Looker for Embedded Analytics: Worth it in 2024?

March 20, 2024

Mile Zivkovic

Looker is a powerful business intelligence tool. But how does it perform in an embedded environment? Let’s find out in this detailed review.

Looker is one of the most capable business intelligence apps out there, commonly used in enterprise companies. When it comes to data modeling, exploration, analysis and visualization, Looker rubs shoulders with industry giants like Power BI and Tableau. But how does this BI tool hold up in regards to embedded analytics?

Today, we take a look at Looker - pun intended - and how it works for embedded analytics and dashboards within websites and web applications.

What is Looker?

Looker is a cloud-based business intelligence tool that businesses use to model, explore, analyze, visualize, and interpret their data. It is most frequently used as an internal BI reporting tool to empower business users with actionable insights for better decision-making.

Looker has many use cases, and one of them is embedding Looker dashboards into custom applications and websites to give end-users real-time access to their data. This is our primary topic of discussion today.

But before we move on, there is one key piece of information to be mentioned. Looker was founded in 2012 and in 2019, it was purchased by Google, still retaining the name Looker. In 2023, Google renamed its Google Data Studio to Looker Studio. Long story short, Looker and Looker Studio are two different products, and we’re talking about the former today.

Looker for embedded analytics - the setup

Looker is built on top of their proprietary semantic model, called LookML. It’s incredibly powerful and allows for a vast range of customizations (through SDK and otherwise), automation, and workflows within the app. You can go around it in some cases by using SQL-based data manipulation but the end result is the same - you need working knowledge of LookML to be proficient in Looker.

looker dashboard

For many data teams, this means a very steep learning curve or hiring an extra set of hands with previous Looker experience. Once you’re past the initial hurdle, you can get on with building dashboards, and this is where Looker shines.

You get a wide range of visualization types: bar charts, pie charts, scatter plots, heatmaps, histograms, waterfall charts, and many, many others. It looks sleek and modern and your dashboard won’t put you to shame in front of your manager or the CEO.

The downside is that there is so much to choose from and if you’re new to data visualization, you’ll struggle with choosing the right visualization type for your key metrics.

Apart from that, Looker doesn’t use the same drag-and-drop interface you might be familiar with in most BI tools. Instead, you’ll need to write queries to create charts. That means that even for the seemingly easy act of creating charts, you’ll need a data scientist or developer.

On a related note, you can customize your dashboard elements extensively, but customizing them within a finished app or website is an entirely different thing. Since it’s an iframe, you don’t have much control over what happens within the embedded code in an app or HTML page. As a result, you’ll often get stuck with the Looker default branding and colors instead of your business’s primary colors and design.

The embedding process

It’s super easy to embed and configure a finished Looker dashboard. There are three main ways you can embed a dashboard from Looker:

  • Private embedding - you can embed the dashboard internally for your team, e.g. within your own app or website
  • Public embedding - anyone with a link can view a dashboard without having to log into Looker
  • Custom applications - for cases when you build a custom application using the Looker API

Besides embedding, you can also choose for white-labeling - creating a fully rebranded Looker environment for your organization.

Although the cases are vastly different, the embedding process is the same. You grab a line of code and copy and paste it to the end destination as an iframe. The only thing you have to do is allow the URL of the final destination in the Looker interface, as it blocks external requests by default.

For private embedding, you can set access using SSO embedding and authentication, so that when someone wants to view a dashboard, they don’t need to log in separately to Looker, but they will get access to the right data automatically.

On the other hand, for public dashboards, Looker can estimate the number of monthly visits you get and based on that, you purchase the number of licenses.

looker review on reddit

If your dashboard is publicly available on your website, this can lead to massive charges for viewer licenses later on.

The upside is that Looker is based on Google Cloud, which combined with row-level security (RLS) means that your data is protected and safe.

