If you think about it, perhaps the most primitive form of customer-facing analytics is your car’s speedometer, odometer, and fuel gauge. They give you key metrics on how the car operates, and you use this data to take action, like slow down (some people, I mean) or go and get gas. In contrast, the diagnostic data accessed exclusively by the dealership during servicing isn't customer-facing because you can't see it.
What is customer-facing analytics?
Let's begin with a definition:
Customer-facing analytics are insights provided directly to you, the customer, about the service or product you are using.
These insights typically rely on large-scale aggregate data and come in the form of dashboards, APIs, reports, or data shares integrated into the product experience. They offer customers insights, helping them make informed decisions and better understand their interactions with the product or service.
Note that this definition does not prescribe how insights or analytics are delivered. They could be presented through an in-product dashboard, email, data share, or API.
What are examples of customer-facing analytics?
Let's take a look at some real-world examples of customer-facing analytics. These demonstrate how various types of consumer and business-to-business (B2B) products leverage analytics as part of their product experience.
1. Stripe dashboard and Stripe Sigma
The Stripe dashboard is a great example of customer-facing analytics. It’s the first thing you see when you log in to your Stripe account and provides an overview of your account's payment activity. It’s intended to give you a bird' s-eye view of your business and help you identify trends.
Although the main dashboard is great, it does not answer all customers' possible questions about their accounts. At a certain scale, customers need custom reports and to be able to answer ad hoc questions that come up, such as, “How did we do this Black Friday compared to last year?”
That’s why what they did with their Stripe Sigma product is interesting. Stripe Sigma provides a SQL-like interface that allows Stripe customers to query the raw data directly. This unblocks customers from answering any questions that might come up using SQL, a language familiar to many people. Stripe Sigma is a paid product.
2. Mux Data
Mux is a video streaming platform for web and mobile apps. One part of its business is streaming video (Mux Video), and the other provides insights into video engagement and the quality of experience (Mux Data).
Interestingly, this customer-facing analytics product started as a raw data export and has become one of Mux’s main products. Mux Data is a paid product.
Mux Data provides a wealth of metrics and dimensions for customers to filter on and offers a wide range of consumption mechanisms, including dashboards, APIs, and data exports.
Here is an example of how to fetch analytics data via the Mux Data API.
3. Twilio Voice Insights
Twilio Voice Insights was the product that inspired us to create Propel. It provides call quality analytics for developers to easily detect, diagnose, and troubleshoot any call quality issue. It includes metrics about network conditions, device capability, and overall call quality. Voice Insights is a paid product.
Twilio Voice Insights provides both aggregated and raw data via in-product dashboards, event streams, and APIs. These delivery strategies allow customers to easily access the data through very opinionated dashboards for this particular use case and utilize the API to build it into their internal tools and applications supporting their workflows.
4. Spotify Wrapped
Every year, Spotify puts together a personalized video of your year in music. This is not your typical SaaS dashboard, but it is still a great example of customer-facing analytics.
It takes a lot of data crunching to pull this unique experience for the millions of Spotify users worldwide. They, of course, deliver it masterfully in a wonderful and engaging animation.
Spotify keeps leaderboards for top artists, songs, and genres as well as metrics such as total minutes listened to achieve this.
5. Visier Embedded Analytics
Visier, a relatively unknown people analytics product, stands out due to its unique feature. Not only does it display vast amounts of analytics (inherent to its nature), but it also allows for embedding its reports and dashboards into third-party apps. Embedding analytics is common for BI tools, but this is a SaaS product that allows its customers to embed its analytics in their party apps.
By offering embedded analytics, they broaden the accessibility of their people analytics product to users of other applications where these analytics are integrated.
Alternative terms for customer-facing analytics
What can be confusing is that customer-facing analytics are often referred to by different terms. These terms have subtle differences, but all essentially describe customer-facing analytics. For example:
- Embedded analytics - traditionally refers to embedding a business intelligence (BI) tool into an application.
- Embedded dashboards - similar to the above, but specifically refer to embedding a dashboard.
- User-facing analytics - this term is often used interchangeably with customer-facing analytics, focusing on app users rather than internal employees.
- Data apps - are a broader term for applications that leverage large-scale analytical data, which customer-facing analytics apps do. Data apps, however, can be applications used internally by employees as well.
Customer-facing analytics vs. business intelligence (BI) vs. real-time analytics?
