Panoply Blog: Data Management, Warehousing & Data Analysis

Why You Absolutely Shouldn't Connect Salesforce Straight to a BI Tool

Written by Anders Schneiderman | Feb 9, 2021 2:15:00 PM

You want to get more out of your Salesforce data—and who doesn't? So it’s time to boot up your favorite BI tool, hook it up to Salesforce, and rock out. 

That's a fine plan if your startup has simple needs. But if you're planning on doing a deeper dive into your Salesforce data, you're going to need a better submarine.

In this article, I'll help you understand why working with Salesforce in a BI tool isn't a good solution and explore how to build a Salesforce data solution that can help your startup thrive.

Why not just use Salesforce for reporting?

Before we explore why connecting Salesforce to a BI tool isn't a good way to go, let's briefly address a common question: why not just do your reporting in Salesforce?

Here's why you don't want to do that: Almost every analyst who tries to use Salesforce reporting for more sophisticated work goes through what I call the four stages of Salesforce Reporting Grief.

1. Denial, Part 1: Awe

Salesforce is a tinkerer's delight; it's highly configurable, customizable, and programmable. All I need to do, you say to yourself, is really bear down and learn how to get the most out of Salesforce.  In the deep recesses of your brain, there's a small red warning light that is saying, ohhh mama. But you ignore it, because: toys! 

2. Denial, Part 2: Gritting your teeth and gutting it out

But as you dig into Salesforce, you run up against more and more limitations. For example:

  • Granularity: for individuals you can easily drill down. But for aggregates, digging into the data can get pretty tricky.
  • Baby data viz: Salesforce creates super simple charts (e.g., meters). They are easy to understand at a glance, but they don't provide the level of detail you need. And trying to build more sophisticated data visualizations in Salesforce can be quite challenging.

And the joy you initially felt gets slowly ground down.

3. Anger, bargaining, and depression

In this stage, you finally hit the point where you can no longer pretend to yourself that your data solution can easily revolve around Salesforce. Why?

  • Odd, frustrating analytical limits: Is it possible, for example, to analyze which accounts have purchased which products? Yes it is. But do you have the time and energy to painstakingly try one workaround after another until you figure out how? No, you do not. 
  • Your head is not a battering ram for breaking through data silos: Theoretically you could import and connect lots of other data sets in Salesforce and do the analysis there. But trust me, you really, really don't want to try.
  • Salesforce's pricing will drag you down: Because Salesforce seats are pretty expensive, at some point you are forced to make a no-win choice. You either drastically limit how many staff can be self-service beyond the basics, or spend money your startup really can't afford on a bunch more licenses.

By this point you've already spent enough time and money wrestling with Salesforce that the idea of chucking that investment is painful. So you alternate between being pissed, feeling hopeless, and bargaining with a Higher Power. (You have my sympathy.)

4. Acceptance

Eventually you come to terms with the fact that it's time to give up on addressing your reporting needs in Salesforce and try something else. Piping the data into your BI tool seems like the answer, but that creates all sorts of other problems.

Why not connect Salesforce straight to BI?

Once you're ready to let go of your Salesforce reporting grief, a BI tool might seem like a great way to go. After all, these days BI tools are incredibly powerful, able to manage a remarkable volume of data. And your favorite BI tool probably says it's pretty good at slurping down Salesforce data.

If you spend just an hour or two playing with Salesforce data in your BI tool, you might very well think that you've found a good solution.

But as you go from a super simple proof of concept to beginning to build out what your users actually need, you're going to run into problems. Here are three of the most common:

1. Limits built into Salesforce

Salesforce wasn't designed to act as a real time data repository for other tools and systems. As a result, it can't handle the traffic and bandwidth that such an approach would require. So, Salesforce limits how much you can use it externally. For example:

  • Rows of data: when you are pulling data from Salesforce via the API, you can't pull down more than 2,000 rows of data. That pretty much nixes using a large data set unless you put together a janky workaround. That’s doable, but is way more likely to break.
  • API calls: Salesforce sets a limit on how much you can hit their API. It isn't hard to hit that limit, and if you do it can impact other users without you even realizing it. For example, if you blow through your company’s API calls, Salesforce will just stop routing inbound leads to the Sales team. It may take a while for them to realize why things are so quiet, but when they eventually figure out what happened you can bet they won’t be happy.
  • Authentication tokens: Salesforce limits you to five authentication tokens per application. So if you have more than five Salesforce data sets to import, you're either out of luck or will need to jury-rig a workaround

2. Difficulty combining multiple data sources

Most BI tools do a pretty decent job of combining Salesforce with one other data source. Maybe your tool can handle two other sources without causing your head to spin. 

But if you need to integrate more than one or two data sources—which you most certainly will if your goal is to level up how your startup uses data—trying to do it in a BI tool is more trouble than it's worth.

3. You won't have a Single Source of Truth (SSOT)

Even if you can successfully juggle multiple data sources in your BI tool, in doing so you’re effectively creating a mini data warehouse. ...And so is anybody else in your startup who's using that BI tool to manage multiple data sources.

That’s a blueprint for a world of hurt because each mini data warehouse contains its own version of your data, how terms are defined, and how the data should be interpreted.

This is such a common headache that there's a standard cure for it: a Single Source of Truth (SSOT). Having one repository of data built around one set of data definitions means everyone in your startup is working from the same version of reality. That’s a great solution to a common problem—but you can’t do it with a BI tool.

And much, much more

These problems are just the tip of the iceberg. The further you dive in, the more issues you're going to encounter. For example, if your Salesforce is configured with lots of custom fields—and whose Salesforce instance isn’t littered with all sorts of custom cruft?—working with those fields in a BI tool can get quite tricky. Likewise, trying to keep the data straight as account owners come and go can be a real nightmare. And the list goes on.

Do you need an enterprise data warehouse for Salesforce? 

So if your BI tool won't cut it, is your only choice to build a full-blown data warehouse, also known as an enterprise data warehouse (EDW)? Salesforce certainly seems to think that's the case.

We’ve covered this in detail elsewhere, so we'll keep it short: Sure, you could create an EDW. And maybe if you worked for a big, blobby corporation you could afford to go down this road. But for smaller companies and startups, an enterprise data warehouse is a black hole for time and money.

A better way to go: A modern self-service data platform

If we were telling this tale five years ago, that would be the end of the story. But in 2021, there's a happier ending: modern self-service data platforms such as Panoply.

When you're working with data rich applications like Salesforce (or Hubspot, or Shopify), a cloud data platform like Panoply gives your company all the advantages of an enterprise data warehouse and the velocity that fast-moving startups need.

And you won't have to become or hire an expert in building and maintaining a data warehouse. For a fraction of the cost and in a ridiculously short period of time, you could have an MVP of your data solution up and running—and you will have laid the groundwork for a sustainable data solution that helps your startup beat its competitors.

So before you decide to spend the next few months trying to turn Salesforce reporting or your favorite BI tool into a data solution that it's not designed to handle, why not give Panoply a quick test drive? It’s totally free, no credit card required, and you can get it set up and start syncing your Salesforce data in minutes.