Leave a comment
Get the GH Bookmarklet

AMAs

"Going Beyond the Install | Growing your app for value instead of volume"

Fontaine is a Product Manager focusing on mobile app analytics at Google. She started working on Google Analytics but recently launched Firebase Analytics at Google I/O this year. Under this umbrella, she has worked on features to expand install attribution, publisher reporting, and in-app activity.  

Prior to Google, she worked as a Product Manager at ModCloth, an eCommerce company. There she was a daily practitioner of analytics and growth tools while she built a mobile website as well as iOS and Android apps.  She has a BS in Civil Engineering from Stanford University.

You can find her on Twitter at @foxfont

She will be live on August 11 from 930 AM PT for one and a half hours during which she will answer as many questions as possible.

  • SA

    Shaker A

    over 2 years ago #

    Hi Fontaine,

    It's great to have you here! I love ecommerce and admire what ModCloth has done, so I'm very excited to learn from you!

    1) Can you talk about how ModCloth looks at retaining users when your user only need to use ModCloth occasionally by it's very nature (ie when they want to buy something)? If they aren't using ModCloth frequently, building up the habit is hard, which makes it even harder to retain users.

    How do you go about trying to stay top of mind, so when the user has a need that ModCloth solves they think of you? What are experiments/initiatives that have worked for ModCloth when it comes to increasing retention?

    2)I noticed that ModCloth has tiered shipping options, ranging from free to express. Did you find retention to be greater when the free shipping threshold was lower or when the threshold was higher? Does free shipping impact the the quantity and frequency users purhcase at? If so, how? What did you learn about free shipping, and shipping in general, and it's impact on customers during your time at ModCloth?

    3)Buying something is a very transactional thing, especially when it's done online. How did modcloth go make the experience about more than just buying something?

    Looking forward to learning from you!

    • FF

      Fontaine Foxworth

      over 2 years ago #

      Hello! I am not currently at ModCloth but happy to answer your question more broadly regarding user experiences.

      For your first question-- Retention objectives should be tied to the overall brand-- are you a brand that is launching products daily, and therefore can provide a reason for your users to come back to see “what’s new”? Or do you have fresh content that could be something that a customer looks forward to consuming when waiting for the bus, for example? If your brand has products or content on a more seasonal basis (as opposed to daily or weekly), then it’s unreasonable to be looking at retention on a daily basis.

      Regarding shipping-- it’s important to test different tiers for shipping. What works for one company doesn’t mean it works for other companies. Free shipping certainly impacts the quantity and frequency users purchase at, but there isn’t a silver bullet answer. What is the price point of your products? How big is the range of prices you offer? If most of your products are $40, then a $50 threshold seems silly. There’s no easy way to reach that threshold. Perhaps a $75 threshold makes more sense. Generally speaking, shipping thresholds are a very important area to test in the context of your products and users to see what makes the most sense for you.

      Buying something does NOT have to be a very transactional thing. You have so many opportunities to make the experience uniquely you-- what is the voice to the customer? We used to use lively confirmation and error messages to spice it up. What story does the photography tell about the products? Do you name your products something unique? (ModCloth is great at that). How did you select what to sell? Did the customer play a part in defining the brand? How do you use push notifications to re-engage your users in creative ways? In many ways I think you have more control over the experience when you’re online-- more variables and tools to play with. Assuming the experience needs to be transactional is doing a disservice to you and your customer :).

  • AL

    Arsene Lavaux

    over 2 years ago #

    Bonjour Fontaine,

    Looking forward to your AMA and new mobile growth learning.

    1) Do you see gaps in the mobile analytics stack?

    2) How do you envision the future of mobile analytics to fill those gaps?

    3) What is the most innovative mobile analytics tool in your opinion? Why?

    Merci!

    • FF

      Fontaine Foxworth

      over 2 years ago #

      Hi Arsene,

      There are lots of gaps in the mobile analytics stack. Whenever I ask app developers what analytics tools they use, they end up rattling off 4 or 5 different tools. Tools for behavioral analytics, attribution, crash reporting, push notifications, etc. I think there’s an opportunity to consolidate the analytics tools into a single offering. Naturally, my recommendation today is using Firebase Analytics (launched at Google I/O) since it’s objective is to provide a single analytics tool that can be used by different folks at the company to do the things they need. Whether it’s crash reporting, tracking your advertising campaigns, understanding user retention, seeing how effective push notifications were, these can all be accomplished in a single tool. This means that developers, product manager, VPs, marketers-- all of these different roles should feel comfortable logging into Firebase to get their tasks done.