Traditional business intelligence (BI) focuses on internal data analysis for strategic decision-making within a company. On the other hand, customer-facing analytics is externally oriented, providing insights directly to the customers to enhance their experience and engagement with the product or service.
Real-time and customer-facing analytics are similar in that they serve use cases where data is processed and analyzed as it is generated, providing insights immediately or within a short timeframe. These systems can handle high volumes of data, offering immediate insights based on the most recent data, a characteristic we term data freshness.
Customer-facing analytics can be a form of real-time analytics, where these real-time insights are presented to end users. That means they serve more users than internal employee-based real-time analytics use cases.
How is customer-facing analytics different from product analytics?
Let’s go back to the definitions:
Product Analytics: How users engage with a product or service.
Customer-Facing Analytics: The insights provided directly to you, the customer, as part of the product experience.
Think of a product, such as Google AdWords, which is one of the most profitable products on the internet.
When you log in, the first thing you see is insights about your campaign performance. That's customer-facing analytics. How the internal team at Google measures where you click on that product and how many times you use it is product analytics.
Customer-facing analytics use cases
The use cases outline the purposes of customer-facing analytics, while the delivery strategies discussed later explain how these analytics are delivered.
Product usage
A perfect example of a product usage use case is Courier, a notifications API platform and a Propel customer. Courier provides developers with insightful usage data. This includes data on notification volumes and performance metrics, such as views, opens, and click-through rates. By surfacing this data, developers can understand which communication channels are generating the most engagement. Armed with these insights, they can effectively optimize their notifications strategy.
Operational insights
Operational insights make a profound impact in terms of understanding and improving the efficiency of operations. For instance, Lumeo, a computer vision analytics startup and a Propel customer, exemplifies this use case effectively. Lumeo provides analytics on computer vision models they run on real-time video streams for enterprises. This analytics layer enables enterprises to gain a clear operational view of their physical locations, allowing them to monitor performance, identify bottlenecks, and make data-informed decisions to optimize operations. Such operational insights are critical for enterprises to maintain efficiency and continually improve their operational processes.
Observability
PlanetScale’s Query Insights is a great example of the observability use case. In their words:
Query InsightsFind and optimize long-running queries in your application, anomalies in your database, and more.
PlanetScale Insights gives you a detailed look into all active queries running against your database. This in-dashboard tool allows you to identify queries that are running too often, too long, returning too much data, producing errors, and more. You can scroll through the performance graph to detect the time that a query was impacted and, if applicable, the Deploy Request that affected it.
Engagement metrics
Engagement metrics provide insights into the user's level of engagement. TikTok provides a prime example of these metrics. They track video views, profile views, likes, comments, shares, and unique viewers. These engagement metrics enable creators to evaluate their content's success and offer TikTok users valuable insights into the content that is worth watching.
Personalization
One of the most compelling use cases for customer-facing analytics is personalization. This use case involves using analytical data to tailor a product or service to an individual user's preferences, behaviors, and needs. Businesses can significantly improve user engagement by providing users with personalized insights and recommendations.
For instance, LinkedIn is a prime example of a company that utilizes customer-facing analytics for personalization. Based on a user's profile, connections, and activity, LinkedIn offers a personalized feed, job recommendations, and connection recommendations. This greatly enhances user engagement as they can access opportunities that align with their professional goals and interests.
For the ML recommendations to be accurate, the analytical data used to inform the feed needs to be served with sub-second latencies, high concurrency, and sub-minute data freshness.
Strategies for delivering customer-facing analytics
The delivery of customer-facing analytics does not follow a one-size-fits-all approach. The definition itself does not prescribe a specific delivery method; instead, the unique product experience guides the strategy. Many successful products tend to combine different delivery strategies, creating a blend that best suits their users' needs. This section explores various strategies for delivering customer-facing analytics, emphasizing that the ideal approach often involves a mix of these methods.
In-product metrics
In-product dashboards are one of the most common ways of delivering customer-facing analytics. These valuable analytics are baked directly into the core product workflows. This means that they are not just an add-on or an afterthought but an integral part of the product experience. They are designed with the user's needs and tasks in mind, providing relevant and actionable insights right where they are needed.
A prime example is the Google AdWords Keyword Planner. This tool provides a dashboard that helps users understand how a list of keywords might perform for a specific campaign. It provides insights into search volume trends and forecasts, competition levels, and suggested bid estimates for each keyword. The analytics are so deeply baked into the product experience that you’d never think about them as a dashboard.
