May 2017 Summaries

When a new feature gains popularity faster than expected, the backend pipeline and databases are put under immense pressure, leading to increased resource consumption and potential system failures. To address this issue, it's essential to build understanding of the system among the development team, identify areas where performance bottlenecks occur, and implement quick fixes to mitigate the impact. Optimizing the system by optimizing data access, using the right data stores, and re-evaluating how a feature is being used can lead to significant improvements in scalability and efficiency. By taking a structured approach, including product research and tweaks, teams can ensure their feature continues to grow without compromising system stability or customer experience.

Systems thinking in design is an approach where companies consider the big picture first and specific interactions later, establishing a design system early to set the product direction and focus design efforts on important areas. A system is defined as an interconnected set of elements that are coherently organized to achieve something. Applying systems thinking involves asking questions such as what are the core elements in the system, how they are connected, and their purpose. This approach can help uncover new possibilities, prioritize work, reuse existing components, and improve products by making them simpler and better. To apply systems thinking, teams can draw out their product's system on a whiteboard, ask questions such as how to achieve goals with fewer elements, and consider modifying the overall purpose of the system.

As startups grow, they're bombarded with opinions about their products from various sources, including customers, teammates, advisors, investors, media, and even non-customers who claim to represent others. However, not all feedback is created equal, and some types of feedback should be ignored due to their lack of relevance or accuracy. Aspirational statements, where people express what they want without considering the context or feasibility, can be misleading and often don't translate into actual behavior. Hypothetical or espoused behavior, such as predicting how customers will behave in certain scenarios, is also unreliable and should not be taken at face value. Third-party statements, which come from non-customers who claim to know what others might want or need, are particularly problematic and can lead to misguided product decisions. To build a successful product, it's essential to focus on real problems that customers are trying to solve, rather than chasing after aspirational goals or hypothetical scenarios.

Designing half of a conversation is at a fork in the road, with two main options: building toward a real understanding of what a person is saying and generating the right response, or creating controllable, scripted responses to a fixed set of commands. The former approach is challenging due to its reliance on artificial intelligence and can lead to creepy sci-fi futures, while the latter offers more predictable but less nuanced interactions. A conversation model called CLEAT, which emphasizes context, shared language, engagement, agreement, and transaction or action, provides a framework for designing conversations that prioritize understanding the other side of the conversation. By focusing on these elements, designers can create software with conversation at its heart, fostering human-like relationships between teams, customers, and technology. The key to better conversations lies in understanding context, reaching agreements, and delivering the right answer in the right way, while also making sure the bot has a job to do and there's agreement about its role. Allowing for imperfection by narrowing constraints and enriching context signals can help mitigate mistakes and build trust between humans and technology.

The concept of product marketing is often misunderstood as a solo activity focused on getting products off shelves. However, effective product marketing involves close collaboration with product teams to build products that meet customer needs and resonate with the market. At Intercom, product marketers play a crucial role in shaping the product development lifecycle, from understanding customers' jobs and pain points to creating compelling problem statements and solution stories. They also conduct competitor research, drive beta testing, create product positioning guides, develop product messaging, and ensure that the product meets end-to-end user experience standards. By working closely with cross-functional teams, product marketers help build products that achieve commercial success by connecting with customers from the very beginning.

The article emphasizes the importance of subject lines in email marketing, citing the Pareto Principle that a small number of factors can have a disproportionate impact on success. It argues that startups often underestimate the power of subject lines and instead rely on generic, impersonal emails that fail to grab attention. The ideal subject line is short, clear, and direct, with studies showing that emails with 5 words or less have higher reply rates. Personalization, particularly using the recipient's first name, can also increase engagement rates. Effective subject lines ask questions, spark curiosity, and clearly describe the content of the email. Ultimately, the article advises against relying on gimmicks or growth-hacking tricks and instead recommends following a simple golden rule: describing the subject of your email in the subject line itself.

