would support supervised exploration and participation.
Content quality is another dependency. Place descriptions need to be accurate, understandable, and appropriate for the age group. I’d begin with a limited set of locations and languages, with accessible controls and lightweight content for slower connections.
8. Measure success
I’d track DAU, MAU, and DAU/MAU for the eligible supervised experience, alongside four-week family retention. Daily use may not be appropriate for every family, so I’d also look at weekly return patterns.
Supporting metrics would include:
Parent setup completion and first meaningful activity.
Activation-to-retention conversion.
Places explored, saved, and revisited.
Completed family-planned outings, where measurable with permission.
Success in supervised map-reading tasks.
Session count and duration would help explain behaviour, but I wouldn’t treat more screen time as the goal. On shared devices, I’d measure supervised profiles or family activity carefully rather than infer that every session belongs to a child.
Guardrails would include navigation abandonment, parent-reported concerns, inappropriate content exposure, privacy issues, and any decline in the core experience for other users.
My recommendation is a supervised Kids Mode that helps children explore places and contribute to family plans. I’d expand it only if families find it useful and continue returning, rather than relying on novelty to increase activity.