Event data analytics help a great deal in gaining potential insights regarding customers’ actions within your product. The data collected from users’ actions can help improve performance of an app or a website. As simple as it sounds, it’s quite tricky. One needs to have the right tools to analyze event data. These tools help track the behaviors and actions of customers within the products so that the researchers can monitor their activity, test new ideas, and gather excellent insights. But with so many tools available to do this, choosing the right ones is quite tough. Let’s take a look at some of the tools out there.
Looker is a great tool for organize modern web analytics, one of the primary reasons being its rapid data exploration and its ease of data modeling. Looker’s simple data platform makes it easy for everyone to understand and find their event analytics data.
You can make use of modern Looker API to push your data into places that matter the most. You can also format it in a way you like. Looker dashboards can be easily accessed through mobile devices; they even allow ad hoc exploration and filtering. Once you’re done with integration, Looker automatically provides you a UI that you can analyze your data without any need for writing complex SQL queries.
Having said that, there are considerable number of pitfalls. Looker doesn’t play well with SQL and it’s not transparent while generating SQL queries. The main problem with Looker is their query language, LookML. The problem is that while it’s advanced and flexible, you need to learn it in order to be able to use LookML. Since they created their language from scratch, learning LookML is not that easy and in fact, most of the time Looker employees map your data and write your LookML specifications but it’s usually quite expensive compared to using the alternatives.
Tableau’s data virtualization is outstanding compared to so many other conventional tools available in the market currently. Tableau is one of the first companies in the world to give its users the provision to conduct complex data visualization in the most effective way possible — a simple yet intuitive drag and drop manner. You need not be skilled in IT or require any assistance from any IT people to highlight sections and display the results in charts. If you are looking forward to have high-powered data visualization, then Tableau is the tool for you. Also; if you need to be able to customize your charts as wish without any limit, Tableau should definitely be one of the tools that you try.
There are some issues with security. Tableau is not well-equipped to provide centralized data level security. Untrained users (those who are not well-versed in SQL, for example) can use Tableau, but it’s hard to get the best out of it without the help of IT department. It’s simply not easy to learn the basic functionalities. Moreover, it lacks essential functionalities that a business intelligence tool should have, like building static layouts, data tables, scale reporting, etc. Also, it’s very difficult to share the results. The notification functionality is ridiculously simple and only an admin can configure it, and not the end-users.
Periscope brings all the important and relevant parameters at one single place, thereby giving you an overall view of operations and results. This helps you make better decisions regarding your event data operations. Developed by a team of hackers, Periscope is one of the most secure tools for analyzing your event data. It’s also easy to learn, has a simple interface, and plays well with SQL. If you’re ok with dealing only SQL interface and if the simplicity is important for you, Periscope might worth trying out.
Periscope is not rich enough and requires you to give your database access to third party servers. It is quite difficult to manage a large number of dashboards. Besides, it doesn’t work well for data exploration.
One of the major disadvantages of Periscope Data is that you are required to have tremendous experience in programming. Also, the UI can be a tad tricky to use.
4. Mode Analytics
Mode Analytics is one of the most collaborative analytics platform there is currently. This is basically everything your team needs to improve your business. Mode Analytics provides your team with the right tools, which speak your language. You can start by exploring data using Python or SQL and then send important reports to your team to empower them. There is no need to learn any other tedious, proprietary data modeling language.
Mode Analytics is very helpful for data scientists. It allows users to create advanced visualizations and reports. It’s a very flexible tool and supports Python and R.
There is no separation of Development and Production environments, thus making it quite difficult to collaborate with others on dashboards regarding reports. The main reason for this that the changes get overwritten while working on different things at the same time. Besides, you are compelled to mention all the extra calculations you want in a single query. This leads to a lot of confusion and a lot of redundant queries.
Rakam focuses on event data use-cases from the day 1. Instead of writing complex SQL queries in order to be able analyze your user behaviour, it provides Mixpanel-like interface for your event-data on top of your data-warehouse. The integration is similar to the previous solutions that we mentioned, you enter your database credentials and connect to your event-data. The next phase is similar to Looker; you map your event data to our system. However; unlike Looker you don’t need to know any custom-built language similar to LookML, you use the UI to map your tables, special columns (event timestamp, user id etc.) and start using the funnel, retention and segmentation UI.
We also have SQL interface that allows you to run interactive SQL queries and visualize the data. The charting capabilities are not rich as Tableau and Periscope but we allow you to parametrize your queries, build custom reports with chart and table, make them interactive with report linking and variable features. The way it works is similar to Google Analytics but in a custom way; you build your own reports similar to the pre-defined Google Analytics reports and save them to be able to share within your team. We cache your data so that it becomes fast as GA.
Also, we provide advance privacy features such as table level, column level and row level access. We’re aware that the product data is important and not all of your employees should be able to all of your customer data so you give right access to your teams in your organization. Additionally; enterprise companies usually have hundreds of event types so we have a feature called taxonomy that allows you to add labels, descriptions and categories to your tables. It’s basically a book for your data in your data-warehouse.
Rakam is not an all-in-one solution, if you want to have a generic product that integrates with tens of databases Rakam might not be a good alternative. Also it uses its own SQL syntax for security so there might be some sugar-syntax that is not supported by Rakam but supported in your data-warehouse.
Also the charting capabilities are limited compared to Tableau, if you need advanced charts you might need to wait for Rakam team to implement it. — We’re always open for suggestions though!
We have evaluated the most common analytics tools in the market and have seen that most of them tries to be generic in order to be able to cover all the use-cases. On the other hand, it opens up a new hole in the market because you need to spend time to adopt the product for your use-case. For the last couple of years, the companies shifted their mindset to be more data-driven so they want to collect all of their company data including user behaviour data to their data-warehouse. The cloud providers also make that easy with managed data-warehouse solutions such as Redshift, BigQuery and Snowflake and there are also open-source solutions such as Presto, Impala and Apache Hive. However there were no product that plays well with these data-warehouse solutions and also focuses on product analytics — that’s why we created Rakam.