      • AL

        Arsene Lavaux

        over 2 years ago #

        Merci Fontaine,

        I'll look further into Firebase since I agree that mobile tools are scattered all over the place for now and we need more growth convenience...

        Who are your competitors?
        Are you the only fully integrated solution out there?

      • FF

        Fontaine Foxworth

        over 2 years ago #

        Firebase is trying to tackle a lot of functionality, so I don’t know if I can name a single player who is offering the exact swath of features. However, check out my reply to Logan for some ideas of other Growth companies.

  • AA

    Aldin A

    over 2 years ago #

    Hi Fontaine!

    Thanks for doing this AMA!

    1)For ModCloth, how did you figure out what to use as benchmarks for things like retention, repeat purchase rate, avg cart size, etc. Where and how did you get the bench marking information to make sure that your core metrics were healthy compared to the industry and your competitors?

    2)Other than sizing charts, what has modcloth done to help buyers identify clothes that will fit them? What has ModCloth done to make up for the fact that people can't try clothes on before they buy them? How has ModCloth helped buyers overcome the fear that the clothes they buy online may not appeal to them when they finally get it, or the size may not fit, or both?

    3)Other then sizing what other friction points did modcloth find that stopped users from buying online? What has Modcloth done to overcome these friction points?

    Thanks

    • FF

      Fontaine Foxworth

      over 2 years ago #

      Hi Aldin, although I can’t speak on ModCloth’s current strategies, I can certainly address how you can use internal & external data to set benchmarks.

      1) Benchmarking can be tricky because each company is different-- different price points, product mixes, and content strategies lead to different retention, purchase rates, cart sizes, etc. I think it can be problematic if you’re just focusing on comparing yourself to competitors since it’s often not apples-to-apples. Instead, I’d recommend measuring against your own data. Are you able to increase your conversion rates YoY? What about using growth tools to boost retention rates each month? When you are comparing your metrics to your historical data, you know that you’re focusing on making the experience better for YOUR customers. Of course, there are benchmarking tools in a variety of growth products (App Annie and Google Play are examples), but you can use those to focus on benchmarking for other metrics like app downloads.

      2) ModCloth has done a lot to help buyers identify clothes that fit them. User generated content is probably one of the strongest elements. Users can write reviews and upload pictures for all of the products they purchase. My favorite feature that ModCloth has (and one that I worked on!) was a feature available in the iOS App called “Fit for Me.” A user can input their bust, waist, and hip measurements and see a list of products recommended for them based on reviews for users like them! They can see what’s highly rated, see the images that accompany the reviews, and the exact details of why a product did or didn’t work for a customer. This makes it so much more customized than “apple-shaped” or “runs small” descriptors; it puts the power of fit into the hands of the community.

      3) There’s a host of things that can be difficult to overcome when launching an online retailer; among them are product quality, shipping costs, high return rates, marketing/branding, technical issues. Each one of them is something that should be considered in the context of your business to make sure you’re designing the right solution.

  • RB

    Ry B

    over 2 years ago #

    Hey Fontaine,

    Excited to have you here at GH!

    1)How did modcloth differentiate themselves from the other ecommerce sites that are in the same space? What were there differentiating qualities, and how did ModCloth express these qualities to prospective and current customers?

    2)As the number of people buying on the site grows, and the number of things ModCloth sells on the site increases, it gets harder and harder to help people find things they will like. One way to help people find things they will like is to use recommendation engines (which ModCloth seems to use).

    What insights have you gained about recommendation engines from your time at ModCloth? What else has ModCloth done to help ensure that customers will find things they will like?

    3)What does ModCloth's content strategy look like? I imagine just blasting people about promotions day after day wouldn't work for very long. What topics does ModCloth write about and how do they decide that these are the topics that there followers want to hear about?

    Thanks for doing this!

