If you want to know how your customers are interacting with your brand, should you use Google Analytics, or do you need a Customer Data Platform? Or maybe you need both. But how do you know, and when do you use which?
This article should help you answer that question. The quick answer is that Google Analytics is good for anonymized data on web usage while a CDP allows you to customize and personalize your interactions with customers, using data from multiple sources. But it does get a little complicated, so read on.
Customer Data Platforms are a great tool that many companies use to consolidate all their customer data into one place so they can activate customer-centric, personalized campaigns. CDPs are also used to gather information on customer behavior so companies can fine-tune their offering, and even find new product possibilities.
Analytics packages are generally used to understand traffic to, and behavior on, your website. They allow you to track visits per day or week, page views per session, who came from organic vs. paid search, how many visitors are using a mobile or a desktop device, how people navigate through your site, etc.
There’s some overlap in many of these areas. For example, a CDP can also track mobile vs. desktop usage.
There’s a lot of variety in the CDP space, and there are a lot of interesting things you can do with Google Analytics (GA) – especially in conjunction with Big Query and other tools – so the principles outlined in this article are not meant to imply that you can’t do one thing or another with a CDP or with GA, but that these are the more typical and out of the box use cases.
GA and CDPs have a lot of things in common. They both …
- Profile each visitor.
- Collect some information on what the visitor does on your site.
- Can enrich their reports on audiences by receiving information from other sources.
- Can export audiences to other applications.
So what’s the difference, and is the CDP worth the expense, when Google Analytics is free? That depends on what you want to do with your audience data.
How Many, or Who?
Analytics packages anonymize individuals while CDPs collect personally identifiable information (PII), like an email address. Analytics tells you how many, not who. If your use cases require you to know a particular customer’s history, you’ll want a CDP.
For example, let’s say you want to identify people who have downloaded a particular white paper, but have not signed up for your webinar on the same topic. That’s a piece of cake with a CDP.
Demographics
Because Google knows too much about almost everybody, GA can show demographic information on all your website visitors, so you can sort your web audience by age or sex. CDPs can’t do that unless (1) you have demographic information on your customers that you can load into your CDP, and (2) a customer has become “known,” which typically means you have collected the customer’s email address. The CDP does not (typically) have demographic information on unknown visitors.
Note, however, that if you do have additional demographic information on your customers, such as whether they are retired, or whether they have a c-suite job title, you can load that into your CDP and create audiences based on those characteristics – for known users only.
Activations with other tools
You can import and export audiences to and from both GA and a CDP, but generally speaking that’s only useful on the GA side for advertisements. Which should be no surprise. Google’s whole business is built around advertising.
By contrast, you can build an audience in your CDP and send that to your email system, a short messaging service, or create an opportunity in your CRM. Since the CDP collects PII, you can get down to the individual and send personalized messages based on customer attributes. For example, when a known user has viewed several articles related to income tax, you can send an email promoting your tax newsletter, or create a task for a salesman to pick up the phone.
Paywalls
Media companies are rediscovering subscription and membership revenue, which often manifests in paywalls on their websites. Most CDPs can manage a paywall – or at least provide the relevant data to the system that does – but a paywall is outside the scope of GA.
Data cleansing
Data is a messy business. You might think you live in one town, but the post office says you live in another. Customers might use their full name on one form, and an abbreviated version, or just an initial, on another.
The goal of a CDP is to create a single customer view, which means you have to reconcile all this mess and decide if Greg Krehbiel in Hyattsville is the same person as Gregory Krehbiel in New Carrollton. Some CDPs include data cleansing options, but any CDP can connect with a data cleansing service.
This level of personal information is outside the scope of GA.
Making it simple
I could go on with specific use cases, but I’m trying to give you a useful overview to get the big picture. As a rule of thumb, GA is good for use cases where you’re only concerned with web data, and you need to know how many (in the aggregate), or you’re focused on web-based advertising. A CDP is best for anything that involves personalization. So …
Analytics
- Page views
- Campaign information in the aggregate
- Fluctuations in traffic
- Referrers and traffic sources
- Time on page
- Entrance and exit pages
- Bounce rates
- Age and sex of all users
CDP
- Lifecycles and customer journeys
- Paywalls
- Personalization
- Activations with other parts of your tech stack
- Campaign information by customer or specific customer group
- Advanced demographics of known users
Wrapping it up
GA and a CDP are both very useful tools for finding out what your customers are doing, and taking appropriate action to improve the customer experience. You should use both. I hope this short article helps you to sort that out. If you have any questions or comments, please use the comment box, or just give me a call. 240-687-1230. (Phones do still work.)
