Table of Contents
Introduction to BI Tools
In recent months, there’s been a wave of acquisitions and fundings in the business intelligence (“BI tools”) and analytics space:
- SiSense acquired Periscope Data on 5/14/19 for an undisclosed amount.
- Qlik acquired Attunity on 2/21/19 for $560M
- Google acquired Looker on 6/6/19 for $2.6B.
- Salesforce acquired Tableau on 6/10/19 for $15.7B
- Metabase announced a $8M Series A on 4/23/19
- Mode announced a $23M Series C on 2/19/19
- The team behind Superset is rumored to have raised $18M in Q1 2019
That’s a lot of movement in a short time for a segment of enterprise software that’s been somewhat static in the past 25 years.
A major driver in the uptick of M&A and venture activity are cloud warehouses like Amazon Redshift, Google BigQuery and Snowflake. They have brought down the cost and complexity to build a data platform, in a shift away from Hadoop, with BI tools as the catalyst to make data exploration and visualization available to a much wider audience.
We believe that the visualizations and charts are a commodity. What seems to matter to buyers of analytics is the “eye-candy” of the dashboards. But much of the hard work to build a data platform happens underneath the visualization layer, e.g. data acquisition and ingestion, ETL pipelines and data discovery layers.
There’s a bit of confusion on how these tools work and their use cases. So in this post, we’re helping you find the answers, by looking at the current BI tool landscape and their differentiators.
Fundamental Separation of BI tools
BI tools facilitate exploration and visualization of data, and for the purpose of this post, we’re only including tools that support the major data warehouse (Amazon Redshift, Snowflake, Google BigQuery, Azure SQL Data Warehouse.)
Let’s start by dividing BI tools into three simple categories:
Open source – “Modern”: The source code is available for download, with a vibrant developer community. There’s a commercial entity behind the open source project, and the business model consists of providing support and enterprise add-ons.
Commercial – “Legacy”. Tools in this category offer an installable version of their product or a SaaS subscription. The key point though is that they ingest data into their platform before analysis happens, i.e. they schedule and run an extract of the full data set from your cloud warehouse on a regular basis.
Commercial – “Modern”. Tools in this category query the data “where it lives”, i.e. in your data warehouse. Unlike the legacy tools, they don’t run an extract of your data set and query data on a scheduled or ad-hoc basis in your warehouse.
- Periscope Data
We believe the distinction between “legacy” and “modern” is key.
In an old world of on-premise warehouses like Teradata and Oracle, the cost of running queries and experiments on your warehouse was expensive. Extracting the data into the BI tool made sense, to drive down the cost and allow users to be flexible in transforming data and exploring new types of queries.
But cloud warehouses have become so performant and scalable that running transformations and queries in-database is much more efficient. With the separation of storage and compute, there are a few reasons why you would run queries elsewhere. And the benefit is that employees explore one common & current dataset.
Let’s go through each of these categories and explore their key representatives in more detail.
Open-source BI tools (Metabase, Apache Superset, Redash)
Metabase (18169 commits, 169 contributors)
Metabase is an open-source business intelligence tool designed for non-technical users to provide data insights and visualizations. One of the main advantages of Metabase is its excellent visualizations and the fact that you can create dashboards almost without any prior experience.
You can create dashboards and visualizations using a simple question creator with no programming. If the question creator is not enough for you, you can use basic SQL to create more complex queries.
Another useful feature is “Pulses”. With Pulses, you can send updates on specific queries (“questions”) to Slack channel or email.
Metabase supports multiple relational and non-relational databases as data sources. Among them are the following: Amazon Redshift, Snowflake, BigQuery, PostgreSQL, SQLite, MySQL, Druid, MongoDB, and more. The Detailed list you can see in the documentation.
Apache Superset (4213 commits, 380 contributors)
Superset is a modern BI tool with a simple interface allowing to create and share dashboards. Superset comes with a lot of different, rich visualization types included. To create queries and dashboards, you need to know basic SQL.
Superset started as a project at Airbnb, to as a fully customizable application to visualize and explore massive amounts of data in a fast and intuitive way. Today, Superset is one of the leading open source BI solutions.
Superset provides an advanced security management system, with flexibleadjustment for levels of product, feature and data set access.. Superset also has integrations with major authentication backends.. Another attractive Superset feature is the SQL/IDE editor with interactive querying.
Superset supports multiple data source types through SQLAlchemy, a Python ORM that is compatible with the most common databases (PostgreSQL, Amazon Redshift, MySQL, Oracle, SQLite, Microsoft SQL Server).
Redash (6538 commits, 251 contributors)
Redash is a simple and powerful BI tool, suitable for advanced users since you need to know SQL to work with it. Redash provides an interactive query editor for sharing both the dataset and the query that generated it.
Redash allows you to use query results as data sources to join different databases and make advanced reports and dashboards. With Redash you can set up reminders and automatic notifications (Alerts) to notify a group of people or an individual user when a specific query reaches a preset value. The biggest advantage of Redash is a very vast array of different visualizations that users can create.
