Google Associate Product Manager Interview Guide

Allen Nikka

9/16/2020

Google Associate Product Manager Interview Guide

Why did I write this?

Tell me about yourself / resume walk

Why Product:

Why Google?

Your questions for your interviewer

Future Plans / Selling yourself:

General Tips for answering Technical Questions:

General Tips for answering Non-Technical Questions:

Must Know Google Topics:

How to answer Product Questions:

Products You Love/Hate:

Product Evaluations

Google Culture and Values:

Estimation Questions:

Strategy

Tips:

Useful Numbers Cheat Sheet:

Pricing and Profitability:

Pricing Models:

Advertising:

Strategy for Strategy Questions (pun intended):

Strategy for Marketing Questions:

Strategy for Launching Questions:

Strategy for Brainstorming Questions:

Example Question Answers

Product Design:

Product Strategy:

Job Description:

Google Recent News / Overall Metrics:

Google Products:

Google Interviewer on Common Pitfalls:

Sources

Why did I write this?

A lot of people have been asking me about Google’s APM program, or APM interviewing in general, so I thought I’d consolidate a version of a document I put together to help me interview, to help others better prepare for their interviews, too.

This will likely evolve over time, but to put things in perspective, it started as a way for me to collect my own thoughts, questions I thought I might be asked, and approaches to answering those questions before my own APM interviews at Google. So while reading through this document might be useful, I think the most useful way to interact with this document is to step through it and make sure you’ve thought through your own answers to these questions.

Note: Many of my examples refer to Google specifically, but you can substitute in the name of any company you’re interviewing with, they should still apply. Also this was originally made in Fall 2019, so some pieces of specific information related to products, news about Google, or otherwise, might be out of date, though this overall approach should still be useful.

Tell me about yourself / resume walk

• Tell the interviewer what the unifying narrative thread of your resume walk is so they can follow it all the way through
• For example: “My experience seems varied, but in retrospect, the uniting characteristic is that I love working with dynamic teams and cutting edge tech to solve important problems by building things people love to use (and don’t even know they need)”
• Point to any key accomplishments you can quickly summarize
• Metrics are great
• Make sure to emphasize why you’re interested in product, try to weave it into your answer
• Don't forget, you’re telling a story, make it interesting!

Why Product:

• You should really think about why you’re interested in this role, because it’ll show
• For me this included:

• Really enjoy flexing all the muscles you need to do well in product:

• Deep technical expertise
• Strong sense of customer empathy
• Creative thinking
• Business/market oriented thinking

• Realized that a role allowing me to combine my:

• technical domain knowledge
• highly developed EQ/soft skills
• Past experience
• In a smart, driven team where I can own product is the perfect fit for me to have maximum impact and maximum growth, win-win

Why Google?

This should align with Google’s missions and values, and hopefully, yours too! This can include:

• Google’s core mission
• Google’s company values
• What you know about the company from friends/acquaintances that work there

Your questions for your interviewer

You really should have some. An interview is a two way street: you’re trying to figure out if a company (Google or otherwise) is a good fit for you, and visa versa. Not having any questions means you’re not really thinking about that first, really important part of any interview process. I’ve included some things I wanted to know the answers to below:

• What do you value most in the Google’s culture?
• How much product/feature ownership do APMs take on?
• How does mentorship work in the program?

• Formal and informal?
• What does the ramp-up look like?
• “Thrown in the deep end”: What tools/lifelines do you have?

• Are APMs limited to software products?
• What is the usual trajectory of APMs after finishing their two rotations?
• How does A/PM role vary across different product lines (consumer, mobile, apps, enterprise, infrastructure)?

• Or more narrowly, on the different teams you have worked with?

• What does a typical day look like for you?

• Writing specs vs working with engineers vs working with designers vs other?

• Most challenging thing about your A/PM role?
• What product(s) are you working on now / have you worked on?

• What makes an ideal Google A/PM in your opinion?
• What does kicking ass as an APM look like?

