Item: GoodData
Rating: 9
Author: Verified User

Overall Satisfaction with GoodData

Use Cases and Deployment Scope

We use GoodData on the introductory plan and have found it very beneficial so far. We made an intentional decision to centralise all of our cross-functional data into one archival database, and we use GoodData to display this data. This reduces our reliance on 3rd party analytics apps that may have multiple methodologies for measuring the same event (common if you are a product business) and it also enables us to see the customer journey end-to-end. In terms of usage, GoodData is really designed for detailed business analytics & data scientists. We do not have that expertise in house, so we rely on their no-code tools to get set up. While there was a small learning curve (the online slack community and help docs are helpful here) we are now in a place where anyone in the business can play with the data. We are also working with our devs to make the archival database more "readable" for non-technical employees, i.e. by using more self-explanatory naming conventions.

GoodData gives businesses an opportunity to expose selected parts of their data set to a limited number of employees. It then gives those employees the power to query this data in a no-code manner, which reduces (or eliminates) reliance on devs and data scientists. Data is loaded into models, which live inside workspaces. You can include / exclude what is needed from each model, and you can also pipe multiple datasets into a model. Similarly, you can include / exclude employees from certain workspaces. Employees have an option to export data to CSV or XSLX if there is no appropriate graph available. This gives you both security and flexibility.

Pros and Cons

Pros

Supportive community on Slack
No code data visualisation
Access control on both model and workspaces

Cons

No option to visualise a pin on a map
Custom metrics and variables can be tricky to get right
Hard to gauge where you are in terms of workspace data limits
Adding and removing DB fields from models is not as intuitive as it could be
Data model loading logs can be quite dense and hard to read, esp. when new
Setting up models to correctly reference between DB tables takes practise
No option to limit the amount of data a model loads once the table is included

Return on Investment

GoodData is currently saving our business ~3h week in manually collating reports

Usability

GoodData looks great but there are still elements that aren't fully optimised for non-technical users. The creation of custom metrics, model creation and model loading could be more intuitive than they currently are. The system shoehorns data into facts, dates and attributes without clearly explaining the difference between these modes. Expect that at least one member of your business will need to do some helpdesk / slack channel diving to understand the nuances. Similarly, the data model utilises a 1 -> N or N > N relationship between tables that may need further explanation for those who aren't from a data science background.

Implementation

We were up within 24 hours. I was able to get the DB keys from our engineers, put them in and start experimenting with the model. Getting the model right took about 2 days. We had several abortive attempts as I got comfortable with GoodData's functionality.

Self-Service Analytics

The ability to create graphs without prior SQL knowledge.

Integrations

N/A

Key Insights

Do you think GoodData delivers good value for the price?

Not sure

Are you happy with GoodData's feature set?

Yes

Did GoodData live up to sales and marketing promises?

Yes

Did implementation of GoodData go as expected?

Yes

Would you buy GoodData again?

Yes

Alternatives Considered

Metabase

We found Metabase very buggy in comparison to GoodData. While using Metabase, we frequently had to reload views and tables, and we also had to "re-do" the joins for every new query that we had. This made Metabase very inaccessible to our non-technical users, even though they were interested in learning the program. In addition, Metabase reads directly from the DB, so you don't really have control over the parts that your employees see.

GoodData abstracts the DB into a mini model of its own. This essentially means that you pick and choose which tables are relevant to your team, upload them into the model, and then let the team interact with that model through a no-code interface. It cuts out the need for manual joins between tables, and it also adds a layer of security in that you can control what people see.

Likelihood to Recommend

I would recommend GoodData to businesses that have non-technical users whose decision-making would benefit from having regular access to cross-functional data. In addition, GoodData supports the creation of 3rd party dashboards. This can be helpful for businesses that need to externalise (some) of their data to a third-party, e.g. a client, on a regular basis.

GoodData Feature Ratings

Comments

Please log in to join the conversation

Small business using GoodData