Most software companies want to know more about how their customers and trial users are engaging with their software. Increasingly, they realize raw data isn’t enough: they need actionable insights for optimizing product development, increasing adoption, and growing sales.
The solution has a name: software usage analytics. But it’s still relatively new, and there are key differences in how potential solutions work. If you want a solution that’s manageable, scalable, cost-effective, and quick to deliver ROI, you need to understand those differences upfront.
Software usage analytics has unique characteristics and requirements. We know, because we built Revulytics Usage Intelligence specifically for the job. For example, we provided native support for distributed applications on Windows, Macintosh, and Linux platforms, and extensive out-of-the-box tools for:
As we speak to software companies, some ask whether they can do the same job with Microsoft Application Insights on the Azure cloud platform (now known as Azure Monitor). It’s a fair question, because Application Insights is a powerful tool for tracking and troubleshooting Azure cloud applications. However, it focuses largely on performance, not usability – and it simply wasn’t designed to instrument desktop applications.
We know developers who’ve tried to use Microsoft Application Insights for software usage analytics on distributed client software. They say it’s a lot like building their own custom telemetry and analytics systems from the ground up. They’ve had to dedicate significant engineering resources to collect, visualize, and analyze usage data, and as their requirements become more complex, their investments in development and maintenance grew significantly. They weren’t expecting it to be nearly so difficult or costly.
To see why, let’s walk through the software usage analytics lifecycle, and compare solutions based on Revulytics Usage Intelligence vs. Microsoft Application Insights.
Plenty of planning knowledge is built into Revulytics Usage Intelligence. Out of the box, Revulytics:
Microsoft Application Insights handles aspects of infrastructure associated with its underlying cloud platform, and stores data in an Azure database. But consider all you still have to plan on your own:
Both Revulytics Usage Intelligence and Microsoft Application Insights let you instrument your client applications via an API that can natively support multiple languages. However, because Revulytics’ solution was purpose-built for software usage analytics, it includes several crucial capabilities absent from Microsoft’s.
For example, since the Revulytics Usage Intelligence SDK automatically generates and manages machine fingerprints and user installation IDs, it’s easy to build unique user profiles, capture usage trends for each installation, and link individual installations to download sources and marketing campaigns.
Revulytics also provides automatic licensing validation, and an easy customer opt-in/opt-out mechanism. In addition, its lightweight SDK was designed from the ground up to avoid application bloat or performance impact.
When it comes to delivering telemetry, both Revulytics and Microsoft provide APIs – but, once more, the differences reflect Revulytics’ targeted focus on desktop usage analytics. Only Revulytics provides built-in mechanisms to:
Depending on the complexity of your telemetry, building comparable functionality for Microsoft Application Insights may take months.
Revulytics gives you the flexibility to maximize the value of your usage data, but Microsoft establishes limits that can prove costly. For example, Revulytics offers unlimited analytics and reporting data export via a Web API, and lets you export raw event data. In contrast, Microsoft charges for its Continuous Export feature (only available with paid subscription). Moreover, you can only export directly to Microsoft Azure, which requires paid storage and an Azure subscription. Only after you’ve exported data to Azure can you re-export it to your own premises-based systems.
Revulytics retains telemetry data for at least a year (extensible to five years upon request), whereas Microsoft retains it for only 90 days. If you want to retain your Microsoft Application Insights data longer to understand trends over time, you must implement a process to regularly export it.
Out of the box, Revulytics includes many flexible reports designed specifically to help software companies answer key questions about how users engage with their products. All these reports can be customized, segmented and filtered to drill down into specifics, without requiring users to learn any reporting syntax. Revulytics’ reports include details about:
Microsoft Application Insights dashboards provide significant data on web application performance and failure. However, if you want reporting on how your customers are using your distributed client applications, you need to build most of it from scratch, or create custom dashboards within Microsoft’s Power BI analytics engine, potentially at added cost.
Since Revulytics’ built-in reports reflect the experience of scores of software companies, they satisfy most stakeholders. You won’t need SQL skills to build new reports requested by your developers, product managers, marketers, or anyone else.
Revulytics also provides User Flow reporting, allowing you to understand the specific paths that users take within your application. This is the only report of its kind, allowing you to visualize the user journey. Not only do you get the benefit of the same drill-down analysis which is featured in other Revulytics reports, but this report is available for any tracked feature out-of-the-box.
If you want to capture more custom data or build specific reports to answer new questions, the Revulytics Usage Intelligence client and reporting framework supports this natively. For example, you might collect specific string data or name-value pairs whenever an event occurs, to capture more detailed user environment or application status parameters.
