Customers are the lifeblood of your business. Understanding how you acquire them is critical to your company’s success.
Equipped with HubSpot’s marketing automation tools and CRM, you gain valuable insight into how contacts behave and interact with your business. You can also get more robust insights using this data by taking it a step further with data blending.
What Exactly Is Data Blending?
In a nutshell, data blending is the process of combining data from multiple sources into a single dataset. For example, you might blend customer, sales, and financial data to create a dataset that has all the information you need to answer questions about customer acquisition cost, lifetime value, etc.
Let’s take a look at why data blending is valuable for your company’s growth, how to blend HubSpot data, and common pitfalls to avoid when blending data.
Why Is Data Blending a Key Element of Corporate Growth?
When you're trying to answer a complicated question, the information you need is probably not in one spot. It’s scattered across different tools, often leaving you to analyze that information in isolation. This makes it difficult to understand how various parts of your business relate.
Blended data, on the other hand, helps you see the relationships throughout your company and account for other variables. It also offers more nuanced insights than a single data source.
Most importantly, blended data helps you understand your company’s inputs and outputs so that you can forecast, strategize, and hit your goals. To keep your business growing, you need to know that each dollar you spend is delivering results.
Because to evolve as a company, you need to go from understanding what is happening to why it’s happening. And as you evaluate your performance, you can better understand how to improve it.
Blending Your HubSpot Data
Now that you understand the importance of data blending, let’s take a look at how you can blend HubSpot data with native reports and external, third-party tools.
Blending HubSpot reports: Deals and contacts blending
Say you wanted to evaluate which referring sites are best. To do this, you need to know how many closed deals were associated to which referring site. To get that information, you would need to blend your deals data with your contacts data.
First, create a custom report in HubSpot:
To get started:
- Navigate in your HubSpot account to Reports > Dashboards.
- In the upper-right corner, click Add report.
- On this screen, navigate to the upper-right corner again and click Create custom report.
- On the left, click Across data sets.
- Then select Deals and Contacts as your datasets.
Next, determine the number of deals—and the amount of revenue associated with each deal— grouped by the referring site:
- Under the Data tab of the report builder, navigate to the left of your screen and click the Properties tab.
- Select Amount (Deal) for revenue amounts.
- Select First Referring Site (Contact) to group by referring sites.
Now, filter this data by Closed Won deals:
- Switch over to the Filters tab.
- Under the Deal Filters menu, click Add Filter and enter Deal Stage.
- Select Deal Stage from the filtered list. You should already be on the is any of option.
- Click the dropdown menu and enter Closed Won.
- Select Closed Won.
- Select Apply Filter.
- You can now get the data by selecting the Get data button.
- Once the data is in, switch to the Visualization tab of the report builder.
- Create a bar, column, or table chart with Count of Deals or Amount (Deal) on one axis and the First Referring Site (Contact) on the other axis.
You should now have a chart that blends referring sites from your contacts object and closed won deals from your deals object. Depending on which chart layout you choose, you can now see where the most deals by count or amount are coming from. You can give this to your marketing team to optimize how much time you’re spending on those referring sites.
You can learn more about building custom reports across data sets in the HubSpot Knowledge Base.
Blending HubSpot data with other sources
Pulling data from multiple sources into one place can be handy in building out a complete picture of your efforts. This means that at times, you’ll want to blend HubSpot data with information from other sources.
Some important third-party sources for tracking how you acquire customers is traffic analytics (such as Google Analytics in addition to your HubSpot tracking) and ad campaign platforms (e.g., Facebook, Google, LinkedIn, etc.). For more ideas on what platforms to use with HubSpot, check out this integrations list.
Using Grow to blend multiple sources easily
When blending data sources manually, you have to export the data from each source and then join them together in a spreadsheet or database. The process can be time-consuming and challenging, often leaving you with out-of-date data—but, it is possible. If, however, you wish to simplify the process of blending data and get access to up-to-the-minute reports, a business intelligence tool like Grow can help.
Data Blending Pitfalls to Avoid
While data blending may sound as easy as joining a few reports, there are some things to look out for. You’ll want to pay close attention to these three specific areas:
1. Business logic
One of the biggest challenges of blending data is the business logic behind your various datasets and sources. Your data sources aren't inherently related, which means each one may have different definitions for the same data point. It’s up to you to understand where the definitions overlap.
For example, two of your data sources may refer to leads, but do they define it in the same way? A lead as defined by one of your ad platforms likely does not match the custom logic you set up in your CRM.
If you're going to blend those data sources, you need to make the definitions match up, or figure out how to account for the difference.
2. Correlation vs. causation vs. omitted variables
The primary reason we blend data is correlation—that is, we expect to see a relationship between the data sources so that we can understand them better. However, this expectation can become a pitfall if you're not careful. That's why it's so critical to understand the differences between correlation, causation, and omitted variables.
Correlation, as we've already stated, is a relationship between two things. For example, if you were comparing data about people's height and shoe size, you would expect to see a relationship between those data points, with greater height generally correlating with a larger shoe size.
However, while height and shoe size tend to be related, we know that a person's shoe size doesn't cause their height, or vice versa. And yet, this is the mistake that many well-meaning business leaders tend to make—attributing causation when two things are only correlated.
For example, let's say you were tracking site traffic against inbound leads. Generally, you would expect to see a positive relationship between these two: As your site traffic increases, so do your inbound leads.
It would be easy to assume that an increase in site traffic caused the increased number of leads...but are you sure? What if there's a third, unseen variable that's impacting both of them, such as increased spend on a high-converting ad? These so-called omitted variables can be tricky to spot, but it's important to ask questions about your data and search for them so that you fully understand your results.
3. Data Cleanup and de-duplication
De-duplication, or removing duplicate entries (sometimes also called "merge and purge"), is a critical part of keeping your data clean. Keeping data clean can involve many tasks such as removing invalid emails from your list or standardizing terms (e.g., "U.S." vs. "USA" in form fills).
A common problem is that your data includes one-to-many relationships. For example, you may have multiple contacts connected to a single deal or multiple deals connected to one company. You may even have many-to-many relationships, such as different salespeople within your company contacting the same lead at different times of the year. However, you have to understand how those relationships work in your data so that you know where you're duplicating entries; otherwise, you could end up counting things twice.
Getting Better Results With Blended Data
To grow a healthy business, your number one goal must be to make every dollar you spend yield more dollars in revenue.
Acquiring customers is crucial in increasing that revenue, but tracking how you acquire customers and the associated costs is difficult. Interactions and spend are sometimes spread out over a number of applications, leaving data siloed and sometimes ignored altogether.
Originally published Feb 20, 2019 9:00:00 AM, updated February 20 2019