Redash provides the most impressive list of data sources from all open source BI tools in our review. You can use almost any data source you can imagine, from CSV to Elasticsearch, and of course the major data warehouses like Amazon Redshift, Snowflake and BigQuery. You can explore the full list here.
“Legacy” Commercial BI Tools (Tableau, Dono, Qlik, PowerBI)
Tableau is one of the leading BI solutions offering a wide range of visualization capabilities, extensive documentation, and a simple interface. Tableau is friendly for beginners, the needed functions are most often achieved in no more than 2 clicks, filters are easy to find, and all operations are clearly documented.
A key differentiator for Tableau is data blending, a method for combining data from numerous databases and sources. Tableau also allows multiple users to work on a report in real-time simultaneously.
Tableau supports a wide range of data sources. These can be files (CSV, JSON, MS Excel, etc.), relational and non-relational databases (PostgreSQL, MySQL, SQL Server, MongoDB, etc.) and cloud warehouse (Amazon Redshift, Oracle Cloud, Google BigQuery, Microsoft Azure).
Tableau provides several pricing plans for different needs:
A free trial is also available.
- Tableau Creator – $70 USD/user/month
- Tableau Explorer – $35 USD/user/month billed annually
- Tableau Viewer – $12 USD/user/month billed annually.
Domo is a powerful cloud-based BI tool with solid data visualization capabilities that integrates with multiple data sources, including spreadsheets, databases, social networks, and almost any other existing cloud or local software solutions.
Domo provides a wide range of connectors (over 100 connectors right out of the box) and impressive sharing features. But its user interface is not intuitive, and this tool is much less friendly to newcomers than some other competitors.
There are several pricing options:
- Up to 5 users
- 4 concurrent data connections
- 5M data storage size (rows)
- Hourly data refresh
Standard – $83/user per month (billed annually).
Professional – $160/user per month (billed annually).
Enterprise – $190/user per month (billed annually).
Qlik Sense is a powerful BI solution. Its Associative Engine automatically finds relationships inside the data. This simplifies the study of data and allows easy combination of any data sources, no matter how large or complex they are, into a single view.
Qlik Sense lets you explore data at any level of detail you need, explore the shape of the data and pinpoint outliers. AI-powered insight suggestions will not allow any meaningful data to escape from you. Qlik Sense also provides an opportunity for you to dive into advanced analytics calculations from R and Python using simple clicks.
Qlik Insight Bot adds AI-powered conversational analytics capabilities to Qlik Sense. With it, you can receive comprehensive reports, ask questions and discover insights using natural language.
Cloud Basic – Free
- Fully interactive apps
- Sharing with up to 5 users
- Access to Qlik DataMarket Free.
Cloud Business – $15 per user/month (billed annually).
Power BI is a business intelligence tool from Microsoft with all the benefits that Microsoft services ecosystem provides. PowerBI connects with the company’s main products, such as MS Excel, Azure Cloud Service, and SQL Server. Power BI is marked as a leader in Gartner’s Magic Quadrant for Analytics and Business Intelligence Platforms.
Power BI interface is simple and clear, it will be familiar to Windows users. Visualizations are created by drag-and-drop. All you need to create any graphics – click on the desired item and drag it to an empty space in the report.
Power BI supports plenty of ways to incorporate or import your data (streaming data, cloud services, excel spreadsheets and third-party connections) and provides a simple API for integration with your applications.
- Power BI Pro $9,99 Monthly price per user
- Power BI Premium $4,995 Monthly price per dedicated cloud compute and storage resource.
“Modern” Commercial BI Tools (Looker, Chartio, Mode Analytics, Periscope Data)
Looker is a BI tool of a new generation. It is fundamentally different from other solutions. It offers LookML, its own data modeling language, which simplifies the descriptions of dimensions, calculations (measures), and data relationships in a database. Using LookML, database queries are more descriptive and subject to fewer errors.
Once you have created a data model with LookML describing in which dimensions you want to group the data and how you want to evaluate it, you can open access to this model to other users who can use it to create their own queries, reports, and dashboards through a simple interface. Thus, the work is divided and optimized, data analysts get a convenient tool for modeling and business users are freed from the need to create complex SQL queries.
The need to learn LookML and data modeling before any data visualizations is made easy through the educational materials for learning this language provided by Looker. (Learn more about Looker and Amazon Redshift Dashboards.)
Looker introduces innovation in how to collaborate on BI projects. Each project or data analyses can be saved as a Git project. In this way, you can use the full power of modern version control systems when collaborating on data.
Looker supports about 45 different data sources. The full list you can see on this page.
Looker’s price ranges from $3,000 to $5,000 per month for 10 users. Additional users are $50 per month.
Looker is an official Integrate.io partner
With Integrate.io, Looker users get all the tools they need to optimize their queries running on Amazon Redshift. For more info on how this feature works, visit intermix.io/looker.
Chartio is a great BI tool for advanced users with excellent data blending, data transformation, data sharing capabilities and robust security options.
Chartio provides two modes of work: drag-and-drop interactive mode or SQL mode. While its UI is quite powerful, it is not so easy for non-experienced users as UI of some other competitors. Chartio has built a visual version of SQL, which enables you to explore, transform and visualize data on the fly through a drag-and-drop interface.
