A Tableau guide to Salesforce
01st April 2021
Author Alex Waleczek
What you need to know about Salesforce and Tableau CRM
Since Salesforce acquired Tableau around two years ago, people on both sides (Salesforce and Tableau) wondered where this union would go. Two years on, a lot of integration has happened, but for the most part, the greatest benefit of this union is seen by Salesforce users.
People who already use Salesforce CRM and Tableau CRM (aka Einstein) are excited about Lightning Web Connectors and Einstein Discovery plugins for Tableau, while most people in the Tableau community read these announcements and move on with a shrug.
Let me shed some light on the Salesforce platform and ecosystem today and look at what is happening now, and then I’ll take out my crystal ball and share what I think might be happening in the future.
But what is Salesforce, anyway?
At its core, Salesforce is a CRM (Customer Relationship Management) system. That’s where it started, but by now it’s much more. With the Customer 360 framework, Salesforce aims to manage every single one of your customer connections, from searching on your website and signing up for an account, via getting customised emails and managing orders to call centres and mobile apps. In short, there isn’t a lot you can’t do in Salesforce. If you have a few minutes, this video will give you a good idea of what it is capable of. For the purpose of this article, I will focus on the core CRM functionality: managing accounts, opportunities and orders.
And Einstein Tableau CRM?
Here it starts to get confusing! As of now, Tableau CRM (TCRM) is exactly what Einstein was one year ago. The only difference is that they changed the name of it. TCRM is essentially Salesforce’s AI/ML engine. It can do graphs but, given the choice, I’d rather not use it for that.
The important thing to realise is that TCRM is NOT Salesforce. It is a separate component that you can use in Salesforce, but it requires a separate licence. You can use Salesforce without TCRM (as many organisations do) and can apparently even use TCRM without Salesforce.
In my opinion, using TCRM without Salesforce doesn’t make a lot of sense right now. It is very much designed to work as a Salesforce component; it can ingest external data but there are limitations around it, it is by no means as easy and powerful as Tableau’s connections, and you will need to buy additional storage space if you use this feature a lot. Machine learning applications in particular benefit from large amounts of data, so, depending on your business and use case, this may cause some unexpected costs.
“But what does it actually do?” I hear you ask!
TCRM has two main features that are coming to Tableau. The first one is ‘Discovery – imagine ‘Explain Data’ on steroids. You give it a dataset and define that you would like to maximise revenue/minimise churn/etc. Then it goes away, sifts through the whole dataset, identifies which attributes contribute most to the defined goal and where the greatest potential for improvement lies. Depending on the use case, it might tell me that one of my suppliers, who supplies product A, can also supply product B and, if we get all our product B from this supplier instead of our old one, we might be able to increase revenue by x% because they sell it y% cheaper.
In the newly released Tableau 2021.1 you can get this natural language output right inside your dashboards (if you have a TCRM licence), so if you change parameters in the dashboard, the story will adjust to the new circumstances.
The second big functionality in the latest release is predictive scoring. Imagine having a bunch of opportunities in your CRM and you would really like to know which ones are likely to close, so that you can focus on the ones that generate revenue rather than the other ones that won’t close anyway. Or you would like to know which of your customers are likely to churn, so you can get in touch with them proactively to keep them happy.
While you usually need a data scientist to make the data usable for prediction, determine which variables are strong predictors for your target and actually develop and test a model, TCRM does all of that on a no-code basis. You go through a wizard and at the end it spits out a score that tells you how likely it is that an opportunity will close, a customer will churn or whatever else you want to know. (These predictions are obviously just that: predictions. They can be reasonably accurate or horribly wrong, depending on your data and what you would like to predict.)
Can I try it?
Yes, you can! Salesforce makes it incredibly easy to spin up developer instances to explore the platform and test features. The only limitation I have found so far is that you can’t add more users to it. Apart from that, it’s basically the same as any other Salesforce instance. The feature to spin up instances as and when you like is actually a core feature of Trailhead, Salesforce’s training platform. If you want to get started with Discovery, head over to this trail, create an account and a developer instance and off you go.
Bobby Brill – Product Manager in the Einstein Analytics team – also wrote an excellent blog post, explaining how to integrate a Discovery story in a Tableau Dashboard.
So, why is it called Tableau?
Architecture diagram: how Tableau and Tableau CRM fit into the Salesforce ecosystem
The short answer is: “I don’t know! It’s confusing for everybody!” It actually makes it really hard to talk to people when they say something along the lines of “But we have Tableau already; we bought Tableau CRM/Einstein years ago!” and you have to explain to them that this is called Tableau, but it is in fact not Tableau (at least not the one you talk about).
The long and more useful answer is that Salesforce is using the “Tableau” brand to bring together all their analytics functions. In the context of Salesforce, “Tableau” and “Analytics” are synonymous and Tableau CRM is one part of their Analytics offering – I have heard it being called “One Analytics”, but whether this will become an official term, though, I don’t know. You can see in the diagram above that “Tableau” can refer to three different parts of the ecosystem, so making sure to establish what people are talking about is – at least at the moment – essential.
How does it all come together?
For now, the big winner of this merger is Salesforce’s customers (not surprising considering the sum Salesforce paid for Tableau). With an existing Salesforce system, these customers have a ton of data that will need to be analysed; and it’s already super easy to connect Tableau to Salesforce for analysis and embedding a Tableau dashboard in Salesforce takes only a few clicks. If they have a Tableau CRM licence, they might have predictive models set up already, and with the latest 2021.1 release they can integrate those easily right into Tableau.
For everybody who uses Tableau already, this instantly unlocks a whole lot of possibilities to make those predictions available to users across the organisation, maybe even those that don’t even use Salesforce but have use for its data. For everybody else, Tableau becomes the default solution for analysing Salesforce data. Sure, it has its own reports and graphs but it’s impossible to overstate how much easier Tableau makes analysis compared to the cumbersome native functionality in Salesforce.
The crystal ball
We have seen TCRM and Tableau become increasingly integrated in recent months. We also know that the TCRM teams are now part of Tableau (the company).
My wild guess would be that TCRM will become an add-on to Tableau (like the Data Management add-on). So, you’d get your normal Tableau licence and, if you have use for it, you’d also get the AI add-on and therefore all the features of Discovery and scoring that are available now. The difference would be that rather than logging into TCRM as part of a Salesforce instance, you would hopefully be able to access it right from Tableau Online, which also means that it would become much easier to use the predictive ability on non-Salesforce data. We might still be a while away from this, but it certainly sounds like a great opportunity!
(Full disclaimer: I don’t have a real crystal ball, and any opinions and predictions offered here are my personal conjecture only!)
If you would like to know more about anything covered here, we’d love to talk. You can get in touch with me at email@example.com.