Looker Reviews

The LookML model provides a single source of truth, which allows us to democratize the data analysis process in our company, getting more people's hands on the data, while still having confidence that the results they obtain will be accurate.

Looker greatly reduced the strain on the few SQL-proficient people in our organization. People who were previously waiting to receive data can now get it themselves and answer their own questions. The ease with which questions can be answered and explored leads to more interesting questions being asked and data being used more broadly across the company.

There are a few areas of the product where there hasn't been enough time spent making the interface user-friendly. Creating complex filters on reports or dashboard-wide filters, for example, requires guessing at the appropriate syntax, rather than having the super easy to use dropdowns that are available for simple filters.

4 months

Only a few intermittent and usually minor bugs.

No.

Excellent. The chat support always responds quickly and is helpful. Some weekend support would be nice.

We previously just ran our analytics out of Google Sheets with charts based on queries of our database each week. We reached a size where this was no longer feasible and we needed a more robust data warehouse and BI stack.

Very complex setup process that involves modeling your entire database in LookML. As noted above, this model is extremely valuable when completed, but the setup process is lengthy. Looker is aware of this and requires you to receive some training/support from them to get your model off the ground when you sign up, but even with this assistance you still have to spend a ton of your own time completing the model, especially compared with the setup time of other tools in this space.

The pricing model makes sense, but can be very expensive, potentially prohibitively so for smaller companies.

Yes, Tableau and Chartio.

Looker is an extremely valuable product that can really transform your company's ability to use data to answer challenging questions. It does require a significant investment, both of money and time, to use it successfully, so it shouldn't be purchased if there's not a willingness to devote resources to it both for the initial setup and on an ongoing basis.

LookML semantic middle layer: Gives me the ability to centralize and normalize all of the data for end users. This includes everything from names & definitions (which becomes our official data dictionary) to KPI calculations to custom datasets. Persisted Derived Tables (PDTS) are also a killer feature. This allows easy creation of custom datasets to expose to all end users.

No more backlog of simple data pull requests from most employees. Over 100 people are answering their own questions. Even PR loves being able to source their own data to build stories.

Their needs to be much more control of visualizations to become more on par with Tableau. We have to jump through a lot of hoops to do simple things like control what range is displayed on the X-axis of a trend chart. Also want more chart types: heat maps, for instance.

We have used this solution for one year.

Looker rolls out releases to customers in stages. When we were in the late stage bucket, we would get the release 2-3 weeks after it first launched and things were very stable. After we opted to be in the early stage, sometimes the product was unstable or buggy for 1-2 weeks after the release. Some bugs significantly hindered productivity, so we have switched back to being a late stage customer.

I would give technical support a rating of 9/10. They are very responsive and often installed patches within hours of reporting problems.

We previously used Omniture which was effective for click-stream analysis but limited in exposing other types of data, like product state data (I.e. how many users are currently subscribed to our newsletter?). We also considered Tableau, but it lacks a semantic/programmatic middle later and so is nowhere near as customizable. I wanted a tool that would be extremely flexible so I felt like we'd never be truly 'stuck' in what kind of dataset we could provide to end users without implanting large ETL projects.

The initial setup was simple. Looker helped us with best practices and we were up and running in a single day. Looker auto-generates all of the LookML required to expose tables, and the language makes it very easy to join tables and expose larger 'explores' (views) to end users. Love features that allow us to rename fields on the fly, add descriptions, and bucket fields into sections. Results in a very intuitive end user experience.

We implemented In-house. For Looker, get educated on data security features and best practices for organizing 'explores' into different 'models'. As with all data exploration tools, the performance is the biggest challenge. In Looker's case the performance bottleneck is not Looker, it's the database. So, have a strategy to optimize query performance. After that, setting up Looker's caching features is hugely beneficial. Most of our users just need daily data, so we globally cache all query results for 8 hours. That way if someone finds something interesting in the data and shares the link with 20 other people, those folks see instantaneous results when they click the link ( and no extra DB load). You can also set caching at more granular levels.

ROI is difficult to measure, but we are absolutely on the positive side mostly because of problems we've caught and corrected.

Several fixes saved us tens of thousands of dollars, and numerous product improvements have resulting in even more. Empowering every employee to make smarter decisions in less time easily pushes us into the black on ROI. Looker is competitively priced, and they are open to working with you on a fair price.

Looker can do anything SQL can do. If you are a SQL shop, you want to centralize data definitions, and you want programmable flexibility, then Looker is a wonderful choice.

SQL runner: Instead of having to connect to a slave DB, I can use the SQL runner in Looker to run SQL queries, which is more efficient in my opinion.

It is easier and quicker to find things in the database.

There is a lot of customization that we needed/need to build ourselves. Other BI tools have more out of the box. For example, products like RJ Metrics and Google Analytics come with pre-built dashboards based on your data; examples such as conversion rate over time or sales by product. We had to build those in Looker ourselves.

Also, it is often slow.

I have used it for 10 months.

I have not encountered any stability issues.

Technical support 9/10.

We previously used RJ Metrics. I was not part of decision to switch.

I was not involved, but before choosing this product, they evaluated Domo.

You must have a strong engineering team and a team member dedicated to set-up and customization of product.