No results found for your search
Ask GH: Best analytics for mobile apps?
I'm torn. I'm using Segment.IO but now I have to decide between GA's Mobile App Analytics, Mixpanel, Localytics and 100 other services.
Any feedback is appreciated.
I agree with @msarlitt generally in that you should break this down to the different stacks or categories.
I've already posted this in a couple other GH discussions but it seems relevant here too, so I'll throw this out there again. We received a lot of similar inquiries from customers, so we went ahead and compiled an eBook called The Ultimate SDK Guide for App Optimization. It's free and it will give you a wider view of the mobile app analytics tools out there that you should consider using.
You can download it for free here: https://www.appsee.com/ebooks/sdk-guide/?cpnid=701b0000000WHTw
I would also say that, from what I can see in my industry, the real breakthrough has come in the form of visual analytics that provide you qualitative data in addition to the quantitative. I'm biased because I work at Appsee Mobile App Analaytics which does just that, but it also makes a lot of sense. Appsee automatically digests the data for you and provides valuable visual insights.
Being able to watch a video recording of a mobile app user, or seeing an aggregated heatmap of where people are tapping, gives you a much more in-depth yet quick answer to many questions, as well as raising new ones. If you're interested, you can read more about the technology here: https://www.appsee.com/?cpnid=701b0000000WHTw
I actually had a similar conversation with @lukethomas. Here's what I recommended:
I think it's best to break a mobile analytics stack into 3 parts: Analytics, marketing automation, and attribution. Most tools out there specialize in one area, then dip their toes into one other.
We (postalpix.com) use Mixpanel for analytics, Mixpanel and Mailchimp for marketing automation, and MobileAppTracking.com for attribution. MAT sends attribution info to Mixpanel, which you can segment by in their reports.
Mixpanel is a pain to set up properly. If it's configured correctly though, I think it's the best bang for your buck in terms of a combined mobile analytics and marketing automation tool.
If I was starting from scratch, I'd go with:
Analytics: Mixpanel (primary), GA (backup)
Marketing Automation: Mixpanel, Mailchimp/Mandrill
Attribution: Appsflyer if you plan to run FB ads, MAT if not (you can't use MAT for FB ads as of April 2013)
Looks like Localytics dropped their pricing since I last checked them out, so they are an alternative to Mixpanel.
This is just a general recommendation that fits most apps, but it really depends on what your app does and your data needs.
What were the challenges you encountered implementing Mixpanel?
Mainly needing to define everything you want to track before you start receiving data. The concepts of events, event properties, people properties, and super properties and their relationships to each other were not intuitive to me. I still find it weird that you can't send notifications to users based on events they trigger, and instead need to add special people properties to store this information. Having said all that, I still think they're a great tool and happy we chose them. ;)
I agree with you in regards to the confusion between the different data sets of Mixpanel not communicating with each other.
There are 2 types of analytics software for mobile:
1. Unbiased ad tracking / Attribution [ MobileAppTracking, AppsFlyer, Fiksu ]
2. In-app analytics [ Google Analytics, Mixpanel, Localytics, Flurry ]
a) Attribution / Unbiased Analytics
If you plan on doing some paid acquisition for your mobile app, you will most likely need one of those analytics product in order to start a campaign on an Ad Network (Pay-per-install and performance-based campaigns). These services are necessary to run with Ad Networks, they act as middle-men for unbiased reporting on the campaign. They provide metrics such as the number of clicks and installs by ad network. These software can also post-back the attribution information of an installation to your server or main in-app analytics system. (Attribution: From which campaign the user was acquired through?). I recommend AppsFlyer which supports most AdNetworks, including Facebook. MobileAppTracking is not supporting Facebook.
b) In-app Analytics
This type of software is used to analyze how people are using your product. This is where the core of your tracking is done. There is a wide variety of products out there. I strongly recommend Mixpanel as main tracking tool to understand how users are using your product. Mixpanel is relatively easy to set up and their interface is very simple, making it easy to query your data. You can set up Mixpanel to get the attribution information from your attribution solution and segment the usage based on the user acquisition campaign.
I strongly suggest defining your business-level questions and your needs prior to choosing a solution. Most of these companies can provide you with a demo account with fake data so you can play with their interface. They are also usually very happy to do a demo of their product and answer your question.
Would you recommend Amplitude over Mixpanel?
CEO of Amplitude here. Mixpanel's great to start up with but can get pricey at scale. Amplitude is also good at telling you exactly why users are dropping out of your funnel. Shoot me an email at firstname.lastname@example.org and I'd be happy to get you set up.
I'd recommend Mobile App Tracking/Adjust for paid channel tracking, (both Mixpanel and Amplitude integrate with both of them). I've seen AppsFlyer mentioned a few times before but I don't have first hand experience with how good their product is.
That fact that you're using Segment.io already makes you way ahead of the game in terms of your thinking about tracking!