Future Plans / Selling yourself:

Think of the qualities in yourself that you think make you a great fit for the role, and be able to sell yourself. This doesn’t mean acting like your selling a used car, but advocate for these qualities, and be able to back up your assertions with times you exhibited them. Some qualities I listed for myself:

• Responsible - found funding opportunities and ways to cut costs to keep my startup alive
• Resourceful: quickly self-taught a company’s proprietary C++ memory management and multithreading libraries, as well as API interfaces, to deploy a feature using a 30M line, 25 year old codebase
• A quick learner - able to go deep fast, not just on technical topics: learning biz dev, customer interviews, i-Corps
• Able to translate between disparate skill-sets and jargon: bridged engineers to designers and PM in previous roles
• A bridge-builder

General Tips for answering Technical Questions:

1. Ask clarifying questions to understand goal
2. Provide structure

1. Help the interviewer take notes
2. Explore options in structure

3. Clearly state assumptions

1. Show decision making is informed, data driven

4. Use the whiteboard & think out loud
5. Sanity/gut check answers
6. Go back and work through other solutions (from 2.b) or try to improve existing ones

General Tips for answering Non-Technical Questions:

1. Ask clarifying questions to understand goal
2. Examine at micro (product) and macro (company, environment) level
3. Understand customers
4. Assess qualities of company:

1. goals, mission, positioning, strengths, and weaknesses
2. Align with mission if possible: organizing the world’s information and making it accessible to people

5. Segment competition
6. Marketing:

1. Market tailored to above

7. Launching:

1. Determine target market, users, MVP, rollout, distribution channels

8. Strategy

1. Identify high level bucket:
2. Diversification, barriers to entry, one stop shop, low cost, reducing key buyer/supplier reliance, testing a market

Must Know Google Topics:

• Google’s market, competition, and trends
• Ads: AdWords, AdSense, CPC, CPM
• Industry Trends (e.g. cloud, sharing economy, mobile, VR, ML)
• 2 google products in depth

• E.g. Translate, keep, others

• 2 startups in depth (especially those that could be good acquisition targets for Google)

How to answer Product Questions:

Product Design:

1. Ask questions to understand the problem
2. Provide a structure
3. Identify users and customers
4. What are the use cases? Why are they using the product?
5. How well is the current solution doing for their use cases? What weak spots exist?
6. What features would improve the weak spots?

Improving a product:

1. What is the goal of the product?
2. What problems does the product face?

1. Expanding user base, revenue, engagement, conversion?

3. How would you solve this problem? → big/risky (Google prefered) vs incremental
4. How would you implement these solutions?
5. How would you validate your solution?

Favorite Product:

1. What problems does it solve for the user?
2. How does it accomplish these goals?
3. What makes it neat / makes people fall in love with it?
4. How does it compare to alternatives?
5. How would you improve it?

Products You Love/Hate:

For each, you should think about: goals, main use cases, key pain points, users/personas, current solutions, what can be improved, competitors, and metrics to measure success

Come up with your own criteria for what makes a “good product”, and let your interviewer know what it is before you start evaluating products to keep yourself consistent. I had three principles:

1. Fits entire user story better than counterparts
2. Intuitive to use (good design, UI & UX, simpler is better)
3. Address an underlying human need, not just a specific task

Product Evaluations

I kept a few just to demonstrate the level of detail and the way I thought about these, but you should definitely think through these yourself - pick products that are impactful to you! Having a good blend of hardware, software, and non-tech products is also good practice.

Blind:

• Goals: ask the questions you always wanted to ask, and get semi-verified results, from an anonymous professional community
• Main use cases: figuring out if salary is competitive, what company cultures are really like, honest feedback on big decisions
• Key pain points: how do I get honest feedback on topics that are generally taboo about work, personal life, etc. with  some soft notion of verified / legitimate responses?
• Users/personas: industry veteran offering advice, industry newbie seeking advice, individual at a big decision point, lonely individuals, individuals feeling disconnected
• Current solutions: reddit, anonymous forums, confidential in-person conversations, anonymous notes
• Possible improvements:

• blind groups: make a cohort of reliable peers to bounce ideas off of
• anonymized sentiment analysis around keywords
• Influencing feed with searches
• Auto-tagging users with expertise given what kinds of comments they make/posts they comment on, and prompting them to reply to relevant posts
• Web client functionality matching mobile (direct chats, etc.)

