A complete buyer’s guide to user analytics tools
What is a user analytics tool?
A user analytics tool reveals the actions users take within a digital product. It ties those users to the events they trigger and tracks their breadcrumb trail through the website or app. Teams can use this information to improve everything from the product’s usability to their marketing conversions, retention, and customer lifetime value.
Why buy user analytics?
User analytics are rich in benefits, both real and intangible. They help solve the mystery of what users want, need, and feel, and give teams greater confidence in their decisions. Unlike an all-purpose business intelligence tool, user analytics are built to show product, marketing, and analytics teams precisely what features each customer uses, the paths they take, and what makes them quit and not return. It’s a guidebook for making the product more intuitive, profitable, and enticing.
A financial services app, for instance, could learn that users acquired from Facebook are worth more than the typical user, and decide to increase their social media advertising. A news site could learn which ledes perform best and provide guidelines to its writing staff, and enterprise software could isolate bugs to prevent support ticket spikes.
Without user analytics, teams fall prey to the vagaries of guesswork and intuition. It’s easy for people who work closely with the product to assume they know what users are thinking but it’s an entirely different thing to discover it through data.
For most businesses, data from user analytics is the most efficient way to reach quick, practical insights. Ninety-seven percent of users who churn do so silently. They don’t offer feedback before it’s too late. And, while user interviews are useful for empathizing with customers, they don’t provide teams the statistical certainty they need to make tough calls, such as whether to sunset an underused feature.
User analytics provides:
- Unified user data: Aggregated information from across websites, apps, and devices for a complete customer view
- Sharing across silos: Teams from all departments can answer their own questions
- Visual interface: Intuitive reporting and segmentation
- Quick answers: Distill data into actionable user knowledge
- Messaging: Send notifications to users and A/B test features
While this may all sound great, there are dozens of options out there, including Mixpanel. Each has advantages and disadvantages and it can take months for a team to narrow their options. This guide is intended to lighten the load.
How to choose the right user analytics tool for my business
Evaluations for big-ticket purchases take time and involve more people than ever before. For analytics tools, this is probably a good thing. As the axiom goes, to save time, measure twice and cut once. User behavior analytics software can be time-consuming to install and teams can’t easily trial multiple options, so it’s essential to conduct extra-thorough research.
What constitutes the “right” tool depends on each business’ needs and constraints. Government agencies may have simple needs but strict constraints on purchasing software whereas a ride-hailing startup might have few constraints, but complicated needs. For every company, there’s at least one right-est solution.
An 11-point questionnaire for comparing user behavior analytics tools:
1. Product feature wishlist
Before teams look at any platforms, they ought to order pizza and shut themselves inside a conference room to decide what they want the behavioral analytics software to do for them. The team should be cross-functional, with representation from the product, engineering, marketing, support, and analytics teams. Each should present the use cases they’d need to make best use of the platform, and talk through the evaluation process.
Teams can rank features by necessity. Which are critical and which are simply nice to have? Is there anything that will auto-disqualify a vendor, such as not supporting iOS devices, or not complying with GDPR?
Teams can always add more items to the list as the evaluation progresses, but writing the list down allows them to compare vendors empirically.
Examples of common feature requirements:
- Track users across desktop and mobile platforms
- Offer a suite of pre-built reports
- Marketing funnels
- Anomaly detection and alerts
- Messaging and A/B testing
- Integrations and an API
- TSIA-certified support
2. Tracking targets
Which user events are teams hoping to track within the platform? Start with top-level goals such as gaining market share, driving ad revenue, or increasing usage, and then determine the KPIs and metrics that indicate progress toward that outcome.
Can the platform in question track the necessary KPIs and metrics with enough granularity? If the platform will track websites, how does it handle corner cases such as referral spam or ad blockers?
3. Reporting capabilities
Good analytics platform design is more than just an aesthetic choice—well-designed and intuitive graphical interfaces help teams find answers faster. Intuitive platforms get used more frequently and more usage leads to a higher return on the investment. Poorly designed interfaces, on the other hand, may check all the right boxes, but sit unused.
A good analytics platform should make it easy to export the data to a CSV file or BI tool if needed. Some vendors make exportation difficult on purpose, a process known as vendor-lock in, which makes teams dependent upon that one system. Avoid vendors that practice lock-in.
4. Technical requirements
Is there anything unique about your business’ product that will make it difficult to track? For example, does the website use Flash, rich media, or sit behind a restrictive firewall? Is there a custom-built CMS tag management system that might interfere user tracking? Any platform will have to clear these hurdles or offer a workaround.
The analytics platform should be reliable. Are there any data, file size, or throughput limits that might form a usage bottleneck? Will the platform develop latency as it houses more data?
