There are many tools you can use to measure your website traffic (see this roundup) and get varying degrees of data and insight about your visitors. For subscription based businesses, apps and some other business models, it makes sense to take things a step further and that's where advanced customer analytics come in.
In this roundup, I'm comparing Kissmetrics, Mixpanel, Woopra, Calq, Indicative and Trakio.​
More...
Overview
What's the difference between a "normal" analytics tool and an advanced customer analytics tool?
In some cases the lines are blurred, but one of the most important points of difference is the emphasis on tracking individual users instead of just providing aggregate data about website visitors.
Advanced customer analytics tools make a lot of sense for a business where people register for an account and where there's a lot of variance in the value of individual customers (e.g. a subscription based business where someone who remains a member for years is worth more than someone who cancels after the first month).​
These tools are built to answer questions such as:
- ​How does a recent change we made to our app affect the retention rate of our customers?
- How does a change in pricing affect not only sales conversions but customer lifetime value?
- How do new users engage with our app or website and where can we make improvements to turn more of them into loyal customers?
Common Features
What all the tools have in common are:
- ​Funnel reports - see multiple steps towards a conversion and drop-off rates for each step.
- Advanced queries and segmentation - e.g. show users who've performed a specific set of actions.
- Insight into individual user profiles - see what actual people do, instead of just an aggregate of pageviews and visits.
- Cohort and retention analysis - insights into how changes affect long-term behavior among customers.
I won't mention any of the basic features in the following and instead, highlight what makes each of the services unique:
Calq​
Calq is a new entry in this market, but it is catching up fast. The terms that come to mind when thinking about Calq are "developer friendly" and "startup friendly".
Developers will feel right at home with Calq's "action query language" or AQL, a language based on SQL, which can be used to create very complex queries into the data in the tool.
Further, users can run any query and get any report in Calq through their API, making deep integrations with other tools possible.
If all this sounds a bit intimidating, fear not: Calq isn't made only for developers. Like the other tools, it also has a visual interface and a visual query builder, so us non-techie can use it and get the answers we need, without having to dig into query languages and APIs.​
If you're a startup, you can get Calq for free for 6 months and then at a reduced price for another 6 months after that.
Platforms: | Web, iOS, Android |
Free Level? | Free Plan (1,000 users/month) |
Price Based On: | Users |
Price Range: | $50 - $1,000+ per month |
Indicative
Indicative is another newcomer in this segment, having only recently come out of beta.
The first advantage I noticed with Indicative is the dashboard feature. With this tool, you can create a fully customized dashboard, giving you on overview of all the metrics that matter most, to your business.
In addition, you can create multiple dashboards (for different teams or team members, for example) and also​ share the dashboards. With this, Indicative has the potential to make dedicated dashboard apps like Geckoboard or Chartio.
Indicative's user interface is unusual, but I had no trouble finding my way around (they also have a great, interactive tutorial that explains the reports very well). The interface is highly visual and I found it easy to slice and dice the data and get (most of) the reports I was looking for.
At the time of testing, some important features like revenue tracking were still missing, but the team at Indicative assured us that they were working on adding them.​
Platforms: | Web, iOS, Android |
Free Level? | Free Plan (10,000 users/month) |
Price Based On: | Users |
Price Range: | $79 per month, per 50k users |
Click here to learn more about Indicative
Kissmetrics
Kissmetrics is probably the most widely used among the solutions in this roundup. It's also the most elusive one, since the website doesn't contain detailed feature lists and no one replied to my request for more information, either.
An interesting feature in Kissmetrics is the so called "path report", which is built to map out the most commonly taken steps users take between first appearing and reaching a conversion goal specified by you. Think of it as a report that automatically digs through your data to find the most common (and most successful) funnels you may not even know about yet.​
A feature I liked in Kissmetrics is that you can track events or people by simply adding a URL parameter to a link. This is great for tracking email links, for example and it is something that doesn't need any technical expertise to implement.​
Platforms: | Web, iOS, Android |
Free Level? | Free Trial |
Price Based On: | Events |
Price Range: | $179 - $599+ per month |
Mixpanel​
Mixpanel is a well-established solution that has been around for several years and has been steadily growing in both user-base and features.
There are two main points that set Mixpanel apart. The first is a suite of features that can be used to send in-app or email messages to your customers, based on their behavior. You can create rules that will trigger messages based on any of the events you track in Mixpanel.