Chartio’s visual SQL allows you to quickly combine and compare data from different sources. For example, you can get data from Google Analytics and then compare it with Campaign history from Mailchimp. All this can be done in a few clicks and you will be helped by extensive documentation.
Chartio supports about a hundred of different data sources including relational and non-relational databases and a bunch of SaaS partners. The full list you can see on this page.
- Startup $400 month
- Growth $900 month
- Enterprise is only available when you request a quote.
Chartio is an official intermix.io partner
With intermix.io, Chartio users get all the tools they need to optimize their queries running on Amazon Redshift. For more info on how this feature works, visit integrate.io/chartio.
Mode is a light-weight framework for business intelligence. You can run it directly in the browser. The tool allows running complex queries using SQL, Python, or R. The query editor is very intuitive and simple even for novices. For Python and R, you can perform data analysis in the Mode Notebooks. More than 60 Python and R libraries for data wrangling and analysis are available out of the box. If needed, you can install other public packages. Also, Mode Analytics supports the creation of the reports, which you can customize and share with different stakeholders.
The target group of use cases where Mode Analytics is extremely good is the fast and light data analysis. In-browser running, drag-and-drop capabilities for reporting, simple interface, a large number of built-in features, integration with SQL and such popular programming languages for data analytics as Python and R makes Mode Analytics a perfect choice for fast prototyping and ideas sharing.
Mode Analytics has two pricing options:
- Mode Studio is free
- Mode Business is offered for free only during the first 14 days of trial period. The full price of this tool is not available on the official website.
Mode is an official intermix.io partner
With intermix.io, Mode users get all the tools they need to optimize their queries running on Amazon Redshift.
Periscope Data is one of the new tools for BI. It provides support for different data source connection and data preparation using SQL, Python, and R. It also allows to perform predictive analytics and unstructured data analysis (for example, during the natural language processing).
Periscope Data has a lot of built-in visualizations. Among the provided templates there are even such exotic charts as radar chart, cohort grid, gauge chart, etc. Charting with maps is also available. In addition, you can use Python and R libraries to prepare your own graphs. There is a lot of space for customization anyway.
As many users say, there is no learning curve for Periscope Data at all. It is extremely simple. Periscope data supports easy-to-understand drag-and-drop interface. If you need to find something, usually it is no farther than a few clicks from your current location in the app. Sharing capabilities and collaboration are also great.
Enterprise pricing plans for periscope data are not available at the official website. However, there is a free trial package available for the customers to get started with the software. Later, they can consult the sales team for their respective pricing packages.
Periscope Data is an official intermix.io partner
With intermix.io, Periscope Data users get all the tools they need to optimize their queries running on Amazon Redshift. For more info on how this feature works, visit intermix.io/periscope.
BI Tool Popularity
[You can embed the code in the article. This will allow to continuously update the graphs. If only including screenshots, we will state the date when they were taken.]
Open Source BI Tool Comparison
On the chart below, you can see the searching trends for 3 open source BI tools described in this article: Metabase, Apache Superset, and Redash. At the time of writing, the most popular tool is Metabase which is approximately 2 times more popular than its closest pursuer, Redash. Also, you can notice that while the popularity of Apache Superset and Redash remains on a relatively stable level, the popularity of Metabase started to grow in the last couple of months (at the time of writing).
Commercial BI tools – “Legacy” Comparison
As was mentioned earlier, “legacy” commercial BI tools include Tableau, Domo, Qlik, and PowerBI. Their search popularity on Google Trends is depicted in the graph below. Tableau is the undoubted leader here. Its popularity is substantially higher than all other instruments. PowerBI ranks second. Domo is a little bit more popular than Qlik, but the level is relatively the same. There is no vivid trend for any platform, with unexpected peaks due to one-time events (i.e. the Tableau acquisition by Salesforce). The stable (side) nature of trend for all of these tools emphasizes their matureness in BI world one more time.
“Modern” Commercial BI tools comparison
The leader in the set of “Modern” BI tools is even more clear than in the set of “Legacy” tools. Looker is like a giant among them. To better understand the gap between these instruments remember that Google Trends displays the relative popularity. Now, Looker at the time of writing has 37 points, while Chartio, Mode, and Periscope Data have 2, 3, and 1 point respectively. Looker had a huge peak in popularity due to its acquisition by Google. The trend is not clear here for any of the instruments. They all remain at the same level.
In this article, we highlighted some of the most popular business intelligence tools. We split them into open source and commercial proposals. In turn, commercial tools were conditionally split into “legacy” and “modern”.
“Legacy” BI tools are probably the most well-known. They are admitted as classic platforms and every BI analyst should know at least some of them.
“Modern” BI tools have emerged relatively recently. These tools provide some new value for data users. In many aspects, they are very good and can be considered as worthy competitors to the “legacy giants”.
We briefly described features of each of the instruments as well as tried to measure their popularity and existing trends. We can say that regardless of the fact that there are a lot of BI tools nowadays, there is no “swiss knife”. In each case, a person or organization should consider their use case, estimate what and how they need, which resources are available and then decide which instrument will be the optimal choice.