• Metrics: installs, uninstalls, posts engaged with / user / day (read, comment, like), posts commented on after prompts, groups created, conversation quantity in groups vs otherwise, velocity of all metrics over time, divided by cohorts (i.e. personas from above), company, etc.

Pulse SMS:

• Goals: provide User with unified, feature rich messaging experience
• Main Use Cases: see personas
• Key Pain Points: I want to text but from more than just my phone, and I’m not on Mac, I want my texts stored encrypted
• Users/Personas: Apple convert looking to message from all devices, tech enthusiast desiring expanded functionality, user concerned with privacy and stored message encryption
• Current solutions: MightyText, Android Messages, etc. email, fb messenger, messaging apps
• Possible Improvements: auto refresh/sync, modified sales model (free for sending, pay for encryption and warehousing of messages) - how often has your “messages” memory started to get huge?
• Competitors: see current solutions
• Metrics to measure success: monthly active users, downloads, free to paid conversions, uninstalls, messages sent, number of devices messages are sent from

Kindle e-reader:

• Goals: Carry your library with you, get books instantly, read anywhere, easy on the eyes
• Main Use Cases: scholastic, leisure, business
• Key Pain Points: weight of books, ease of acquiring books, maintaining central repo of documents
• Users/Personas: student whose back is breaking, leasure reader who wants books on demand, businessperson who needs documents and journals to read on demand
• Current solutions: lug a backpack, go find a bookstore, carry all your documents or journals with you, paper books, reading books on computer/tablet
• Possible Improvements: easier highlighting over page boundaries, subscribe to reading lists with teasers of books, better social integrations (think spotify)
• Competitors: Kobo e-readers, B&N nook, reading books on computer, paper books
• Metrics to measure success: book downloads, boundary highlight attempts / completions (closer to 1 is better), active book/document viewing time, book purchasing cadence before/after kindle purchase, quantification of social interaction use

Moleskine Notebooks:

• Goals: record or revisit thoughts, seamlessly, beautifully, and durably
• Main use cases: academics (notes), business, design, professional writing, journaling
• Key pain points: ease of location (of the notebook, and individual notes), aesthetics, carrying/storing
• users/personas: students, writers, business professionals
• Current solutions: loose leaf paper, spiral notebook, electronic note-taking device
• Improvements: reversible spine so writing of left pages doesn’t cause right hand to run into right page (opposite for left-handed people)
• Competitors: loose leaf, electronic notes, smartphones, spiral notebook
• Metrics: units sold of each variety (size, spine type, color sku’s), correlated to date/time of year and geography

Daniel Wellington Analog Wrist Watch:

• Goals: know the time and date, demonstrate fashion sense, display wealth/status
• Main use cases: jewelry, wealth display, telling the time
• Key pain points: limited range of culturally acceptable jewelry options for men (US), needing to know the time (discreetly, i.e. without pulling your phone out)
• users/personas: first time watch buyers, buyers looking to expand collection inexpensively, medium income fashionista/os
• Current solutions: sundial, asking someone for the time, cellphone, computer, smart watch, bracelet
• Improvements: ‘smarts’ built into the back of analog watch face to allow haptic-based interaction, swappable watch face housings to compliment swappable bands
• Competitors: smart watches, cellphones, bracelets, rings
• Metrics: units sold by geography, demographics of buyers if online, timing of sales, number of units synced to phones, number of units sending haptic notifications

Products the author hates:

• Push to start cars (but not keyless entry, just keyless start): create problems of losing keys and reduced safety of thief starting car, don’t solve any real problem/pain point

Google Culture and Values:

Things that are good to know before your interview

• Thinking is biased towards large impact - billions of users
• Looking for the 10x leaps, not the incremental improvements
• Teams across the company with deep expertise that you can worth with, and simultaneously learn from
• People are willing to, and enjoy, sharing knowledge
• It is a true meritocracy, varied backgrounds are embraced and leveraged
• Core tennant is that benefits of technology should apply to everyone
• “Create/Design/Build/Code” for everyone

Estimation Questions:

Strategy

1. Clarify the question (Revenue vs profit? US vs worldwide?)
2. Catalog what I know / wish I knew
3. Make an equation
4. Consider edge cases & alternative formulations
5. Break down each component of equation so it can be estimated independently