I don't know Amplitude so I can't judge. Based on your goals, I would consider opting for Mixpanel + AppsFlyer for attribution. Mixpanel has Funnels / Segmentation / Retention reports.
I also heard good things about Localytics and that it can provide attribution out-of-the-box. This can be another interesting option but I never tested it.
I have two goals:
1. Attributing users to the correct traffic source so I'll know which paid channel yields the most ROI.
2. Figure out which part of the funnel needs improvement so I can iterate. i.e: Activation/retention/referral?
Some companies combined these to so that you don't have to use different solutions. For instance www.adjust.com has in-app and attribution analytics together (incl. cohort analysis and other tools).
I've been 100% focused on mobile for the last two years since I switched careers from finance to marketing. I love it. Thus I will chime in here even though I'm late to the party.
Note: the remarks below are related to marketing mobile apps as opposed to mobile websites. Mobile web is less complicated.
I agree with a few other posters that there is no single solution that has the ability to effectively manage the full mobile user cycle. In my view you need:
1.) A Click-Stream Analytics tools (e.g. Google Analytics, Omniture). No in-depth explanation necessary here. These tools will allow you to measure top-of-the-funnel data like how many users are in the app, where the users coming from, how much time are they spending in the app, how many screens are they viewing, what devices are they using, what are the language settings on their devices, where are they entering the app, where are they exiting the app, what search terms are they using to find the app. On the e-commerce side you can setup goals and to do funnel marketing, but for some reason I don't know anyone that uses GA for mobile post-install event analysis. My GA is not configured for this, but just from playing around with it, I suspect that a more technical person could get 100x more out of GA than I can. Perhaps the need for more advance technical skills to fully exploit GA is a reason why premium solutions proliferate despite GA being a fully-functional free tool.
2.) A 3rd party install tracking tool (e.g. Flurry, Adjust, AD-X, Kochava). Despite popular opinion that these are analytics tools, I believe their core value proposition is not analytics - it's that they allow user acquirers and user sellers to have an objective 3rd party tell each side how many installs were delivered or in-app events took place. They also eliminate the need for acquirers to setup a server-2-server integration (which would require a new release of the app) with every media partner you work with to get an install count that both sides accept. In my case, without Adjust, I would need something like 15-20 S2S integrations. That would seriously increase the weight of the app and be monumental waste of resources considering how many media partners fail to deliver.
3.) A post-install analysis, testing, and CRM suite (Mixpanel, KISS Metrics, SWRVE). These are tools that combine a bunch of disparate functionality into a single tool. None of them does everything, but they are significantly reducing the need to work with multiple vendors. For example, not too long ago, if you wanted to do push notifications you needed to call up Capptain or Urban Airship. If you wanted smart links you needed to call up 360 Dialog. If you wanted to do A/B testing you called up Apptimize. If you wanted to do mobile deep linking, you needed (and still pretty much have to) develop your own standard. Platforms like Mixpanel and SWRVE are consolidating some of these services into a single suite. I've been fighting internally to get one of them integrated since late 2013 so I can drop some (or all) of those tools I mentioned above. The fight continues. Nevertheless, I've been following the evolution of these tools and I believe they are distinct from tools like Google Analytics and 3rd party tracking tools like AD-X.
To wrap it up, the tools in each category have some overlap. However, they each specialize in something that makes them distinct enough to warrant using them all despite some overlap. If you used only one tool, I think you would be at disadvantage. How much of a disadvantage depends on which tool you're lacking. You can probably get away without using GA, but without 3rd party tracking, you basically can't do paid acquisition on a CPI basis unless you want to bill off your suppliers install counts. The analytics and testing suites are going to arm you with the information to really know what the hell is going (e.g. funnels, cohort analysis, in-app surveys) on and do something about it (A/B testing and push notifications). You can go to individual vendors to get each service piecemeal, but then you end up with 10 different SDKs in your app, plus a data consolidation headache, and the extra admin.
Every app/mobile website is in a different situation. So I'd advise figuring out what you're optimizing for, what key questions, you need to answer, and what can you afford. Take that info and do some research and make the best choice for your specific situation - not the solution most popular solution, with most bells and whistles and a nice shiny interface with a bunch of pretty (but useless) dashboards.
Flurry and GA provide traditional analytics solutions. They measure key metrics such as new users, what OS they're using, active users, geographic breakdown, etc. and they do it quite well.
But, if you want to dive into the user behavior and user experience, their solution won't tell the the full story since they emphasize the numbers instead of the reasons. So, you'll understand the 'what' and not the 'why'. For example, you may detect that you have a low user retention, but you won't understand why users are not returning to use your app.
By applying the use of visual mobile analytics tools, you will be able to see the WHY, by diving deep into the user experience & behavior, and essentially seeing the app through users' eyes.
I work at Appsee - http://www.appsee.com - where we provide a visual mobile analytics solution. Appsee helps app developers understand how their users interact with their apps by providing an extensive user experience & user behavior analytics platform and analytics reports. Among the unique key features included in our solution are user recordings, touch heatmpas, UI analysis, real-time in-app analytics and more.