    • FF

      Fontaine Foxworth

      over 2 years ago #

      Hi Ry, although I can’t speak on ModCloth’s strategies, I’d be happy to speak to how companies can differentiate in very crowded industries.

      1- Arguably the most important part of differentiating is figuring out your brand. ModCloth has a very distinct brand and connection with their community. A lot of the brand initially came from the “Be the Buyer” initiative which was a program that allowed the community to be part of the fashion buying process. It was a very distinctive take on fashion, but resonated with a lot of customers. Finding your voice as a brand is a critical part of establishing yourself in the market.

      2- I disagree with the statement that “As the number of people buying on the site grows, and the number of things ModCloth sells on the site increases, it gets harder and harder to help people find things they will like.” In fact, I find the opposite to be true-- the more purchases you have and the more products, the more opportunity to make smart recommendations to your customers. Recommendations are key to building a good online business; it can be overwhelming to delve through hundreds of products. Instead, take the data at your fingertips and do something cool with it. Coming up with recommendation algorithms was a really fun part of working in ecommerce. You could approach it in so many ways. Maybe recommend the most popular items. Maybe use colors, shapes, and lengths a user gravitates towards. Maybe find items you think will fit them based on other users reviews. There are so many directions you could take; it’s a really fun challenge.

      3- ModCloth is launching products daily, so in many ways some customers treated ModCloth like they might a lifestyle inspiration blog-- it’s fun to see the latest styles whenever you have a free minute. If content is simply promotions, I don’t think you’ll go far. But if you have other types of content-- products, blog posts, inspirational imagery-- your users will want to come back simply to explore and not necessarily to just shop.

      3 Share
  • MM

    martín medina

    over 2 years ago #

    Fontaine,

    Thanks for doing this AMA!

    As a Product Manager what do you believe your primary role and responsibilities are regarding the growth of a new product? How long do you typically work on a product?

    What is the most challenging thing about working at Google?

    What are some of the most exciting things you have seen regarding mobile analytics?

    • FF

      Fontaine Foxworth

      over 2 years ago #

      I like to think my job is taking a bunch of inputs and turning it into a single set of requirements to build. Inputs can come from so many areas-- customers, other parts of Google, competition, executives, myself-- so understanding needs and turning that into a solution is a key part.

      I think it’s important as a Product Manager to be committed to a certain area of the business; so you might work on specific features for several months, you are always driving an area of the roadmap. Particularly for Product Managers, I think it’s important to commit several years working on a product if you really want to make a difference; you’re responsible for driving the vision from start to finish, so it’s key to be ready to buckle down for a while :)

      Working at Google is incredibly exciting and empowering. One of the biggest challenges is making sure all of my stakeholders are happy(ish). There’s never a silver-bullet solution to any problem, so finding a solution that works well for most stakeholders is always the fun part.

      Mobile Analytics is a totally different space than web analytics, so in many ways I feel like it’s unfolding before our eyes. It’s fun to be part of something that I feel like we’re shaping as we go. There are no obvious powerhouses in the space, so it makes for lots of creative solutions and companies!

      4 Share
      • FF

        Fontaine Foxworth

        over 2 years ago #

        Transparency is #1. If I can show my stakeholders the other constituents’ needs, it instills trust and understanding in our decision-making process.

      • AA

        Anuj Adhiya

        over 2 years ago #

        With respect to keeping everyone happy(ish) - are there are any strategies/solution you've seen work better than others?

  • ES

    Edward Stephens

    over 2 years ago #

    Hi Fontaine,

    Can't wait to have you on for this AMA. Couple of questions this end:

    1) What challenges are Google facing at the moment to get users to conduct more search on mobile and do you think it makes sense to do targeted acquisitions like FB of key mobile products (e.g. the Snapchat's or Instagrams)?

    2) When you set up the analytics suite for ModCloth what tools did you use, and how did these change at scale and which of them would you now revise from that list in hindsight?

    3) How can mobile app developers encourage users to make updates? I've noticed this is a huge problem with some of the start-ups we work with. They drive a load of installs, then make a critical update with new features but new app users either aren't aware or don't bother installing the update before disengaging.

    Looking forward to hearing your answers!

