Logo

    #50 - Growth Office Hours with Anu Hariharan and Gustaf Alstromer

    en-usDecember 01, 2017
    What is the primary focus before pursuing growth?
    Who emphasized the importance of user retention?
    What metric indicates meaningful product usage?
    When should companies start building in-house analytics tools?
    How do companies assess user retention in various industries?

    Podcast Summary

    • Ensure strong user retention before focusing on growthFocus on improving user retention before scaling growth efforts, as a strong retention foundation is essential for sustainable growth.

      Before focusing on growth, it's crucial to ensure strong user retention. Anu Hariharan and Gustaf Alstromer, partners at Y Combinator, emphasized this. Growth hacking, a buzzword for quick growth solutions, is not a sustainable approach. Instead, a growth team, which applies a data-driven approach, comes into play after achieving product-market fit. Retention is the foundation of growth, and companies should first check if they have stable retention that is better than benchmarks. This may involve assessing whether users remain active and engaged over the long term, depending on the specific industry. For instance, a social network would require a minimum of 50-60% retention. Therefore, before scaling growth efforts, it's essential to evaluate and improve retention.

    • Focusing on repeat usage or retention for valuable insights when starting outCompanies can gain insights into their product's evolution and effectively scale growth by prioritizing retention and measuring key metrics such as repeat usage or number of second fixes within the first few months after the initial purchase.

      Measuring and evaluating the performance of a product or marketplace involves identifying key metrics that indicate meaningful usage. When starting out, it may be challenging to have a substantial amount of historical data, so focusing on repeat usage or retention can provide valuable insights. For instance, Airbnb tracked bookings as a sign of product usage, while Uber focused on completed trips. Stitch Fix, on the other hand, initially prioritized retention by measuring the number of second fixes within the first four months after the initial purchase. By improving retention, companies can better understand their product's evolution and effectively scale growth.

    • Understanding user retention in the early stages of a startupDetermine ideal user behavior, prioritize retention for cost savings, and establish a clear understanding of user behavior for growth foundation.

      For startups, focusing on user retention is crucial for long-term growth. Stitch Fix, for instance, understood this concept early on and prioritized it for their first four years, leading to organic growth and a shift to growth strategies in the last one and a half years. Retention is essential because acquiring new users is costly, and retaining existing ones saves marketing dollars. When starting, entrepreneurs should determine the ideal user behavior and frequency of use for their product or service. This information can be obtained by understanding the use case or analyzing typical retained users' behavior. Having a single source of truth data is also essential for accurate and effective metric analysis. In the early stages, startups may not have all the answers, but establishing a clear understanding of user behavior and retention is a solid foundation for growth.

    • Building In-House Analytics Tools: A Growing Trend for BusinessesBusinesses facing data management challenges may opt to build in-house analytics tools for better control and customization, involving setting up a database, ensuring data availability, and building tools for segmentation, event tracking, and experimentation.

      As businesses grow and collect data from various sources, they often face the challenge of storing and analyzing that data efficiently. Initially, they may use external tools like Segment or mixed panels to manage their data. However, when they reach a certain scale, they may decide to build their own analytics tools in-house for better control and customization. This process involves setting up a data event database, ensuring data availability, and building tools for segmentation, event tracking, and experimentation. The decision to build in-house tools depends on the size of the growth team and the resources available. At Airbnb, for instance, they started building their own analytics tools when they had around 35 engineers, and within a year, they had the basic components in place. Although it requires significant time and effort, building in-house tools offers benefits like better integration, customization, and control over data.

    • Making Data Accessible for Effective Decision-MakingCEOs of early and later stage companies should prioritize making data accessible to everyone, not just data engineers. Use external tools for quick access, maintain a common source of truth, and engage CEOs for alignment and action.

      For early and later stage companies, focusing on making data accessible and available to everyone throughout the organization is crucial for effective decision-making. A common mistake is making data someone's sole responsibility, such as a data engineer, rather than making it a company-wide priority. Building technology to make data accessible can take a lot of time, so starting with external tools like Mixpanel or Amplitude is recommended. Having a common source of truth is important, but CEO engagement and alignment on taking action based on data are even more essential. Balancing intuition with data is key for a CEO, and Zuckerberg at Facebook is an example of a successful CEO who did this. A great growth team can help quickly discover the impact of decisions and course correct when necessary, acting as a defense against potentially detrimental changes.