Embedded analytics
Calendly, one of the most popular scheduling solutions introduced Calendly Analytics in early 2023. It’s a dashboard that helps admins understand key metrics such as popular meeting days, top performers, and in-demand meeting types. This feature is clearly built with a more traditional embedded analytics approach. It is very self-contained and not integrated with the rest of their product.
Analytics APIs
Brightcove, a video streaming platform, offers its customers analytics and insights via an API. It probably has large customers who need to integrate this data into their own applications and experiences. Analytics APIs are also common when products have a large partner ecosystem.
Data sharing
Customers sometimes need access to the raw data to perform their own analysis. When they are sophisticated enough to have a robust data platform and team, they also need access to the raw data to integrate into their reports and pipelines. This was the case with Salesforce, which announced support for direct data sharing with Snowflake. Salesforce customers can get their data directly into their Snowflake account.
Snyk, for example, uses email as a delivery mechanism for its customer-facing analytics. It sends a weekly email summary to each organization, providing essential insights. This email includes the total number of issues, active projects, and dependencies being monitored, along with a link to view the dashboard for a comprehensive status overview. It also displays the organization's plan status, indicating total usage, and suggests an upgrade when the limits are nearly reached.
If you haven’t read it, I highly recommend this blog post on Product usage emails.
Self-service reporting portals
Self-service reporting portals represent a powerful delivery strategy for customer-facing analytics. This approach empowers customers with the tools to customize their own reports, tailoring the experience to their unique needs and priorities.
An example of this strategy in action is Property Vista, a Propel customer in the real estate management space. Property Vista offers self-service reporting portals to its high-value customers. This feature enables these customers to customize their reports and completely tailor the experience to what matters most to them in their real estate management operations.
ML Models
Machine Learning (ML) models may not be the first thing that comes to mind when considering customer-facing analytics, but they are fundamentally related. The core ingredient for these models is the same: real-time, large-scale analytical data.
ML models play a crucial role in a vast array of customer experiences. For instance, they power churn and fraud prediction systems that can alert companies to potential issues before they escalate. They also drive e-commerce recommendations. These models analyze a user's behavior and preferences to recommend products they might be interested in.
In summary, while ML models might not traditionally be considered a form of customer-facing analytics, they actually are in a big way.
AI Agents
AI agents are programs that can perform tasks autonomously, often mimicking human-like behavior to a certain degree. They are equipped with the ability to analyze data and make decisions, learning from their interactions and experiences to improve their performance over time. The development of AI agents is one of the latest advancements in large language models. This allows them to process vast amounts of information rapidly, understand and respond to human language, recognize patterns in data, and even predict future outcomes based on historical information.
In the context of data analytics, AI agents can significantly streamline the data analysis process. They can automate the collection, cleaning, and sorting of data, freeing up analysts to focus on interpretation and strategy. Moreover, they can identify patterns and trends in datasets much faster than human analysts and with a lower margin of error.
It's early, but this might change quickly. We are very excited about this area.
Why do leading companies invest so much in customer-facing analytics?
You might wonder why leading companies like Stripe, Shopify, Hubspot, LinkedIn, and Twilio have heavily invested in customer-facing analytics. Some of the leading companies have thousands of engineers working on their customer-facing analytics products. Goals and objectives vary, but they can be distilled into serval core motivations:
Generate new revenue streams
The founding team at Propel spent a decade at Twilio, building APIs, data, and AI products. Like many modern companies, Twilio generated vast amounts of data, which was primarily used internally. The team realized that exposing this data externally unlocked a significant business opportunity.
For instance, one of the analytics products they launched, Voice Insights, quickly became one of Twilio's fastest revenue-growing products.
This product is an add-on product to Twilio’s core Voice offering, and it’s a must-have for any mission-critical voice deployment.
Similarly, for Mux Data, it seems to be a very strong revenue driver.
Reduce support load
Customer-facing analytics can significantly reduce a company's support load. By providing customers with the data and insights they need directly, companies can minimize the need for customer support interactions. This not only lowers the cost of support operations but also improves customer satisfaction, as customers can find the answers they need without waiting for a support agent's response.
A great example is Amazon Web Services (AWS). They offer extensive analytics and monitoring tools to their customers. These tools enable customers to keep track of their usage, monitor the performance of their services, and troubleshoot any issues that arise.