When it comes to guiding customers to success with a product, a one-size-fits-all approach won't work. Instead, differentiate onboarding to help all customers master the product by tailoring instructional methods to individual learning styles and needs. This can be achieved through providing scaffolds such as great written documentation, how-to videos, and best practice blog posts that cater to different learning styles. Additionally, building time into conversations with customers, offering one-on-one or group tutoring, and using data to customize message schedules can help achieve a more personalized onboarding experience. By matching organizational structures to product logic and leveraging peer-to-peer learning techniques, businesses can create tailored support systems that engage users and foster loyalty.

Empathy engineering is a crucial aspect of software development, bridging the gap between code and customer needs. This approach involves engineers dialling into the needs of their customers, understanding their pain points, and empathizing with them to deliver better products. By setting up conversations that add value for both engineers and customers, engineers can identify areas of complexity and recurring bugs, track customer feedback, and make data-driven decisions to improve the product. Talking to existing customers helps engineers understand how code is being used, while talking to new customers provides an opportunity to explain how they can achieve specific use cases and avoid common issues. By incorporating empathy into engineering, teams can reduce the rate of incoming issues, gather positive and negative feedback, identify non-product feature requests, and prioritize tasks that improve product quality.

The text provides guidance on how to ask the right research questions to get valuable feedback for product development. It emphasizes the importance of defining clear goals and outcomes before starting research, and suggests various question types to achieve different objectives such as evaluating feature success, understanding user intent, gathering initial perceptions of a new feature, and investigating users' mental model of the product. The text also offers tips on how to analyze and act upon the feedback, including treating every clustering of feedback as a hypothesis and verifying it before building a solution. By asking the right questions and following these guidelines, companies can gain actionable insights that lead to better products and improved customer satisfaction.

We can make a significant impact on performance and optimization by sharing responsibility across our organization, having the right information available to the right people, and using easily comprehensible metrics. A small team of engineers improved the load time of Intercom's website from 4.5 seconds to below 1 second through compression, delay loading non-critical items, caching, and monitoring page load times. By making performance metrics available to a wider audience, we can encourage care and responsibility for optimization across our organization, leading to better visibility and even better performance.

At Intercom, the product development team spends a significant amount of time on problem prioritization and problem definition before designing or building anything. This approach is in contrast to many other companies where senior important individuals often come up with ideas that are then hastily designed and built without much consideration for the underlying problem. The author believes this approach is crucial for success, as it ensures that solutions are tailored to real customer needs rather than just solving a perceived problem. Despite its importance, this process is often undervalued due to its lack of glamour, difficulty in measuring progress, and challenges in selling its value to management.

At Intercom, a customer support team is embedded in the product team to ensure that customer feedback and voice are always considered during product ideation. This approach allows for tighter feedback loops between customer support, product teams, and engineering teams. The embedded role gives customer support reps a unique insight into both sides of the business, enabling them to champion the customer's cause and advocate for impactful changes that improve the product. By being part of the engineering team, customer support reps can better understand the technical limitations of the product and provide valuable feedback to improve bug reports and issue resolution. This approach has resulted in improved products, faster issue resolution, and more career progression opportunities for the support team.

The text discusses an alternative approach to building mock servers for mobile apps that communicate with a server through HTTP. The authors, at Intercom, implemented their server logic in Go and embedded it into their iOS and Android apps using the Speakeasy tool, which generates AAR files for Android and frameworks for iOS. This allows them to easily test requests that cause data updates in the backend and reduces duplication of work across platforms. By implementing this approach, they have seen benefits such as reduced workload in half when making changes to mock server logic, easier testing of real-time communication, and improved self-containment of their app. The tool is available on Github and can be used by creating a Go package that implements the `speakeasy.Engine` interface, running `speakeasy build`, and adding the server as a dependency in the app.

We're dropping the term "content marketing" at Intercom because it's become synonymous with spam and hacks, rather than a genuine way to create authentic content. The term "content" is problematic on its own as it commodifies what we do, and adding "marketing" to it undermines the value of what we create. Instead, we believe in focusing on publishing great content first, which will attract people to our product with minimal marketing effort required. This approach prioritizes creating something people want to read or watch over trying to market and sell from the get-go. By putting readers first and applying editorial principles, we aim to create high-quality content that resonates with our audience and sets us apart.