For example, you can send an email to customers who signed up, but haven't used/downloaded your product after a certain number of days.
Mixpanel also has a strong focus on mobile and includes a visual A/B testing tool for mobile apps. You can use it to create any kind of split test inside ​your mobile app, without requiring a 3rd-party testing tool.
You can also use Mixpanel to display in-app surveys and gather further insights from your users.​
Mixpanel is the only solution that bills for both actions and users. For example, if you are tracking 40,000 people ($150) and 600,000 events ($350) per month, the total cost for the tool will be $500/month.
Platforms: | Web, iOS, Android |
Free Level? | Free Plan (25,000 events/month) |
Price Based On: | Users & Events Billed |
Price Range: | $300 - $2,000+ per month |
Click here to check out Mixpanel
Trakio
Trakio is a service that has undergone an evolution from an advanced analytics tool to a pure (and higher end) customer analytics service.
That means Trakio's purpose is deliberately not to provide a full range of data following the process from first visit, through signup etc. Instead, it aims to only look at what happens within your product, membership or app after users have signed up for an account (i.e. after they are no longer anonymous).​
In practice, this means it is concerned with engagement and retention rather than conversions.​
Like Indicative, it features a customizable dashboard where you can get an overview of some of your business metrics. Trakio also goes into very granular detail in individual customer profiles and features an "alert" function that can tell you when important accounts are showing negative signs (anything that indicates they might be about to lose interest and cancel).
It's also worth noting that Trakio is only available with yearly contracts.​
Platforms: | Web, iOS, Android |
Free Level? | Free Trial |
Price Based On: | Users |
Price Range: | $3,000 - $72,000 per year |
Woopra
Woopra has been around for a long time and in the past few years it has undergone a dramatic transformation from being a "regular" web analytics tool to becoming a high end customer analytics tool.
One of Woopra's main selling points is the AppConnect feature. As the name implies, this feature creates a connection between Woopra and various apps or 3rd-party services such as Stripe, HipChat, Freshdesk and many more.​
In practice, this means you can integrate multiple services without having to touch any code. And the genius of AppConnect is that it works in 2 directions: on the one hand, you can check your customer profiles in Woopra and not only see how they interact with your site, but also see data about support tickets they created, live chats they participated in or emails they opened.
On the other hand, you can trigger actions in the 3rd-party apps, based on rules and segments in Woopra.​ For example, you could trigger an invitation to a live chat to display, based on what pages a visitor has been looking at.
Platforms: | Web, iOS, Android |
Free Level? | Free Plan (30,000 events/month) |
Price Based On: | Events |
Price Range: | $80 - $1,200+ per month |
A Note About Pricing Bases
I've added price range information for each of the review candidates, based on the prices they publish on their websites. Please note that the prices for each of these solutions are open-ended, as they all have a "call us" level of pricing that kicks in after a certain usage volume.
A more important point is about whether they charge based on users or actions/events.​
To understand exactly what that means, let's consider this scenario: a customer visits your website, logs in, watches 2 tutorial videos and starts a download.
This equals to 5 actions taken by the user (or 5 events reported). It also equals to one single user.​
In my opinion, tools that charge on an action/event basis are inhibiting and this pricing model reduces the practical usefulness of the tool. ​
Here's why: unless you have an unlimited budget (anyone?), the price of the tool will matter to some degree. If you're being charged based on events, you'll inevitably spend some of your time and effort thinking about the price implications of your tracking. You'll probably make decisions about which events to send to the analytics tool and which ones to leave un-tracked.
The whole point of data analysis is that you don't know in advance what matters and what doesn't. You don't know which interactions your customer has with your product or app are important and which ones aren't.​
That's why it's best to have a large bucket with all of your data, to dig around in and uncover the answers to questions that will drive your business forward.​
In short, the pricing model based on events is certainly more lucrative for the service provider, but the pricing model based on users is more useful to the customer.​
Price Comparison
Because different services charge based on different factors and at different usage tiers, it isn't possible to make a truly objective price comparison between the review candidates (which is a deliberate choice made by the marketing teams of the respective services, of course).
Normally, I don't spend this much time investigating pricing, but since these tools represent a significant investment, I'm making an exception.