    • Balancing Data and Company VisionGrowth teams should find a balance between data-driven decisions and the company's long-term vision. Involve the entire team in the experiment review process to make informed decisions.

      Having a growth team in a company requires alignment between the CEO and the team's goals. Experimentation is crucial when dealing with complex products, but relying solely on data and ignoring the company's vision can lead to unsatisfying results. Facebook's famous "move fast and break things" approach may be effective for large companies with billions of users, but for smaller companies, breaking things can be detrimental. Experimentation is essential to understand the impact of changes, but the results are not always clear-cut, and different people may have varying opinions. Therefore, it's essential to involve the entire team in the experiment review process to make informed decisions. Growth teams should not rely solely on data and ignore the company's vision. Instead, they should work together to find a balance between data-driven decisions and the company's long-term goals.

    • Coexisting Product Vision and Growth StrategiesCompanies can have both strong product visions and optimized growth strategies. It's essential to communicate the purpose behind growth tactics to avoid manipulation and maintain authentic user experiences.

      While optimizing a product for growth and maintaining a strong product vision can seem like opposing forces, they can coexist. Companies like Facebook and Instagram, which have both strong product visions and optimized growth strategies, prove this. However, it's crucial to communicate the purpose behind growth tactics to avoid coming off as manipulative or "hacking" people's minds. For instance, SEO is not an abusive tactic, but rather a way to help people find the best answers to their queries on search engines. Similarly, Facebook and Instagram, despite being part of the same company, have distinct growth strategies due to their unique team dynamics and product visions. The key is to establish clear guidelines and heuristics for what growth tactics are acceptable within each team. Ultimately, the goal is to create a user experience that feels authentic and beneficial, rather than pushy or manipulative.

    • Establishing initial heuristics for a growth teamFounders need to understand users and establish a growth team with essential disciplines like engineering and data-savvy product management. Flexibility is crucial for international expansion and adapting to local tastes.

      When building and scaling a growth team for a company, it's crucial to establish initial heuristics while remaining open to adjustments for local adaptation. Every founder knows their product best, but understanding the users and their unique needs is essential for growth. A growth team typically includes disciplines like engineering, product, design, data science, and user research. In the beginning, engineers and a data-savvy product manager are essential. The term "growth team" is used to define a specific role in the company dedicated to growth, as it's challenging to assign this responsibility to an entire team. Being flexible with guidelines while maintaining core values is essential for international expansion and adapting to local tastes.

    • Aligning Growth and Product TeamsHaving a growth team within the product team reduces tension and ensures alignment. Ideal growth team members are data-driven, experimenters, and quick movers. CEO's endorsement is crucial. Effective communication between product and growth heads is essential.

      Having a dedicated growth team within an organization can help optimize specific areas of opportunity and eventually teach the rest of the company to adopt similar growth-focused mindsets. However, the location of this team within the organization, specifically whether it's part of the product team or a separate entity, can lead to tension and communication issues. The majority of growth experts recommend keeping the growth team within the product team to reduce tension and ensure alignment. The ideal growth team members are those who are willing to experiment, data-driven, and able to move quickly. The CEO's endorsement is crucial for the success of a growth team. The main reasons for potential tension include differing perspectives between product and growth teams and the perceived ownership of growth within the organization. Effective communication between the heads of product and growth is essential for success, regardless of the team's location within the organization.

    • Defining clear accountability metrics for each teamClearly defining metrics for growth and product teams ensures focused efforts and prevents confusion, allowing for optimal scaling of a company.

      Having clearly defined accountability metrics for each team, whether it's a product team or a growth team, is crucial for success in scaling a company. This was discussed in the context of Facebook, which maintains separate growth and product teams, and Uber, which had separate rider, driver, and marketplace teams before merging. The importance of this structure lies in ensuring that each team has a clear understanding of their specific responsibilities and metrics, preventing overlap and confusion. For founders, transitioning from a hands-on product development role to a more data-driven approach at scale can be uncomfortable, but the introduction of growth thinking is necessary for making informed decisions and achieving optimal results.

