Tableau, Salesforce integration now in full swing

It took more than a year after Salesforce acquired Tableau for $15.7 billion in June 2019, but the vendors finally unveiled plans for their integration during Tableau's virtual user conference.

During the opening keynote, Tableau unveiled the first product collaboration, an integration between Tableau and Einstein Discovery, an AI-fueled predictive and prescriptive modeling tool that's part of the Einstein Analytics suite Salesforce first rolled out in 2014 when it was known as Wave Analytics.

But Tableau president and CEO Adam Selipsky revealed far more than just a single capability.

He said the Einstein Analytics team joined Tableau's team earlier this year and is now under Tableau's guidance, Einstein Analytics is no more and has been renamed Tableau CRM, and that future integrations between Tableau and what was formerly the Salesforce analytics platform are in the works in order to ultimately create seamless interaction between the two.

Francois Ajenstat, chief product officer at Tableau, meanwhile revealed that the integration between Tableau and Salesforce will extend beyond what was formerly Einstein Analytics and include joins between Tableau and MuleSoft, a data integration vendor Salesforce acquired in 2018, and Datorama, an AI marketing software vendor Salesforce also bought in 2018.

Ajenstat recently discussed the integration of Tableau and Salesforce, including the reasons behind starting with a synthesis between Tableau and Einstein Discovery, challenges the vendors have faced as they join forces amid a pandemic, and Tableau's roadmap beyond its integration with Salesforce.

As Tableau and Salesforce began collaboration, why start with an integration between Tableau and Einstein Discovery?

Francois Ajenstat

Francois Ajenstat: It was the most natural and the one that extends the platform the furthest, earliest. But I'll step back for a second. Whenever there's an acquisition, there's this fear that one of the products will disappear or the focus will change, and we've seen this fear dozens of times in this industry. At Tableau we took a different approach. We went in with the perspective of not trying to rationalize these technologies but trying to figure out how to get the best of these technologies -- find what's truly unique and make a combination that's bigger than the sum of its parts. You take something like Einstein Discovery and see something like that doesn't exist in Tableau. It's powerful but limited to Salesforce, so we thought, 'Wouldn't it be great to extend it to any data, any user, anywhere?' So we just started going at it.

It's the same thing with the other pieces of the technology -- looking at what's unique, what's differentiated and how to leverage it. Not everything is the same size and some things will take longer to build, but this was one that was just the most obvious.

The integration between Tableau and Einstein Discovery won't be released until the first half of 2021, but are you able to give any more detail about when it might be generally available?

Ajenstat: We kept it vague for a simple reason. Our general approach is that we release software every quarter, so if for some reason we miss that one quarter it will be released the next. But because it's a brand-new technology we're going to be doing extensive beta testing with customers and getting feedback, and then if we feel that the quality and the feature meet the bar it will come out as soon as it's ready.

As for further integration plans between Salesforce and Tableau, do you have a sense of which features will come first and can you give a general timeline for their release?

Ajenstat: The next big one that will come out is around data interoperability, making sure that data in Einstein Analytics -- now Tableau CRM -- can work in regular Tableau and data in Tableau can work in Einstein Analytics. What we believe is that if the data is the same across both, the numbers will be the same, so we really want to make sure that foundation is clear. That's reading and writing back and forth between the two. That's actively being worked on right now, and there's a session during the conference that's showing an early view of that integration where they're querying 10 billion records in Salesforce from Tableau and getting instant responses.

We haven't disclosed a timeline. It's next calendar year, but as soon as we start getting that clarity about when we can get out of beta, we'll make sure to share it with everyone.

Another big revelation was that the staff of what was previously Einstein Analytics is now part of Tableau -- when did that first happen?

Ajenstat: That happened at the beginning of this calendar year. We brought together the engineering teams, the sales teams, the marketing teams all under one umbrella.

Was that before the start of the COVID-19 pandemic?

Ajenstat: Yes, it was pre-COVID.

Were you therefore able to get in some face-to-face collaboration before the pandemic?

Ajenstat: Yes, there was some, but I wish there was more.

You take two teams that don't know each other and start working together and at the beginning it was a little hard, but as we started aligning on what we're trying to achieve it got easier and easier.

Given that most of the work on the integration between Salesforce and Tableau -- the staff of Einstein Analytics included -- has taken place during a pandemic, what have been some of the unique challenges?

Ajenstat: COVID is hard for everyone. Adjusting to that new work-from-home experience, especially for people with families who have to be parents and teachers all at the same time, puts pressure on all of us. This has just been hard for everyone. But having said that, I'm really proud of the delivery of the engineering team. When I think of what our plan was for the calendar year, we've delivered more than 200 new features, most of them developed virtually, and I think that has been an incredible achievement.