Kissmetrics suggests that you should assume an average of 10 events per month, per user. Based on this, here's roughly how the prices stack up:​
The dots in the graph represents the actual price tiers (data from 11.11.2014). The trend lines are just there to help compare the prices, but you won't actually be charged anywhere outside the fixed pricing tiers (in most cases).
A problem with this comparison lies with the Kissmetrics suggestion that 10 events roughly equal 1 user: this simply won't be true in many cases. The average amount of events you have per user is going to vary drastically based on your business model. A quick glance at some data we sent from one of our sites to one of the services showed that we seemed to be sending around 20 events per user.
So, just as an example, here's what the graph looks like if we assume and average of 25 events per user (per month):​
Please take this comparison with a grain of salt. What it really shows, above all, is the reason why I prefer the "per user" pricing base, since it's much more predictable. With services that charge per event, you simply don't know how much they're going to cost you until you start using them.
Also consider that the prices and tiers are likely to change over time, for the different services.
Personal Impressions
In all of the above, I tried to be as objective as possible, so allow me a brief segment to voice some purely subjective viewpoints now.
Both Calq and Indicative are promising newcomers in this field and Indicative is the tool I was hoping we could use for our own business. Unfortunately, both of the solutions don't provide everything we need to get all the critical data from our business. In Calq, I was missing a way to group reports by different properties (to be fair, this kind of grouping isn't available in all the other tools, either). In Indicative, it's the missing revenue reports and missing user aliasing that kept us from using it.
Having said that, I think both of the tools have potential to grow into strong competitors.
Kissmetrics is feature rich, but for me, it's let down by the user interface. Kissmetrics feels a bit slow​ and it never made me feel like the answers I was looking for were right at my fingertips.
Mixpanel makes a very good impression. In fact, I'd give this a resounding recommendation if it weren't for the steep pricing. Mixpanel gives you easy access to a range of different reporting tools and they come with good tutorials as well. The tools are also fairly easy to use and report data quickly. This gives me that feeling I was missing in Kissmetrics: that there's always a way to get the answers I'm looking for and that I can find them quickly. What I'm missing from the tool is a dashboard where I can get an overview of the business KPIs.
Woopra have made a smart move with their AppConnect feature. Depending on your business needs and what other services and apps you use, Woopra could be the perfect tool for you, just based on AppConnect.​ I like Woopra's user interface and the detailed visitor tagging and segmentation tools that it provides, as well.
Trakio has very prohibitive pricing​, to the degree where this is the make-or-break point of the product. You have to ask yourself: what exactly do you need from your analytics tool? If Trakio is the only solution that can provide what you need (from this group) and the price difference to the other tools is warranted by the extra value you get, you can start evaluating it as an option.
Conclusion
There's no doubt that advanced customer analytics tools can bring massive value to your business. One of the big advantages all the tools in this roundup have is that they don't only have more of a people-focus than Google Analytics, they are also a lot easier to use.
Once you're passing all your important data points to the service (which requires coding - and I recommend you use Segment for it), it's easy even for the non-technical to get answers to important business questions.
​In the scope of this review, it's not possible for me to tell you "this tool is the best". The truth is that depending on exactly what your business model is, a different tool will be the right one.
In closing, I want to mention Avinash Kaushik's 10/90 rule: ​for every $10 you spend on an analytics tool, you should be spending $90 on a very smart person to make use of the tool. The same is true for smaller teams (who don't have those $90) and the amount of effort invested: unless you invest a lot of time to actually get useful data out of the tool, it won't do any good.
Do you have any questions or suggestions? Let me know by leaving a comment below!​
P.S.: if you're still looking to get the analytics basics sorted, before jumping into something advanced like this, check out my roundup of Google Analytics competitors.
Thanks for the comprehensive overview.
I am currently playing around with Mixpanel (they sent me a T-Shirt for doing that ;-). It quickly turned out that one should not start implementing complex analytic tools before he knows which questions he wants to have answered by the tool (this also defines what should be implemented and if a fancy tool is actually needed) and I agree for user/account-centric sites this is gold. Many business owners just want to know how many visitors they had in the past month, if numbers go up, its good else not so much and thats it. Analytics can provide so much more insights nowadays, it would be worth to create a post just about all the answers it can provide.
I didn’t know Indicative, will have a look at them.
Yes, a very important principle I use for analytics is the “ask the question first” approach. As in: you need to know what business related questions you want answers to before starting to look at different tools. I will actually be releasing a post about this soon, on the Thrive Themes blog.