1. Solve each independently and then merge, don’t combine them all

6. Review and clearly state assumptions
7. Do the math
8. Do sanity checks

Tips:

1. Use round numbers
2. To calculate time until something doubles, divide by 72 (70 or 75 in practice)

1. Growing something x% per year, doubles in 72/x years

3. Given a * b = n, #digits(a) + #digits(b) = #digits(n) +- 1
4. Be confident and label ALL units
5. Consider all sources (of sales, etc.) - can call out and purposefully ignore some
6. Record ALL intermediate steps
7. Record ALL assumptions (running list, circle them as I make them, etc.)

Useful Numbers Cheat Sheet:

Pricing and Profitability:

• Cost plus pricing → hard for online
• Value pricing → if service has direct value to user
• Competitive pricing → price based on competitors prices
• Experimental pricing → experiment live with different prices and pick based on sales volume, can make people angry!

Pricing Models:

• Free, add supported → hard unless you have special ad angle
• Freemium → must keep a close eye on conversion
• Tiered → different features at different price points
• A la carte → price each feature/service separately
• Subscriptions
• Free trial
• Razor / blade model

Advertising:

• Key ratios: click through rate (CTR), cost per click (CPC, usually via auction), conversion rate
• Pricing:
• Pay per click → advertiser pays only when someone views their ad
• Pay per impression → pay each time an ad is shown
• Pay per action → pay each time user goes from ad to specific action (like purchase) rarely used, hard to track, FB does it with downloads and page likes

Max $ paid for CPC ad = conversion rate (2-5% for FB and G) * profit per conversion

Conversion Rate: How many website or app visitors complete an action out of the total number of visitors.

Strategy for Strategy Questions (pun intended):

• Two major facets:

• Micro → product
• Macro → company analysis (goals, positioning, etc.)
• Strategy should be on both levels

• Strategies:

• Diversity revenue sources
• Building barriers to entry
• Being the “one stop shop for ___”
• Being a low cost leader
• Reducing reliance on key buyer or supplier
• Testing a new market

Strategy for Marketing Questions:

1. Understand the company:

1. goals, mission, strengths and weaknesses

2. Understand the competition:

1. segment them

3. Understand the customers:

1. Who are they? What are their needs? Where & how do they live/work/play?

4. Understand the landscape:

1. Legal issues? Regional forces?

5. Market the product based on a combination of the above

Strategy for Launching Questions:

1. Product:

1. Discuss vision, strengths, weaknesses, risks

2. Launch goals:

1. Determine what we want now and what can come later
2. I.e. users, profitability, etc.

3. Launch design:

1. How will we achieve goals from (2)?

4. Launch implementation: (Pre, During, & Post)

1. Target market
2. User types
3. MVP or full product (i.e. polished MVP)?
4. Distribution channels / plan
5. Rollout
6. Buzz
7. Partnerships
8. Risks

Strategy for Brainstorming Questions:

1. Suspend disbelief, throw out some crazy ideas
2. Think about strengths and key assets of the product and/or company
3. One vs many:

1. What can you do with one item versus with many items

4. As-is vs. with modifications:

1. What can we do with an item unmodified and then post modifications

Example Question Answers

Again, these are examples I came up with, the best exercise would be to come up with your own, and then compare/contrast with some of the things I came up with here.

Product Design:

• Improve restaurant search:

• OCR and NLP to parse menu contents for common food keywords
• Allow filtering on semantically similar groups of keywords like “cheese” or “fish” to see if menus do/do not have those items
• Streamline process for picky eaters, eaters with dietary restrictions, and their friends/families

• Next great search feature:

• Verified results: allow users with verifiable credentials to endorse results
• Medical, news, business, etc.
• Credential linked against google account, prompt user to link based on the types of searches they are doing (i.e. lots of very specific medical searches - you are likely a medical professional)
• Prompt feedback after credentialed user visits a topic in their domain, or passively detect which sources they use more regularly and/or look at for longer periods of time
• OR:
• Video search → like shazam but for video with all video on the internet, any video that contains the portion of video you are searching for should pop up in the search

• Frame based convolution and pooling at fixed time intervals in video
• Operate on encoded media (H.264, etc.) to speed up processing time

• OR:
• Product sentiment analysis:

• Keep index of sentiment analyzer in context of brands’ products
• Scrape entire internet, blogs, social media
• Allow brands to measure how much people love (or don’t love) their product

• Potential pitfall:

• Privacy concerns
• Would have to be very clear about how this data is anonymized
• Might not be able to do so on social networks with password lockout, etc.