And then to the question of integrating staff, you take two teams that don't know each other and start working together and at the beginning it was a little hard, but as we started aligning on what we're trying to achieve it got easier and easier. It became easier when we started engaging with customers and asking for feedback. When the customers said, 'Oh my God, yes, when we can we have that?' it just motivated them to make it work. It's a great proof point, and it's going to fuel the other teams to see what's possible, and the community reaction we saw on Twitter and LinkedIn [to the integration between Tableau and Einstein Discovery] is so motivating for these teams. I think we'll see even more energy and more innovation coming out of them.

As the teams began collaborating, did you notice some difficulty at first and then things get easier as time passed?

Ajenstat: It's been mixed. At the beginning we saw a slight dip and then an acceleration of work, and that was fascinating to see because we were monitoring everything that was going on [with the pandemic]. We actually got more productivity quickly, but I don't think people expected this to be a marathon. They expected it to last a month, maybe three months, and as the summer hit we saw a little bit of a dip. Now, what we've really tried to focus on is how to get balance and ensure that people don't burn out, ensure that people have an environment that supports sustainability because we can't keep going at that early pace. Now we're seeing this standard progression where teams know what they're doing, they know what their philosophy is, and they know what they're able to do and not able to do.

Are there particular platforms you've relied on for collaboration while the integration between the Tableau and Salesforce teams has been going on?

Ajenstat: All of the above. Between Slack, Zoom, Webex, Google Docs, we're living in a work-from-anywhere, all-digital world. In a way, you look at developers that tend to want to work in their environment and we've enabled them to be even more collaborative than they were before.

Given that everyone is working remotely, have people been able to make the best of a bad situation and perhaps travel while still working to get them out of their homes for a bit?

Ajenstat: We've had people go to Lake Tahoe, to Lake Chelan. One of the people I work with got rid of her apartment and is living in a van and traveling across the Western United States. It's great. In a way it's given people more flexibility to be anywhere they want. But the most important thing is that people have their well-being in check. We check on that quite a bit, and we encourage people to work in different ways that make sense for them and their unique situations.

Regarding the rest of the new features that are part of Tableau's roadmap, including a Slack integration and enabling users to do their analysis in their web browser, why develop these particular features at this particular time?

Ajenstat: Across all of them, we're always talking to our customers. We're listening to what they need, where they like to go and how analytics can help them. From that regard, nothing has really changed. We're continuously listening and working with customers and being customer centric. So when you look at what we've announced -- a Slack integration makes sense when we're all living in collaboration tools. We talked about Slack, and you'll see the other [collaboration] platforms coming in the future, but Slack is one of the popular ones, so that makes sense. People want to be notified and engage with the data where they live.

The browser one has been a journey that we've been on for many years where people want to have more flexibility to do their analytics everywhere. Having Tableau Prep in the browser and a full offering experience in the browser is something our customers have been asking for a lot. Centralized row-level security, as we're going more enterprise, addresses this need of having more centralization, more governance, more security, and it becomes even more important when you have everybody working on data. When you have 10 people it still matters, but it's not as critical as when you have 60,000 or 100,000 people and you need that trust. And then we talked about the Einstein Discovery one already, which addresses the theme of not telling me what just happened but helping me figure out what's in front of me so I can start adjusting my business accordingly.

Is there anything else we should talk about?

Ajenstat: I'd say three things. The first is just echoing what was in Adam's section of the keynote about how data has really gone mainstream. You look at everything that's happened this year and the number of crises, and for me the thing that's been recurring in my conversations with customers is the need for speed and agility, empowerment, and being able to look at all the data you have has just accelerated in importance compared to before. If you have a technology that takes six months to implement versus one where you can get value in minutes or a day, that's drastically different for the survival of a company, and I think that plays really well to Tableau's story.

The second piece I wanted to add is just the importance of our partner ecosystem. We've talked a lot about what we're doing, but as you look at this conference there are a lot of partners that are supporting us. All of these players are super important in the space, and we're continuing to work deeply with all of them to add more value to our customers.

And then the third point is that whenever there's an acquisition, there's this fear that you'll lose your focus -- you either focus on the acquirer, or you slow down because you're distracted by the acquisition. We thought deeply about this so we wouldn't make the mistakes of the past, and when I reinforced that we delivered a lot this year -- we delivered on schedule, every quarter -- that's really important because those features had nothing to do with Salesforce. They were part of our mission, so we're continuing on that mission and expanding it, and that was a very deliberate stance and sometimes one that can get lost.

Editor's note: This Q&A has been edited for clarity and conciseness.