How would you improve Google Maps?

  Google
Add Your Answer
Answers (4)

Approach to the product improvement question 

1. Describe your understanding of the product, features, users and business model of the company (revenue generation)

2. Clarify with interviewer whether he has any expectation on

  • Any Specific Pain Point to address

or

  • Is it to improve an existing feature
  • Add a new feature to improve the product overall
  • Improve a platform (Web/Mobile)
  • Improve a particular segment of Users?
  • Improve a company goal: Brand, Revenue, User Aquisition, Traffic
else (if the interviewer is ok for you to identify a pain point)
  • State what you see as the BIG pain point

3. Identify the use case to BIG PAIN POINTS or which have BIG IMPACT

  • Usecase
  • Solution
  • Value to the customer (High/medium/low)
  • Development/deployment Complexity (High/Medium/Low)

4. Prioritise the Usecase/Solution

5. Explain how to measure the success criteria related to the feature you described

6. Wrap Up – Impact to the Goal, Alternate Solutions, Self-Assessment – Problems with existing solutions

——————————————————————————————————————

1. Describe your understanding of the product, features and the users

Google Map helps users

1.  Suggests an optimal route to move from Point A to Point B using various transportations.

2. Helps in searching various point of interests with ratings and location address

3. Automatic Re-Routing and alternative route suggetions

4. Rate a place

The Company generate revenue via – Advertisement on their platform

The Main user segments are – General commuters, Travellers (Tourist, Business travellers), General users searching for Bricks and Motors, private vehicle owners, Bikers, people using ride-sharing Apps etc

Main Pain Point I see is for Traveller who visit new places or countries they have issues related to

  1. Clear Guidance in terms of places to visit (Historical, Shopping malls, Pubs, Restaurants)
  2. Language – Communication
  3. Clear Guidance in terms of places to visit

The Solution

Usecase & Solutions – Google Smart Guide for travellers

Usecase Solution Value To Customer Complexity
List top Locations to visit (Based on User Behaviour) When the user selects to generate “Guide Button” Google Maps Generate a list of places to visit based on user ranking under various segments (Place to visits, Pubs, Restaurants, etc) Medium Low
Intelligent Itinerary Once User selects the prefered place to visit in each segment, it consolidates and automatically generates an itinerary connecting various places and approximate time at each places and the medium of transport between them. Multiple options are provided to the user and the user can cancel or add places, which results in re-creating another optimal itinerary.
Itinerary can be modified based on available time.
Medium Medium
One-Stop payment guide Tour Intelligent Itinerary can also have complete transportation assigned, by linking with prominent cab services like Uber or reliable government service, with a single booking (includes transportation and tickets to site seeing). Itinerary can be modified based on users budget allocation. High High

Now Google becomes the local guide, the interaction of the user is more with google maps rather than a local person, this eliminates the communication problem due to different language.

Feature development prioritization

The priority will be as per the above table, as the first two needs to be developed before the last point to be achieved, so I start from the first item.

To get user attention: whenever google sense you have travelled abroad/ or far away distance and opens google maps, a Pop-Up nudge is presented of the new features with the top list of places and restaurants to visit during the time he is in. (Google can also use the date, place and time by the flight travel itinerary and plan an itinerary which spans few hours to few days)

Success Criteria: (specific to the feature you are suggesting)

1. We can see how many people have clicked on the Nudge and the session time per user
2. How many people repeatedly visited the features each time they visited
3. How many people actually travelled to the place which was suggested (using GPS location)
last but not the least
4. Revenue generated per user based on commission made using this feature
Wrapping: (customer first then revenue)
1. This solution will provide users with a very optimal, safe and realtime guide… we can call this app as Google Local Guide.
2. Users can feel safer as google has integrated with credible transportation and is tracked.
3. Users can less worry about arrangements and relax more with friend and families as the alternative transportation arrangements are made in case of delay.
4. This will open up additional stream of revenue (from restaurants, Site seeing places, Transportations, Hotel Stays).
Self-assessment:
Places, where there is no dependable transportation ex: Uber, Lyft or in developing countries and rural areas the Google Guide will have challenges in providing an End to End solutions.

I am assuming, objective of improving Google Maps is to increase user engagement.

To do so, there are a couple of things that we can look at:

  1. Improve user interface
  2. Add a new feature

I would like to discuss adding a new feature unless you want to discuss improving UI

One feature that I believe would increase unser engagement is having google maps for indoors for malls.