    • FF

      Fontaine Foxworth

      over 2 years ago #

      1. Google has a variety of formats on mobile that aren’t just confined to Search. There is the mobile display network both on apps and the web, Google Play, Youtube, etc. In many ways we’d like to remove the cognitive load of creative ads from marketers; they should be able to easily come to Google and create an app promotion campaign, designate their goal, and start running. Simply acquiring startups doesn’t equal more advertising opportunities; I think the focus should be building easy experiences with the host of products and formats we already have.

      2. Before coming to Google I was an avid user of Google Analytics on both apps and web. So, when I joined the Google team, I had a lot of opinions about how to make it better. ModCloth was a very data-driven company, so we had robust tracking throughout the properties. This helped our teams make the most informed decisions. :)

      3. I’ve seen a variety of approaches when it comes to encouraging users to make updates. An extreme example is to force users to update their apps by limiting functionality if they haven’t updated to a certain version. If you are launching a critical feature, then this is a route to consider. Sending push notifications with highlights of what is being introduced can be a fun way to get users to update. At the end of the day, app updates are a part of application development, so it’s something that we have to be comfortable with as app developers. We will never have 100% of our app users on the latest version of the app; make sure that when you’re doing your analysis of users you are considering the most recent versions of your app.

  • DH

    Dani Hart

    over 2 years ago #

    Hi Fontaine,

    Excited to hear your answers today.

    I'm curious... Is Firebase Analytics a competitor to other mobile analytics tools already out there or is it complementary (just like GA is with others)? If it is complementary, what tools do you think best complement what Firebase Analytics offers for app publishers to get the most complete picture of their usage through the AARRR funnel?

    Thanks!

    • FF

      Fontaine Foxworth

      over 2 years ago #

      Hey Dani,

      I’d turn the question back around-- what do you mean when you say “mobile analytics”? “Mobile Analytics” means so many different things to different people-- crash reporting, push notifications, attribution, behavioral, etc. Firebase can certainly be used as a complement to other tools, but if developers aren’t using other analytics tools yet, then I think it could easily take on many of the types of analysis that people have historically used 4 or 5 products to do. With respect to other tools that could be used simultaneously, check out my response to Logan on another thread.

  • JF

    Javier Feldman

    over 2 years ago #

    Hi Fontaine, It's awesome to have you here! Here's my question:

    How are growth teams structured at Google (or if its easier to answer - in the part of Google that you work within). Why is it that way?

    Looking forward to read your answers.

    Cheers!

    • FF

      Fontaine Foxworth

      over 2 years ago #

      Hey Javier,

      Great question. I would consider my area to be unique because I’m actually working on building Growth tools myself (via Firebase). So, when we talk about “Growth teams,” it’s more about delivering features. As the data layer of Firebase, it’s my product’s job to connect and supercharge a lot of Growth features.

      I think historically teams have been very siloed when it comes to growth-- advertising functioning independently of other organic channels like push notifications or app invites. I think the industry is starting to have a more holistic approach to growth, so there’s more fluidity between areas of the organization. A common chasm I see is between developers and marketers-- they do very different things, but at the end of the day they’re both looking to drive more users. By introducing products that both functions could easily use (like Firebase), we hope to start establishing more communication between different teams.

      2 Share
  • DO

    Danielle Olivas

    over 2 years ago #

    Hi Fontaine,

    Thanks for joining us today!

    How has being a product manager at ModCloth been different than at Google? What lessons did you take with you to Google, and on the flip side, what did you learn at Google that you wished you knew earlier?

    • FF

      Fontaine Foxworth

      over 2 years ago #

      In both places, it’s my role to take a bunch of inputs from all over and turn them into features, but of course I go about it in a very different way.

      First of all, my “users” are very different-- at ModCloth they were largely women consumers in their 20s and 30s. I could easily whip out my phone at the dinner table and ask my friends for feedback on a new feature. At Google, I work mostly with app developers. Building for businesses rather than consumers means I engage with them in a very different way. I meet with them at conferences, in their offices, in coffee shops.

      Secondly, the nature of my products are very different. The apps I worked on at ModCloth were very compact. At Google, I’m building a product that needs to integrate with a bunch of other areas of Google. Integrations are a much bigger part of my work at Google, since at the end of the day, data is most powerful when it can be acted upon.