Hope this helps! Happy to answer any additional questions you may have.
Check out flurry.io, I've had clients which strongly prefer that to mixpanel, but this was ~1 year ago and it seems that space is moving at breakneck speed. I've only used heap for the web, but keep on meaning to try it on a mobile app as well.
In case you land up with GA, Google is about to do a course on Fundamentals of Mobile App Analytics: https://analyticsacademy.withgoogle.com/course04/preview
Has anyone tried Amazon mobile analytics? -> http://aws.amazon.com/mobileanalytics/
Didn't even know that existed! Interesting that they're getting into the space.
Yeah, it's very basic at the moment but looking forward how this will evolve!
There are a bunch of players in this industry that bring a host of dynamic solutions to the table.
However, there are very few intelligent mobile marketing platforms in the market that endeavour to bridge the gap between app analytics and effective user engagement. This is precisely what CleverTap excels at.
Using granular, actionable user-level insights, you can gain an in-depth understanding of user personas and in-app behaviors. Based on these you can easily automate your user segmentation driven by specific in-app actions/non-actions taken by users. Accordingly, you can craft incredible omnichannel engagement campaigns - using channels such as push notifications, in-app messages, emails, and SMS - to engage with the right users in a personalized and contextualized manner. Not only does this data-driven approach help you engage with users at every stage of the app user lifecycle, but also maximize monetization and user retention in the long-run.
Learn just how you can leverage clever app analytics and unleash the true potential of your app business here - https://clevertap.com/live-product-demo/
Over 4,000 customers across 100 countries, including leading brands such as Sony, Star, Domino's Pizza, Vodafone, Go-Jek, and BookMyShow, harness the dynamic capabilities of CleverTap's ML-powered platform to deliver incredibly differentiated customer experiences.
Good question. I don't have any recommendations.
I'm no expert when it comes to mobile analytics but I continuously hear that Mixpanel is the best from friends I've talked to in the mobile industry. Definitely do your research, hopefully someone can provide a more in-depth answer.
I run an analytics company called GoSquared, but we don't currently specialise in mobile analytics. I've heard a lot of good things about Countly: https://count.ly/ they appear to offer an open source version too if you're ok with hosting things yourself.
There are a few options in the market, but to evaluate them you need to define how are you going to use the solution. For me a mobile analytics platform should give app publishers a full view on all that is happening in an app, from all available users profile data, sessions, events / actions performed and all info regarding the devices used. This should be the basis for any analysis and for all the actions a publisher wants to make after.
Most of the current options on the market lack some of this full view aspect, either because they are not covering enough data on the users, or they are not offering a full understanding on sessions data.
This forces app publishers to use more than one platform and consolidate the data in an external platform, which isn't practical and tends to lead to quality issues.
I’m the Lead Business Developer of Liquid, and these were some of key reasons why we have created our platform, so an app publisher can have a one stop shop for analytics.
Other key reason on why we created Liquid was due to our belief that analytics needs to lead to actions. So besides the typical communication tools (push notifications, segmented lists), we add personalization capabilities to our platform, allowing app publishers to change an app on the fly without changing the code.
Check our tool at lqd.io and feel free to contact me if you'd want to see a demo.
I suggest you to take a look at FlightRecorder.io It's free.
With FlightRecorder’s all-in-one SDK you can dive into deep data, track users' behavior and get insights to improve your product with unique features such as Screen Recording, Crash Reporting, Cohort Analysis and much more..
Have a look at http://www.FlightRecorder.io
I suggest you to watch the short video about mobile app screen recording with pinches, gestures and swipes:
FlightRecorder's key features are;
Screen Recording: Next Step On Understanding User Behaviour FlightRecorder records screen captures of sessions that show you each and every action your users take drilldown by events and watch the videos that matter.
Crash Reports: Not only the crashed line.Watch the crashed sessions, understand the crash reason. FlightRecorder gives you whole session not only the line of a crash.
Understand The Big Picture.A user centric vision & action capability to the sessions. You just need to call FlightRecorder SDK's User ID method, that's all. Use Segmentation and get results based on users.
Your app at a glance. Immediately see the reports that matter to you and understand the big picture. We've a beautiful dashboard, not only numbers and graphs. We summarize your data and make it meaningful.
All-In-One SDK! Integrate FlightRecorder SDK to your mobile application and get solutions for UX, Analytics & Segmentation, Crash Reports, Push Notifications and Beta Distribution.
If your mobile application has more than one SDK integrated, it means that your app is less stable. FlightRecorder is one-stop solution for all your needs and we release updates frequently.
Please, feel free contact me if you have any questions.
Use the feedback box below if you have a question, comment or general feedback.
Your feedback has been sent.
Sweet! The link has been copied to your clip boardy board!
Flash isn't supported. Please copy the link manually.