Revulytics also supports integration with third-party frameworks with little additional development. Moreover, our Web Reporting API permits real-time reporting in JSON or graphical formats. All this means it’s easy to integrate usage intelligence with your other business systems, no matter who you bought them from.
Getting the right message to the right user at the right time is crucial to building customer adoption, receiving critical feature usage feedback, and promoting continued revenue streams, whether via upgrades or ongoing subscriptions. Often, the best time to communicate with your users is when they’re engaging with your software. For example, you may want to:
A full software usage analytics solution should include robust in-app messaging for distributed client software. Revulytics Usage Intelligence does. Our ReachOut feature lets you customize messages to different segments of users, and then deliver them either via an automated pop-up, or in specific message areas built into your product’s UI. It can even present quick surveys and forward users’ answers to your survey framework of choice. With ReachOut, you can:
Microsoft Application Insights contains none of these features. To build equivalent functionality capable of building effective, contextually relevant messages, you’d first have to match message delivery with usage intelligence data collected from each installation. Revulytics ReachOut does that out of the box.
As we’ve seen, choosing Microsoft Application Insights is likely to drive significant development costs related to client instrumentation, telemetry, caching, data integration, and reporting. You avoid these costs with Revulytics.
But there are other costs to consider. For example, a Revulytics subscription includes free support; Microsoft offers no free support. Revulytics subscriptions charge a flat monthly fee based on monthly active users, regardless of their activity. With Microsoft’s basic plan, costs increase proportionally with event data. That’s a disincentive to use Microsoft Application Insights for deep research about customer usage. Microsoft’s Enterprise plan sets data collection limits per node, with cost directly dependent on how many concurrent users send telemetry every hour. Customer usage patterns or increased activity in certain time zones can impact your costs, and may be difficult to predict.
Costs are only one side of the equation. The other is effectiveness: what does it take to get actionable, useful answers?
Consider an example. Software developers often need to understand whether specific features are being used, how they’re being used, who is using them, and whether usability problems might be interfering.
Out of the box, Revulytics Usage Intelligence captures all the information most software companies need to answer questions like these. It can segment and filter granular usage information based on software editions and versions, user hardware and software environments, and any custom-defined property. You can assess the number and duration of attempts to use a feature; related features used or left unused, and many other factors. All this helps you quickly pinpoint problems and opportunities related to new features, so you can precisely target development investments.
Since Microsoft Application Insights offers no comparable out-of-the-box data capture or reporting, software companies must first identify the metrics they want to collect, without the benefit of experience across many software vendors and products. Since Microsoft’s solution is designed for cloud rather than distributed applications, you need to build client services to collect data. Once your data is being collected in an Azure-based database or data warehouse, you must construct SQL reports to perform all the necessary data mining.
Given enough time, resources, and skills, you could do all that – but it might take months or years to start receiving useful analytics. In a fast-moving business environment, that’s an eternity.
Of course, getting analytics up-and-running is only the beginning. When analytics become available, multiple stakeholders quickly recognize its value. They rapidly start asking new questions and requesting new reports. Even more questions arise as products, competitors, and markets evolve.
Many of these questions are anticipated by Revulytics’ out-of-the-box data capture and reporting. In contrast, with Microsoft Application Insights, new reports must be custom-built – and if these reports require new source data, costly changes to client instrumentation may be needed.
To summarize, Microsoft Application Insights excels at measuring the performance of Microsoft Azure cloud applications. You can overcome its shortcomings in tracking distributed Windows, macOS, or Linux applications, but you’ll need to dedicate significant development and extensive ongoing support. Chances are, you’d rather focus those scarce resources on building great customer features and experiences – not on building analytics infrastructure your customers will never see.
Keith is Revulytics’ VP, Software Analytics and was the co-founder and CEO of Trackerbird Software Analytics before the company was acquired by Revulytics in 2016. Following the acquisition, Keith joined the Revulytics team and is now responsible for the strategic direction and growth of the Usage Analytics business within the company. Prior to founding Trackerbird, Keith held senior product roles at GFI Software where he was responsible for the product roadmap and revenue growth for various security products in the company's portfolio. Keith also brings with him 10 years of IT consultancy experience in the SMB space. Keith has a Masters in Computer Science from the University of Malta, specializing in high performance computing.
Remember the still-maligned Windows Vista operating system upgrade? What about Google’s Lively 3-D virtual world ...
Revulytics sponsors a series of Product Management Today webinars featuring innovative ideas from top software product management ...
We recently shared a Q&A with Nir Eyal about a webinar Revulytics sponsored that covered some of the ideas in his New York Times ...