Let’s first discuss why I think this is important than we will discuss the feature in detail

WHY:

It’s daunting for mall visitors to find the store they are looking for because of the huge size of mall and new store constantly opening

 

WHAT:

For indoor maps to work I can think of following feature in the priority order:

  1. collecting mall’s information
  2. updating google maps app
  3. updating in-mall directory
  4. Moonshot – smart notifications
Collecting information:
To collect information on where each store is in the mall, we can have robot on each floor. These robots will also collect information on events like seasonal stalls, wet floor, under construction etc.
Updating Google Maps:
If a user has Google maps installed on the phone and they have notifications turned on, we can notify them about this new feature as soon as they enter the mall. Notification can be something like “Easily find the store you are looking for”
Updating Mall Directory:
Often malls have directories installed at multiple places which is just a blueprint of the mall, which has to be updated constantly as and when a store closes or a new store open, we can provide date (collected by our robot) to this directory and replace static image with a digital image so that this directory is always updated. on top of the directory we can write very bold and clear “Powerd by Google Maps” along with this message “Open Google maps on your phone to find the store you are looking for”.
Moonshot:
This is a crazy idea that’s the reason I have it last on my priority list, suppose a user enters a mall that has Macy’s and MAC cosmetics store. This user has a shoping list on Google keep that has Lipstic. As this user walks towads Macy’s, she passes by MAC store, as this person crosses MAC store, Google Maps gives her a notification that “you can buy lipstic from MAC cosmetics which is on your left”
To summarize, to improve User engagement on google maps, I would introduce interior maps for malls which will allow users to find stores they are looking for along with events happening in the mall.

First off, I’d like to outline my understanding of the product to make sure I fully understand all the important elements.  I understand that google maps has a few different use cases

  • It provides maps of pretty much the entire world. The maps have various levels of detail, traditional street maps, satellite overlay, and terrain map. It also allows you to look around in 360 with google street view.
  • It allows people to navigate places, both on desktop as well as turn by turn navigation while travelling in a car, bike or walking.
  • It enables users to search for, discover and find places of interest. Whether that be a business, a restaurant, tourist attraction. Furthermore, it allows users to look at reviews and find out important information about those location like hours of operation. It also has various integrations, such as allowing you to make bookings at restaurants or hotels.

So to summarize, it provides three things

  • General mapping
  • Navigation
  • Discovery, information and interaction with points of interest.

Does that sound right?

Interviewer: Sure that sounds right.

Great.  So, it’s clearly a very big product, and for the purpose of this exercise, I’d like to narrow it down to focus on just one area of the product.  I’d like to focus on improving the point of interest section. However, even within that part of the product that are a lot of different features, so I want to narrow it down even further. I’m going to focus on improving how people interact with the restaurant feature of google maps.  I want to focus on this part of the product for two reasons

  1. It’s an area where there’s a lot of opportunity for improvement. A lot of potential upside in terms of functionality.
  2. Competitors currently hold a significant market share for restaurant-related searches and interactions. For example, Google is currently losing out on a lot of traffic and engagement to services like Yelp and Foursquare that they could capture with enhanced features on their own platform. This is a big missed opportunity for Google.

Does that sound good?

Interviewer: Sure, let’s do it!

Okay, great. Next, I’d like to focus on a metric to move through this.  I definitely think we should be focusing on engagement here. When I think through the main categories of metrics (acquisition, activation, engagement, and monetization), Google has already acquired and activated a huge portion of the market.  That is, almost everyone has looked at a restaurant on Google Maps at some point in their lives. However, a lot of people (perhaps even the majority of people) still go to products like Yelp, and Foursquare for restaurant discovery and reviews, they go to Opentable for reservations and places like Doordash and Postmates for delivery. They could be doing all of that on Google maps. Therefore I want to focus on increasing the engagement. Broadly, our goal is to increase the frequency that people engage with google maps. Specifically, I would measure this by looking at the number of monthly active users who make at least one search for restaurants on Google maps on a daily, weekly, and monthly basis. I’d also want to increase the average number of searches per user on a monthly basis.

I also find it helpful to focus product improvement efforts on a specific user group.  Broadly speaking, we can segment users in terms of how often they’re already using the restaurant features of google maps.  On the high end we have people that are already engaging with it really frequently, let’s call them ‘power users’. On the low end we have people that aren’t engaging with it at all. Let’s call them inactive users’. Then in the middle, we have people that are engaging with occasionally. Maybe they’ve played around with it a bit, but they’re not using it as their primary go-to for the types of searches and value that it can provide.  Let’s call them occasional users. I want to focus on increasing engagement among that middle segment. The occasional user of google’s reach data around points of interest.

Does that sound good?

Interviewer: Sounds great.

Great.  So next what I’d like to do is think about some pain points and problems that users are currently facing. Do you mind if I  take a minute or so to think of some?