    Recent Episodes from Y Combinator

    Consumer is back, What’s getting funded now, Immaculate vibes | Lightcone Podcast

    Consumer is back, What’s getting funded now, Immaculate vibes | Lightcone Podcast

    What's happening in startups right now and how can you get ahead of the curve? In this episode of the Lightcone podcast, we dive deep into the major trends we're seeing from the most recent batch of YC using data we've never shared publicly before. This is a glimpse into what might be the most exciting moment to be a startup founder ever. It's time to build. YC is accepting late applications for the Summer 24 batch: ycombinator.com/apply

    When Should You Trust Your Gut? | Dalton & Michael Podcast

    When Should You Trust Your Gut? | Dalton & Michael Podcast

    When you’re making important decisions as a founder — like what to build or how it should work — should you spend lots of time gathering input from others or just trust your gut? In this episode of Dalton & Michael, we talk more about this and how to know when you should spend time validating and when to just commit. Apply to Y Combinator: https://yc.link/DandM-apply Work at a Startup: https://yc.link/DandM-jobs

    Inside The Hard Tech Startups Turning Sci-Fi Into Reality | Lightcone Podcast

    Inside The Hard Tech Startups Turning Sci-Fi Into Reality | Lightcone Podcast

    YC has become a surprising force in the hard tech world, funding startups building physical products from satellites to rockets to electric planes. In this episode of Lightcone, we go behind the scenes to explore how YC advises founders on their ambitious startups. We also take a look at a number of YC's hard tech companies and how they got started with little time or money.

    Building AI Models Faster And Cheaper Than You Think | Lightcone Podcast

    Building AI Models Faster And Cheaper Than You Think | Lightcone Podcast

    If you read articles about companies like OpenAI and Anthropic training foundation models, it would be natural to assume that if you don’t have a billion dollars or the resources of a large company, you can’t train your own foundational models. But the opposite is true. In this episode of the Lightcone Podcast, we discuss the strategies to build a foundational model from scratch in less than 3 months with examples of YC companies doing just that. We also get an exclusive look at Open AI's Sora!

    Building Confidence In Yourself and Your Ideas | Dalton & Michael Podcast

    Building Confidence In Yourself and Your Ideas | Dalton & Michael Podcast

    One trait that many great founders share is conviction. In this episode of Dalton & Michael, we’ll talk about finding confidence in what you're building, the dangers of inaccurate assumptions, and a question founders need to ask themselves before they start trying to sell to anyone else. Apply to Y Combinator: https://yc.link/DandM-apply Work at a Startup: https://yc.link/DandM-jobs

    Stop Innovating (On The Wrong Things) | Dalton & Michael Podcast

    Stop Innovating (On The Wrong Things) | Dalton & Michael Podcast

    Startups need to innovate to succeed. But not all innovation is made equal and reinventing some common best practices could actually hinder your company. In this episode, Dalton Caldwell and Michael Seibel discuss the common innovation pitfalls founders should avoid so they can better focus on their product and their customers. Apply to Y Combinator: https://yc.link/DandM-apply Work at a Startup: https://yc.link/DandM-jobs

    Should Your Startup Bootstrap or Raise Venture Capital?

    Should Your Startup Bootstrap or Raise Venture Capital?

    Within the world of startups, you'll find lots of discourse online about the experiences of founders bootstrapping their startups versus the founders who have raised venture capital to fund their companies. Is one better than the other? Truth is, it may not be so black and white. Dalton Caldwell and Michael Seibel discuss the virtues and struggles of both paths. Apply to Y Combinator: https://yc.link/DandM-apply Work at a Startup: https://yc.link/DandM-jobs

    Related Episodes

    How to Build An MVP with Michael Seibel | Startup School

    How to Build An MVP with Michael Seibel | Startup School

    YC Group Partner, Michael Seibel, explains how to build a minimum viable product (MVP) for your startup idea. Using examples from real YC companies, Michael walks through how to determine your MVP feature set, build prototypes and demos for user testing, and present your MVP to early customers or investors.


    Apply to Y Combinator: https://yc.link/SUS-apply Work at a startup: https://yc.link/SUS-jobs

    Data Processing Agreements (DPAs) that Embrace Reality

    Data Processing Agreements (DPAs) that Embrace Reality

    In this episode, Arohi Kashyap tears down the Salesforce Data Processing Addendum (DPA).