The integrations

One area where Looker excels is connectivity. Out of the box, you have support for the following:

  • Google BigQuery
  • Snowflake
  • Amazon AWS Redshift
  • A variety of SQL dialects
  • And many other popular data sources and other tools

You can connect with multiple databases at once, so your data analytics operations don’t get locked in with a single vendor.

Embedded analytics pricing

If you’ve ever looked at Looker pricing for their main BI tool, you know that this analytics solution is pretty secretive about the cost of their plans. Pricing is not publicly available and you have to get in touch with sales to figure out how much Looker costs.

The same applies to the embedded analytics platform, but with a little bit of searching, you can find some relevant information.

For starters, the pricing for embedded can be broken down into three parts:

  • The Looker platform (administrative costs, data integrations, modeling, data analysis)
  • The user licenses (how many people edit and create dashboards and how many end-users view them)
  • The management costs

If you take a look around, you’ll find the cost of your typical setup pretty easily. Online sources state that they typically pay $5,000 and up per month for Looker with the embedded functionality.

On top of that, you pay for individual users who can edit and view your dashboard. For true self-service analytics, you want to give edit access to end-users. But if you just want viewers, prepare to pay $400 per year for every person who views your dashboard. If you have lots of end-users, this can result in jaw-dropping annual fees.

So, the good news is that you can, with some accuracy, predict your monthly invoice based on how many customers you have. The bad news is that it’s wildly expensive and that the only way to offset the cost is to pass it on to those end-users and charge extra for the embedded dashboards.

The end-user experience

The end-goal of using Looker for embedded analytics is to provide your app users with data insights. And unfortunately, this is not where Looker shines.

looker on reddit

The key problem? The way that the Looker API is set up and the way the iframe embedding works. Looker was primarily built for internal dashboard reporting, where lightning-fast dashboards aren’t critical. Your internal team members can probably live with a few seconds of extra wait time when they load up a sales report.

But the end users of a SaaS app they are paying for expect their data visualizations to load with one click or tap, and Looker fails to deliver. There are ways to reduce the load times, such as using faster types of data sources like a better data warehouse. 

The ultimate problem is that your web or mobile app can load super quickly and it won’t matter. The dashboard loads separately in an iframe and it loads on its own terms.

If the end users can move past that, they can get pretty solid data experiences. Bear in mind that if they are viewers, they are going to have an easy time viewing and analyzing their data. But if you give them editor access through whitelabeling, they may be overwhelmed by the selection of visualizations and the complexity to create them.

Who is Looker for?

Looker is made for businesses with complex data modeling, analysis and visualization needs. The companies that should use Looker embedded are those that:

  • Have in-house expertise and previous experience with Looker/LookML
  • Have a high product cost that can justify the end-user viewer license fees
  • Have users that don’t mind the wait when Looker is fetching their data through an iframe

The bottom line is that Looker is perfectly fine for business intelligence use cases, but for embedded analytics, it’s slow, difficult to customize and painfully expensive. If you already use Looker internally and don’t have many viewers who need dashboard access, then Looker is the ideal tool for the job. Otherwise, look for some alternatives. Speaking of which…

The best alternative to Looker for embedded analytics

Looking for all the key features of Looker, but with a better user experience, lower cost and in-depth customization?

At Luzmo, we have an API-first approach to embedded analytics. Building and embedding your dashboards is a walk in the park for any developer with the basic skills. Within a few hours, you can embed and customize your dashboards to look like a part of your product.

luzmo dashboard

We offer true self-service analytics, allowing your users to view, customize, edit and create their own dashboards in Luzmo. Set your own permissions and rules and determine who can view and edit dashboards within your app or website.

The best part is - Luzmo’s pricing is fully transparent and you know exactly what you’re going to pay every month.

If you want to learn more, get a free demo with our team today!

Build your first embedded dashboard in less than 15 min

Experience the power of Luzmo. Talk to our product experts for a guided demo  or get your hands dirty with a free 10-day trial.