The ROI of Looker, part 1: scaling and cost-savings

  • 28 March 2022
  • 0 replies

Userlevel 5

This content, written by Dillon Morrison, was initially posted in Looker Blog on Aug 15, 2016. The content is subject to limited support.

We have a saying here at Looker: “We’re in the business of getting people promoted”.

By that, we don’t only mean getting the person who buys Looker promoted for making a great decision. We mean getting every Looker user promoted as well. With broad access to reliable data, and the ability to perform their own analyses to make better business decisions, employees can perform their job better. We’ve heard from tons of users who were promoted thanks to learnings they made through Looker, and nothing makes us happier.

One way many customers like to show the impact of Looker on a business, is through the lens of ROI. We’ve seen this so frequently, we wanted to share the logic with our larger customer base and the public.

As with any software, the ROI of Looker can be difficult to fully quantify. Simple calculations on cost-savings account for some of the returns, but fail to capture the far-reaching and more nuanced returns of improved business processes. How can you accurately measure the financial impact of operationalizing data, or of competitive advantages? It’s not easy.

For part 1 in this blog series, we’re going to look at the ROI through the lens of FTE savings as a company scales. In part 2, we’ll get into some customer examples that demonstrate the returns of those improved business processes achieved through use of Looker.

“Before, the work that two of us are doing on our data analytics team would probably have taken 5 or 6 different people. With Looker, it’s a few clicks away.” - Sandeep Kamath, Senior Manager of Analytics, ShopRunner

We’ve often described the typical experience most companies face with inflexible data pipelines and tools. In this all-too-common scenario, business users seek answers to business questions from their analysts, who add that request to a long queue of questions from other users. Because of the bottleneck here, analysts typically end up reusing old queries and business users end up relying on the same periodic reports and . Business users are never given the ability to ask and answer new questions for themselves.

As a business grows, the number of users who need data, the amount of data, and the complexity of analysis all grow as well. This means the business needs to hire more data analysts to service those needs with more manual SQL. Costs are increasing, just to keep pace with the growth of the queue, but even so, business users are still facing the same bottleneck problem.

With Looker, each analyst’s work is far more leveraged, so the user-to-analyst ratio is far higher. That means that the number of analysts, and the associated costs, grows much more slowly. Looker translates an intuitive UI into optimized SQL queries to be executed across your entire database, down to row-level data, so any user can self-serve with reliable data. Analysts define the business logic once, in Looker’s modeling layer, and every user leverages those definitions

This modeling layer, , is the critical component that saves the need for additional analyst resources as a company scales.

“The department has 1,000 people, there's very few data analysts in this new model for us [with Looker]. The analysts are required for the insights into the data, but for just querying data, the people can do it themselves.” - Suresh Duddi, VP Engineering, Yahoo

Once an analyst has defined the business logic, users have infinite flexibility to query the data. And when the underlying data or schema inevitably changes, all the business logic can be adjusted with just a few lines of code, rather than it breaking all existing reports and dashboards.

We often see companies that are growing quickly turn to online SQL visualization tools. These tools are definitely an improvement over basic SQL editors, and make it easier to create visualizations and dashboards, but there’s no leverage. Every query still has to be written by a data analyst, so even if the executives have nicer dashboards, anyone who doesn’t speak SQL is still totally dependent on a data analyst.

That’s the fundamental difference between a true modeling layer and reusable SQL queries: software scales far better than analysts writing every query by hand.

“There was a lot of wasted time and energy when you think about starting from square one every time. Now with Looker, someone's already done that for you, and agreed on this is the best way to look at this data and you can get right into making decisions with data.” - Robert Olsen, Director of Data & Analytics, Digital Ocean

With salaries for experienced data analysts surpassing $100k/year in many cities, Looker easily pays for itself in short order. We regularly hear from companies that were hiring a new analyst for every 20-30 total new hires but, thanks to Looker, have been able to meet the analytic needs of hundreds of employees without hiring any new analysts

Some of our customers have great examples of these savings in practice. I’ll go into detail on several of these customer stories in part two of this blog series.

“Looker has definitely helped us in keeping overhead down for the analysts in the company. It's allowed us to focus on working in the teams, and on doing the forward-looking analysis, rather than working on maintaining our data infrastructure.” - Erik Johanssen, Data Analyst, Transferwise

If you’re interested in projecting the ROI of Looker for your specific company, or in learning more about how Looker can save you costs, please and we’d be happy to chat!

0 replies

Be the first to reply!