To start with, I’ve mapped out the user journey for someone going out to eat at a restaurant. It looks like this:

  • Decide to go out for dinner
  • Find an appropriate restaurant
  • Coordinating with others.
  • Make a reservation (or recommending when to arrive if it’s walk-in only)
  • Choose the best meal
  • Eat
  • Provide a review
  • Share what they’ve eaten with others on social media

Now I’ll think about some pain points in that user journey. I’ll also prioritize them based on 1) impact to user 2) impact on engagement

User journey step Pain point Impact to user
Decide want to go out for dinner Not applicable to this feature
Find an appropriate restaurant Overwhelmed with choice. Can’t choose a restaurant. . High (very common issue)
Coordinate with others Difficult to coordinate with others in choosing a restaurant that everyone. Finding the right time, the right type of food, and the right location Moderate (only an issue some of the time)
Make a reservation or get a table at a  walk-in restaurant Don’t know when to show up to easily get a seat at walk-in (no reservation) restaurants Moderate (only an issue for some restaurants some of the time)
Choose best items on menu Can’t decide what food to order High

(very common issue)

Eat Not applicable to this feature
Provide a review Don’t know what review to write Low (not that painful to the user. Bigger issue for the platform/restaurant)
Share what they’ve eaten on social media Not sure what to write / difficult to take a good photo Low (only an issue for small portion of users)

So based on this prioritization, I’m going to pick the following 2, which were rated as high impact on the user

  • Being overwhelmed by choice when choosing a restaurant.
  • Deciding on what food to order.

Now I’m going to come up with some solutions for each of these pain points. Please give me a couple of minutes to come up with them

One of google’s competitive advantages and core competencies is in building algorithms that incorporate multiple data sources to rank order things by preference. That’s what search is all about.  So why not apply these to the user pain points we identified.

Choosing a restaurant. 

  1. Selection of top three restaurants you should try next based on prior selections and trending restaurants
  2. Showing if your friend went to the restaurant and what they rated it as.
  3. Showing most popular restaurants in area in general

Deciding what dish to order

  1. Suggest dishes based on prior dishes you’ve ordered and ratings by other people. For example, right now google maps shows you popular dishes, we could easily take this one step further and highlight 1-2 items and say “recommended for you”, and even put a match amount on it.
  2. Have people fill out a questionnaire where they answer a bunch of questions, and then get restaurant and dish recommendations based on that. You could even have email program and push notification if there’s a new restaurant or dish that has a strong match for your preferences.
  3. Show dishes your friends have ordered or rated highly and what they rated it as.
  4. Show most popular dishes at the restaurant.

So I’ve identified 7 possible solutions. I’m immediately going to cross off a few here that I don’t think are worth.  Most popular dishes and most popular restaurants already feature I believe. So no point looking further there. Also, option 5 (have people fill out a questionnaire to identify preferences) is more an implementation detail of option 1 and 4 (recommend top restaurants and dishes), so I’ll merge this into those solutions.

So that leaves me with 5 solutions that I’ll rate on two dimensions: 1) impact on engagement and 2) level of effort to build

Solution Impact on engagement Effort to build
Show top three restaurants you should try next based on prior selections and trending restaurants.

In terms of the interface, I’m envisioning a card at the top of the restaurants tab showing the top three restaurants, and a match rate, similar to what Netflix does when showing match rate for shows.

As for the data, we already have trending restaurants based on geolocation data, number of ratings, and number of views of the listings. We can use the same framework for inferring that an individual has visited a certain restaurant, and also use ratings when they’re made by the user.

As mentioned, we can also have the user will out a profile/questionnaire to indicate their preferences, and then rank score restaurant match based on that.

High because this directly addresses a top problem Medium. Have all the data already (or can infer), but need to make new algorithm
Showing if your friend went to the restaurant and what they rated it as.

I envision that this would be something similar to how facebook shows when a friend likes a post or page.

Medium

Relies on people friend’s and is not going to necessarily tell people what they specifically like.

Medium

Problem is that it relies on friends to actually rate the restaurants, and that will may take a long time to get sufficient data

Suggest top 1-2 dishes to order at a restaurant based on prior food you’ve ordered, and ratings by other people. For example, right now google maps shows you popular dishes, we could easily take this one step further and highlight 1-2 items and say “recommended for you”, and even put a match amount on it.

As mentioned, we can also have the user fill out a profile/questionnaire to indicate their preferences, and then rank score dish/menu item match based on that .

High

Would be a very helpful and often used feature if built

High

It relies on users and others to enter what they ate. Google has a way of getting a sense of popular dishes for the general public based on reviews and uploaded photos, but difficult to do for an individual if they’re not doing reviews or photo uploads themselves.

Show dishes your friends have ordered or rated highly and what they rated it as

I envision that this would be something similar to how facebook shows when a friend likes a post or page on facebook.

Medium.

Not going to necessarily tell people what they specifically like.

Medium

Relies on setting up a way to rate/like specific dishes, which could be difficult.  But once have that data, it would be relatively simple to implement.