The platform should also be compatible with the company’s current information architecture. Some analytics solutions are software-as-a-service (SaaS), some are server-hosted, and some are a blend. If there’s a hardware requirement, will it require the team to purchase new hardware? If so, factor that, plus the ongoing upkeep, into the total cost.
5. Legal requirements
Every country operates under its own data protection laws, and platforms can open the door to misuse. Any analytics providers should be able to show that it keeps its customers’ interests in mind and provides guide rails to prevent them from accidentally violating laws such as GDPR.
Teams should also consider long-term threats, like what happens to their data and support if a vendor goes bankrupt or is acquired. Consider the vendor’s financial condition and how long it’s been in business. Past performance isn’t a perfect predictor of longevity, but it’s a signal.
6. Analytics data
The quality of user data should be a paramount consideration for all teams. Understand how often the platform syncs data and how close to real-time it really is. Understand what users, events, and parameters it can track, and any limitations, such as how easy it is to access the data, and how many parameters it can store.
Plenty of analytics platforms have data limitations that don’t show up in the product documentation. For example, while a platform may collect billions of data points, it may only display limited sample data to keep the system from crashing. Or it may only allow businesses to segment data by a certain number of fields or over a certain period of time. For some buyers, these limitations can be deal-breakers and it’s crucial to root them out early.
And of course, who owns the data? Before settling on a vendor, always review its terms of service to ensure your team retains proper ownership.
7. Professional support
Ample support can be a double-edged sword. Companies that offer too much support may be indicating that their product is difficult to use. Companies that offer very little support may be indicating that their product isn’t very versatile. Somewhere between, there’s a happy medium: A platform that’s versatile and intuitive but provides white glove support if needed.
Quality of support also matters as much as quantity. Does the company offer 24/7/365 support? Are there different levels of support and do they cost money? How much onboarding does the company provide? How much time will the engineering team need to devote to implementing the platform, if any?
Investing in additional training, onboarding, and support may add cost but it can pay dividends. Teams that implement user tracking software faster and learn more quickly see a shorter time to ROI.
User analytics must integrate with your company’s tech stack, martech stack, and data ecosystem. Does the vendor offer ample pre-built integrations to common platforms such as CRMs, ERPs, and support tools, as well as a robust API?
If an online software, how will it integrate into your website? Are the cookies first or third-party? Can it measure dynamic content and decipher dynamic URLs?
Ultimately, can the system use the data from these systems to highlight interesting correlations? Could a SaaS product team, for example, see what users were doing within their platform before opening a help ticket in the support system? A platform that’s versatile and extensible will grow more useful over time and the team will find ever more uses for it.
Can the platform adapt to the team structure? Behavioral analytics software should offer user roles and permissions so interns can’t spill coffee and delete the code base, and partners, contractors, and users throughout the business don’t have a risky level of access.
Can the platform be upgraded without an onerous migration process or data loss? Are there interesting third-party add-ons that can enhance the system’s utility?
10. Demonstration or trial
Gather the cross-functional planning team for a demonstration and, if needed, a trial. It’s important all business units that will use the analytics software are represented and get answers to their questions. If one demonstration isn’t satisfactory, it may be easiest to split it into several smaller demonstrations where the vendor’s sales engineering team can cover each use case in detail.
Some software lends itself to trials; user analytics typically does not. Analytics implementations can be difficult and verifying that multiple new platforms are installed, tagged, and operational can be prohibitively time-consuming. If it’s possible for teams to reach a conclusion and rule out all other vendors without having to do a trial, that’s often preferable.
Before purchasing, speak to references. Have similar companies been successful? Have current users used the platform repeatedly throughout their career? Speaking to a happy customer isn’t necessarily an indicator that everything will go smoothly, but if companies can’t produce any at all, it’s a red flag.
Some platforms scale based on total data, total users, or a mixture. Does the platform’s pricing model match your revenue model? Will increases in platform cost be more than covered by increases in revenue?
Investigate the platform’s total cost of ownership. From additional hardware to add-ons, taxes, support, and upkeep–make sure to measure each platform based on complete costs.
For more, read Forrester’s report: The total economic impact of Mixpanel.
How to quantify the value of user analytics
What’s the value of greater insight? For most companies, that’s difficult to measure until after the fact. It’s both tangible and intangible—higher conversion rates, for example, may lead to deeper customer insights which lead to more conversions in a tangled web of causality. It can be easier to measure by displacement—how did a like business grow in the period following the introduction of user analytics? For this, case studies and references are vital.
To narrow down vendors and estimate their value as neatly as possible, create a comparison table, or use ours. Based on your company’s needs, weight your scoring system and measure platforms side-by-side. Take time and select carefully to find the best one for your business the first time around.