      ModCloth instilled an obsession with user feedback in me; making sure that everything we did was tested with users was an important part of the flow. It’s been fun to exercise that at Google as well. At Google, I’ve gotten to see how one product area can impact the greater business as a whole. It’s cool to understand how Analytics fits into the bigger developer and advertising ecosystem.

      2 Share
  • GH

    Glen Harper

    over 2 years ago #

    Hi Fontaine, thanks for being here today. May I ask, what are the biggest mistakes you've seen/personally experienced when it comes to implementing an international expansion strategy? I've been involved in a few and would love your perspective. Cheers!

    • FF

      Fontaine Foxworth

      over 2 years ago #

      Hey Glen,

      I might highlight 2 things-- first, prioritization. Find out what markets to expand to and force rank them. Where do you want to internationalize your descriptions? What currencies should you be supporting? Don’t try to do all markets at once.

      Secondly, spend time to get to know the markets. Take trips to visit them, talk with users there, and see how their needs might be different. I just did a tour around Asia, and the feedback and needs I heard from app developers were totally different from those I’ve heard in the states. Now I understand how to build features to better address those markets.

      2 Share
  • AS

    Alex Sherstinsky

    over 2 years ago #

    Hi Fontaine, and thank you so much for taking the time to do an AMA with us! We love analytics around here! Question: Is the app store still the main way to discover an app today? Given the big inefficiency there, do you see this as changing any time soon? If so, how do you think the change will come about?

    • FF

      Fontaine Foxworth

      over 2 years ago #

      App stores are certainly one way, but it’s not just about the charts. Search certainly drives app discovery. Our partner TUNE determined that 67% of app store installs were preceded by a search query. So, we are seeing the power of the search box repeated on mobile phones. As the market matures, we are starting to see a broadening of the channels that matter for mobile app marketing. Phil Schiller at Apple said that 65% of app installs were the result of an app store search query. It’s this appreciation for the power of search-- that motivated us to offer paid search placement in the Play Store. We also have Universal App Campaigns at Google which allow you to run a single campaign that distributes your app across a variety of inventory; this way you can get in front of the user no matter where they are.

      At the same time, most of my personal app downloads come from word of mouth, social media posts, sometimes even QR codes. I think the methods of reaching an end user will just continue to expand, so we need to make sure we can provide the tracking that marketers need to understand the complete picture of their channels.

  • AM

    Alexandre Moura

    over 2 years ago #

    Olá Fontaine,

    First, thanks for your time!

    When building Product Squads which roles do you think should be 100% dedicate to a single squad (e.g.: back end dev) and which resources do you think could be best allocated as a shared role across more than one squad/product (e.g.: front-end or UI).

    Also, what is your experience when it comes to hiring a dedicated resource vs bringin temporary consultants (e.g.: improving some specific UX to the next level).

    Thanks!

    • FF

      Fontaine Foxworth

      over 2 years ago #

      Ahh-- a timeless question! This is something that I’ve encountered at both places I’ve worked. If possible, my favorite has been working with teams with a variety of functions-- API engineers, FE engineers, QE engineers. This way, you are not dependent on having another team prioritize your features. Waiting for another team to build out an API can be a big blocker when you’re trying to ship functionality. What have been some observations you’ve made when it comes to resource allocation?

      Regarding contractors-- another great question. I’ve had the opportunity to work both with lots of in-house teams and lots of contractors. I think contractors work best for 2 reasons-- first of all, if they are subject matter experts in the area of development then they can train your in-house team. I found this particularly powerful when first diving into app development. Secondly, if the project is relatively contained, then it can be something that an outside firm could easily drive. If it’s a part of your code that is the backbone of your system or touches dozens of other product areas, it can be hard to have an outside group effectively deliver on the needs.

  • AA

    Anuj Adhiya

    over 2 years ago #

    One more if you have the time:
    What principles/frameworks/foundations that you learned getting your BS in Civil Engineering do you think, apply (if at all) towards building tech products?

Join over 70,000 growth pros from companies like Uber, Pinterest & Twitter

Get Weekly Top Posts
High five! You’re in.
SHARE
88
88