So based on the prioritization, I would choose to focus first on showing people the top three restaurants to try, because it had high impact and medium effort to build. As part of this solution, we can also implement the questionnaire/profile to determine people’s preferences for restaurants and food.

Finally, I want to look at how I would measure success post-launch.  I would look at

  1. Total number of restaurant searches per month
  2. The median number of restaurant searches per active user per month.

Both of these would give me an idea for whether this feature is driving engagement.

So to summarize the full exercise:

  • I’d focus on improving engagement within the restaurant feature of google maps.
  • I identified two major pain points: 1) Being overwhelmed by choice when choosing a restaurant and 2) deciding what food to order.
  • I identified 6 solutions, and narrowed it down to one: Building out an algorithm to match restaurants to user’s preferences, and then displaying the top 3 recommended restaurants.
  • I would measure success by looking at two metrics: 1) Total number of restaurant searches per month 2) Median number of restaurant searches per active user per month

Here is my approach to this Google product improvement PM job interview question. There are eight steps in the framework:

  1. Describe Google Maps – Google Maps allows user to go from Point A to Point B efficiently and provides several commute options such as driving, walking, public transport, and even like to ride sharing apps such as Uber and Lyft. User also uses Google Maps to search nearby places such as restaurants, gas stations, events, and things to do.
  2. Ask Clarifying Questions from the Interviewer –
    1. Is the goal to improve the entire platform or a specific feature?
    2. Is the improvement channel specific – Web or Mobile?
    3. What’s the objective of improvement? – engagement, increase in revenue, acquisition. etc.
  3. Once interviewer provides clarification to above question, I would start by first laying out the approach I am going to take to (a) start by understanding who are users (personas) of Google Maps are, (b) Pick one persona and what their goals or motivations are when using Google Maps. (c) Then I will identify pain points (user needs) and suggest new features/ solutions to increase customer satisfaction leading to better customer engagement (d) next, I will prioritize new features based on cost vs benefit, (e) Lastly I will summarize the overall analysis
  4. Users of Google Maps
    1. Private vehicle owners
    2. Bikers
    3. Commuters using public transport
    4. Walkers
    5. Tourists
    6. People using ride sharing app like Uber, Lyft etc.
  5. I am going to pick commuters using public transport such as train, bus, subways to go from point A to point B. Commuters hate when there is interruption is services that could be because of trains running late, major breakdown, change in train timing and route. People don’t like waiting even if the train or bus is late by couple of minutes. Also not all train and bus stations are same when it comes to amenities such as access to elevators, public bathrooms, and information booth. Commuters such as old age people, family traveling with infants, and physically challenged people need these information in advance to plan their trip efficiently.
  6. Considering the above pain points that I just mentioned, here are few feature/ use cases that Google Maps can offer to delight the commuters
    1. Personalized real-time notification to commuter when there is interruption in the service (this is especially useful to people commuting daily say from home to office and back). Google knows some very vital information about the commuter like at what time they start from home, from which station they take train, what time they head back home. These information can be used to send real-time train/bus service alert to commuter, so that they can prepare in advance, say may look for alternate route of mode of transportation.
    2. Book uber/ lyft for commuter, to the nearest alternate train/bus station, so that they can reach their destination on-time.
    3. Train/ bus station information catalog showing all amenities and contact information. The catalog should get updated on regular interval,. So that commuter is aware of the change say elevator at station X is going through periodic maintenance
    4. Collaboration platform that will categorize station based on people input on incidents such as number of theft, people fallen on track, people got stuck in between train and platform, or even broken machine. This will make people self-aware and will tell them to ride safe.
  7. Now, I will prioritize the solutions across the user goal and complexity of development. The user’s goals is to reach from point A to point B efficiently with most up to date information.
Solution Impact to User Goal Complexity
Personalized real-time notification High: prompt and timely notification about delay or change in service will allow commuter to look for backup plan. They may opt to work remotely on that day. Also providing targeted notification is more valuable to user Medium: Google maps already capture information like train running late, where the congestion is.
Book Lyft/Uber for commuter Low: not all commuter have same affordability level, they may find Lyft/Uber expensive option.  Also not everyone has Uber/ Lyft account.  This may serve just a subset of overall commuter population Low: Google Maps already shows ride sharing option to user when they search for route.
Information catalog High: having up to date information about the station will help commuter plan their trip Medium: easy to develop initially but require ongoing maintenance to make sure that information is up to date and accurate.
Train/Bus station rating High: getting review from other commuter increases trust. High: will require developing collaborate feature in Google Maps that will allow other commuters to update and provide information such as how many ticket vending machine are working, which machine is not accepting cash, etc.
  1. From the table, the two features that provide the highest impact for the lowest effort are (a) personalized real-time notification and (b) Information catalog. My recommendation would to start with these two solutions while also start brainstorming on building collaborative feature with in Google Maps as this will definitely change the way people see and use Google Maps today.