    Companies that deal with private data every day may not see managing data as mission-critical. However, as Kashyap points out, corporate counsel needs to understand and manage data risks.

    Kashyap uses Salesforce's agreement to illustrate a few important principles:

    •Roles must defined, particularly as they shift during an engagement,
    •Liability should be clear in the event of a plausible data breach, and
    •Counsel must understand that not-urgent is not the same as not-important when it comes to data risks.

    Review the contract here: https://www.salesforce.com/content/dam/web/en_us/www/documents/legal/Agreements/data-processing-addendum.pdf

    Potluck - CSS × Angular × Dev job preparation × Svelte × File organization × Gear × More!

    Potluck - CSS × Angular × Dev job preparation × Svelte × File organization × Gear × More!

    It’s another Potluck! In this episode, Scott and Wes answer your questions about CSS, Angular, Svelte, dev job preparation and more!

    Sentry - Sponsor

    If you want to know what’s happening with your errors, track them with Sentry. Sentry is open-source error tracking that helps developers monitor and fix crashes in real time. Cut your time on error resolution from five hours to five minutes. It works with any language and integrates with dozens of other services. Syntax listeners can get two months for free by visiting Sentry and using the coupon code “tastytreat”.

    Sanity - Sponsor

    Sanity.io is a real-time headless CMS with a fully customizable Content Studio built in React. Get a Sanity powered site up and running in minutes at Sanity. Get an awesome supercharged free developer plan on Sanity.

    Show Notes

    3:23 - Q: How big is too big for state? I have an object of around 1000 items which will be used in a search box on the frontend of my site, but the list of items won’t be updated very often. What is the most efficient way to store these items and deliver them to the browser? Would you use state, a JSON object, a MongoDB document or some other way I haven’t thought of?

    7:17 - Q: How much do I need to know before I start applying for dev jobs?

    9:26 - Q: Do you folks have any opinions on Skeleton loaders, and any thoughts on how best to implement them? At my day job we’ve recently implemented skeleton loaders in a way that inherits the parent styles. Would love to know if there’s any best practices for this or ways we can improve our setup.

    13:01 - Q: Do you think we still need to use CSS resets like normalize and sanitizer?

    14:56 - Q: Scott, can you tell the story of how you landed an interview for a job in Japan? I am moving to Japan next year and wondering how that opportunity arose for you.

    20:04 - Q: Do you guys have an opinion on angular good or bad? Or is it that you’re more comfortable and have used React and Vue more often, so they’re largely at the center of any framework discussions?

    22:54 - Q: Could you talk about naming and structuring files/folders in your applications? How do you name components and folders? Do you follow some sort of path of how to structure files in a web app?

    29:01 - Q: Do you think that Node is almost too powerful to handle some of the simpler tasks (a site that just needs a contact form for instance), or am I just missing some very simple configuration? It seems like other languages, such as PHP, might be better suited for some of these simple tasks and Node is really just meant for larger scale application? I guess what my question really boils down to is, for a freelancer, I understand that I need to have my hands in a lot of different tech stacks, but should I really dive in on PHP (or something else) to handle some of these simpler features a client might need?

    33:24 - Q: What do you guys think about Svelte? Before everything goes WASM will Svelte be the future of web programming?

    37:24 - Q: What smart phones and laptops do you use for work? (specific model and why)

    42:38 - Q: Should I apply for other jobs even if my contract might get renewed? Should I feel loyal to a company?

    ××× SIIIIICK ××× PIIIICKS ×××

    Shameless Plugs

    Tweet us your tasty treats!

    From Subscription Surprises to E-commerce Enigmas with Paul Jarrett

    From Subscription Surprises to E-commerce Enigmas with Paul Jarrett
    Dive into the dynamic world of e-commerce with Paul Jarrett, CEO and co-founder of Bulu, a leading force in subscription boxes. In this episode, unveil the nuances of blending e-commerce with subscription models, grasp the finesse of negotiation, and uncover insights that can revolutionize your e-commerce journey. Whether you're a novice or a veteran in e-commerce, there's something for everyone. Listen now for a wealth of knowledge!