By offering these insights directly to customers, these companies can significantly reduce the volume of support inquiries they receive.
Boost user engagement
Metrics drive engagement. Facebook and Twitter/X know this very well. One of the first things Elon did when he took over Twitter was to add the view counter. This was simple UX and engagement improvement, but it was a massive engineering undertaking at Twitter’s scale.
Inform purchase decisions
Customer-facing analytics can also significantly influence purchase decisions. Presenting relevant data and trends allows customers to make more informed decisions. A great example of this is how Kayak, a travel search engine, uses customer-facing analytics to show trending ticket prices. By providing insights into whether the prices are rising or falling, customers can decide the best time to purchase their tickets, enhancing their decision-making process and overall purchasing experience.
Improve retention
Customer-facing analytics can significantly improve customer retention rates by continually providing value and insights that help customers achieve their goals. When customers see the tangible benefits of a product or service through data and analytics, they are encouraged to continue using that product or service.
For instance, fitness apps like MyFitnessPal or Fitbit provide users with detailed analytics about their exercise routines, diet, and health progress. These insights keep users engaged and motivate them to continue using the app to monitor their fitness journey.
Similarly, Netflix uses customer-facing analytics to power personalized movies and show recommendations based on a user's viewing history. This tailored experience makes users more likely to continue using the service, as they receive content recommendations that align with their preferences.
Technical requirements for customer-facing analytics
There are three fundamental requirements to provide a great customer-facing analytics experience. First, the ability to deliver a native product experience, meaning that analytics should be seamlessly built into the product experience. Second, sub-second query latencies and support for high concurrency ensure fast load times for thousands of users. Lastly, sub-minute data freshness guarantees that the most recent and relevant data is being used.
In the following sections, we'll unpack each of these requirements.
Native product experience
One key requirement for customer-facing analytics is a native product experience. This means that the analytics must be seamlessly integrated into the product, making it a core part of the user experience. The analytics shouldn't feel like an add-on or separate tool but rather a natural product extension.
To achieve this, the analytics interface should be designed and built to align with the overall look and feel of the product. This requires APIs to integrate deeply into the product. Approaches that rely on embedded iframes typically result in clunky and inelegant product experiences.
Analytics should be contextually integrated into the product workflows. They should be readily available where users need them, not only in a “Dashboard” or “Reporting” section.
Sub-second latency and high concurrency
Since analytics products serve end users rather than internal employees, they must be designed to support a dramatically higher number of concurrent requests than traditional data tools. This high concurrency is necessary to ensure that all users can access and use the product without experiencing slowdowns or interruptions, even during peak usage times.
In addition to handling high concurrency, it's also crucial that these products offer sub-second latency. Customers today expect fast, responsive product experiences. While internal employees might tolerate waiting several minutes for a query to run, this is not acceptable for user-facing applications, particularly when analytics is part of the core workflow that the product enables.
Sub-minute data freshness
The second fundamental requirement for customer-facing analytics is delivering fresh, up-to-date data. The value of data has an exponential decay curve, making it less valuable as time passes. Outdated data, even if only by a few minutes, can lead to customer confusion and frustration and can harm the product experience.
Sub-minute data freshness refers to the system's ability to ingest, process, and display new data in under a minute. This data refresh rate ensures that users always interact with the most recent and relevant data. Whether it is a dashboard displaying live traffic data for a website or a financial application showing real-time stock prices, the need for sub-minute data freshness is critical in many customer-facing analytics scenarios.
Conclusion
In conclusion, customer-facing analytics are becoming increasingly significant for modern products. They offer immense benefits, including improved user engagement, reduced support load, informed purchase decisions, and increased revenue streams. However, delivering a high-quality customer-facing analytics experience requires careful consideration of factors such as native product experience, sub-second latency, high concurrency, and sub-minute data freshness.
The unique technical requirements of customer-facing analytics can present challenges, but these challenges can be overcome with the right technology and approach. As we continue to evolve in our understanding and implementation of customer-facing analytics, it's clear that they will play a pivotal role in shaping the future of product experiences.
If you are considering implementing customer-facing analytics or looking to improve your existing system, check out Propel. Propel is a serverless ClickHouse platform designed to build efficient, high-performance customer-facing analytics. With Propel, you can leverage the power of real-time analytics to deliver exceptional product experiences.
