We have one deployment on AWS cloud and our enterprise instances use the Alation cloud.
We use it primarily for supporting our data scientists. We also use it for sharing data governance, document policy information, and data classification information.
Our use cases are for a lot of data literacy and data usability.
We had some of our data science engineers, who had built out some cloud-based data and data products for users. They were spending a lot of their time, up to 30 to 35% of their time, answering questions about the data and helping people understand how to use it. Once they were able to document that information within Alation, they were only focusing on around 5% of the high-value questions that really couldn't be solved through the catalog itself. It freed them up to be more productive, it made their end-users more productive, and it made the adoption rate of their data products rise more quickly because people were able to get to using them.
On the engineer side, they went from spending 25% of their time answering questions down to less than 5% of their time answering questions. They could just point people to the catalog instead.
Alation has increased the adoption in our organization. Using the business-led type of governance, where we're finding the experts where they sit in helping empower them to govern their own data, leads to higher levels of data literacy as self-service data, data tools, and development throughout the company.
It enables users across our organization to efficiently search for and discover data regardless of location. It provides a federated search interface that people can use to search across all of the assets in one day. So as long as we've connected those sources to the catalog, they can use the same search interface and the same types of keywords, tags, metadata fields, to search, to narrow down their searches, and to filter the information that they find, and find what's relevant to them and their needs very quickly.
Alation has not helped to reduce our team's workload on the governance side. The workload is a little bit higher because we're monitoring and curating the catalog itself, as well. But it has lowered the workload of other teams when they're interacting with metadata and they're trying to share information about data.
It has a really great capability to connect pieces of information and people to build out a graph or an ecosystem of the data that's being used; whether it's the system, the people, or the artifacts that are being generated with the data.
It's important to us because one of our key pain points for our data scientist is understanding the context and the data. It builds that context natively into the tool so that's served up to you. You don't have to go looking for that contextual information that you need about data.
Alation's active data governance, as opposed to command and control style data governance, is the way of the future. That's the approach that we have shifted to over the past few years, really trying to use a more adaptive governance approach. Where we're reacting to what the business is telling us we need, instead of trying to dictate how things should be done. It sits nicely with their analyst-focused product.
I know that this has been on their roadmap and they're working on it, but as a whole, there's an untapped reservoir of capabilities within their article features that could be used and built upon.
We've been using Alation for almost three years.
We found it to be very stable.
We found it to be very scalable. We haven't seen any issues in that regard.
In our corporate catalog, we have on average 300 distinct users monthly. In our enterprise catalog, where we're around 800 distinct users monthly.
Now that we're on the cloud service, they handle the infrastructure pieces there. We do have a server admin dedicated. Then we have maybe 1.5 full-time employees dedicated to it. Although that may be increasing in the near future. It's more for just driving training and adoption. It's not necessarily for maintaining the tool itself.
We will continue to roll it out. We have five different operating companies, so we will continue to roll it out across all of these companies.
Their technical support has been good. It's gotten better over the past three years. There were times where it was a little slow, but they've been taking steps in the past year and a half to really improve the turnaround time on their support tickets.
Neutral
We had a metadata repository called MetaCenter from Data Advantage Group. We switched because it was a very technical, somewhat limited metadata repository tool. And we were looking for something that was more end-user, business, user data, user-focused, and provided a more holistic catalog experience.
The initial setup was easy and very straightforward.
The actual application itself can be installed and deployed within a few hours. But then there is a process of setting up the data source connection and that really varies from data source to data source, depending on the administrative and infrastructure around each of those different types of sources. The actual setup within the tool itself is very simple and straightforward. It can be done by anybody through the GUI. Any kind of delays come from getting the prerequisites in place kind of on your company side, as far as your security needs and things like that.
We see fast time to value by being simple to implement and manage. We're implementing a new catalog now for our enterprise and we're seeing a lot of quick adoption in people who haven't had access to a tool like that in the past.
Pricing might be a little high for the fact that there are different kinds of users. It's taken them a while to get to the point where they're evaluating how to price these different users appropriately.
There are additional costs if you want to use the new apps, like the governance app.
We also looked at Waterline and Informatica Enterprise Information Catalog.
Waterline wasn't quite as mature at the time. It was a newer company. Informatica was really advanced and technically sound but there seemed to be a need to be within the Informatica family of tools to get the most out of it. But it also didn't have the same interactive experience for end-users, where they could easily communicate with each other and share information with each other.
We didn't see significant differences between them in terms of cost. When it really came down to it, we were looking at Informatica and Alation. They were, at the time, fairly similar.
It's a very effective tool but it helps to do pre-planning, have a good rollout, and architecture strategy place so that you can get the most out of it in a staged and progressive way. Start small and focused so you can provide real and demonstrable value and you can build on that as you go.
It's a one-stop-shop type of solution. I would rate Alation a ten out of ten.