• Monetize a product more effectively:

• Pulse: free for minimal off device storage and multi device texting
• Monthly fee for larger text storage, encryption, and backup of messages

• Rather than 1 time payment that exists now

Product Strategy:

• Should google’s CEO be concerned with microsoft?

• GCP vs Azure: yes as their market share is verging on even. Rely on deep integration to the rest of google’s developer ecosystem, android, and supremacy in mobile to take the lead
• chromeOS vs Win10: yes, because of incumbent inertia and penetration of Win10 into desktop market, Google has an uphill battle but can win for a niche set of non-power users

• Should google create its own version of stubhub?

• Initial thought: no, as it is outside google’s core business, relationship based market, having better tech doesn’t make you better than an incumbent with established relationships  
• Revised: a marketplace product would be a good idea and make sense, as overhead is likely low and could reuse many of google’s core competencies (i.e. reskinning Google flights, but allow sellers rather than airlines to list tickets). Large market that is growing due to the “spotify” effect on the music industry and similar effects of digital distribution on entertainment industry writ large. This business unit would also be readily internationally scalable (see: google scale).

Job Description:

This was the job description when I applied for Google APM, but double check on the career site for the latest and greatest.

• Key points:

• Adaptive, excited to work on projects affecting millions of people a day
• Cross functional, connect business and tech worlds
• Take projects from conception to launch
• Tailor solutions to users
• Emphasis on impact

• Qualifications:

• BS in CS
• SW dev experience: C, C++, Java, JavaScript, Python
• Intern, TA, Lab work experience
• Experience with entrepreneurial efforts, outreach within organizations, or PM experience

• Preferred Qualifications:

• Demonstrated ability to learn new concepts, interest in creating and analyzing products
• Effective analytical, problem solving, organizational, verbal, written, presentation, communication and interpersonal skills

Google Recent News / Overall Metrics:

These were recent in Q4 2019, but you should find out what’s going on now.

• Mobile search, youtube ad revenue, GCP are leading growth

• GCP: high demand from customers for compute and data analytics

• HW business (Pixel, Nest) is slow due to strong competition

• Sundar is bullish on google home → describes it as “market leader”

• Asia-pacific is the fastest growing market due to google play and YouTube Music (esp. In India)

Google Products:

Yes I really did this.

Google Interviewer on Common Pitfalls:

1. Lack of depth in product insights: Most candidates can't dissect a product and provide any thoughts on what makes a product good or bad, what makes a design good or bad etc. When an interviewer asks you to pick your favorite Google product and improve it, don't just name the last random bug you encountered in a product and tell them to fix it, but think big picture, think about the market the product operates and then figure out how to move the products metrics by 10x.
2. No understanding of the market or competitors. This one is a stretch for undergrads since they have never been in the industry before, but try to do some competitive assessment of the products you're discussing and talk about what the threats and opportunities are. E.g if you said Spotify is your favorite product, don't say that it's great because it streams music (RealPlayer did that in the 90s), but talk about what makes it better than YouTube Music or Pandora and why Apple Music is a threat to it.
3. Lack of structured thinking: You're a CS grad, you have taken engineering courses, so use those analytical skills to break a problem down and analyze what the root cause is and how you would solve it. Don't just blurt out the first solution that comes to your mind, but take a few minutes to think through the underlying problems - whether they are technical, economics, demographics or whatever and suggest solutions. Not a single solution, but multiple solutions and then provide a framework to rank them and pick one.

Sources

I didn’t write this in a vacuum - in fact, lots of it is heavily informed / influenced by Cracking the PM Interview: How to Land a Product Manager Job in Technology by Gayle Laakmann McDowell. Otherwise, most if it was randomly amalgamated from the web too long ago for me to track exactly where everything came from.