Analyze & Uncover

Customer Insights

Analyze & Uncover Customer Insights

Once you have a single source of data truth for your customers, the real work can begin. But don’t worry; your CDP will be doing most of the heavy lifting. This type of task is perfectly suited for artificial intelligence and machine learning. It’s all about processing massive amounts of data to identify trends, spot patterns, and ultimately recommend next steps.

This process will give you a clearer view of the customer journey than you would ever get by manually processing data. Customer data analysis on a CDP can uncover:

  • The traits common to your most valuable audience – which profiles are most likely to purchase, make repeat purchases, and become brand advocates
  • Which profiles are highly likely to buy soon, and which require further nurturing
  • Which customers are likely to churn
  • Recommendations for cross-selling and upselling

Here’s an example of the results of this type of analysis. In this case, an algorithm identifies the profiles that are most likely to make a purchase, creating segments based on its predictions. Marketers can then fine-tune the segments manually before automating a follow-up plan (as we’ll cover in the next phase).

Data enrichment expands your customer view

Let’s take a look at some of the customer insights you can discover through analysis on your CDP.

Create a More Accurate Attribution Model

In the early days of digital marketing, the “last-touch” attribution model was the most prevalent. If a customer looked at four blog posts, visited a retail store, then clicked a banner ad, then the banner ad got all the credit for the sale.

It’s easy to see how misleading that type of attribution can be. Based on a last-touch model, you might increase your ad budget and dial back on blog content, when it was the blog that started your customer’s journey. And a first-touch model isn’t any more accurate — it wasn’t just the blog that sparked the eventual sale.

Request a Demo of the Arm Treasure Data CDP

Request a Demo of the Arm Treasure Data CDP

Customer Segmentation: the How and Why of the Customer Journey

Customer Segmentation: the Customer Journey

Modern marketers are still searching for the perfect attribution model. It’s a worthwhile quest: When you have a clear idea of which touchpoints are contributing to success, you can optimize what works and jettison what doesn’t.

With your data united on a CDP, you can create a customized multi-touch attribution model based on your unique customers and their journeys. It breaks down to these four steps:

  1. Start with a large data set of customer data with unified customer profiles.
  2. Identify the touchpoints in your customer journey that impact a purchasing decision.
  3. Determine what KPIs you use to measure attribution.
  4. Use the analytic capability of your CDP to normalize, correlate and analyze the data.

Identify Omnichannel Opportunities

Analyzing your customer data can give you a clearer understanding of your customer journey across channels. The analysis is likely to uncover trends and correlations you wouldn’t have seen otherwise.

For example, your brick-and-mortar customers might be more likely to open an email from your brand right after a store visit. Or the customers that interact with your chatbot on Facebook are less likely to abandon their eCommerce cart.

In short, your analysis can follow each thread in the complex web of interactions that leads to a purchase. With these insights in hand, you can create smarter marketing triggers to respond to these cross-channel interactions.

Imagine if a customer to your brick-and-mortar store got a follow-up email with a “how to use your new product” guide, rather than a generic promotional offer. Or if a customer with a customer service complaint pending didn’t see retargeting ads for your brand everywhere, but instead got a personalized follow up email.

Each of these small tweaks to personalize your marketing can drive purchase decisions, lead to deeper relationships, and ultimately create lifelong customers. And all of these optimizations can be deployed at scale through your CDP (more about that in the next step).

Power Up Your Segmenting and Targeting

One major benefit of analyzing customer data on a CDP is increasing the focus and granularity of your audience segments. You can zero in on the profiles most likely to purchase in the next few weeks, next few months, or those who are just casually doing preliminary research.

For example, Subaru uses ongoing customer data analysis to find which profiles are ready to make a purchase, which are “just looking,” and even which ones are only “fantasizing” about getting a new car. Customers from each of these segments may eventually make a purchase, but they need drastically different messaging and next steps. By identifying these segments and personalizing their approach, Subaru has seen a 14% increase in closing rates, 38% cost reduction per acquisition, and a 250% increase in conversion rate for their most valuable audience.

CDP analysis can identify the trends and behaviors that indicate a customer’s current progress in their journey. Then it can identify which next steps have helped customers at that stage move closer to a purchase decision. You can use this analysis to set rules for automated, personalized follow-up.

Create Lookalike Audiences to Extend Reach

Marketers rely on demographics to target customers. But demographic data alone can miss the mark: Not every 42-year-old male from Houston, Texas has the same wants and needs. Even if you get incredibly specific — 20-25-year-old single mothers with an annual income of $40,000 — it’s still hard to predict individual behavior within the demographic.

That’s where lookalike audiences come in. This tactic targets a potential new audience based on the analysis of your existing customers. It uses machine learning to identify commonalities within your customer base, and use these trends to model which prospects you should be targeting.

Machine learning is crucial here, as it can parse thousands of customer attributes to find the behaviors that signal a high possibility of conversion. These behaviors tend to be very specific and occur at a low frequency, so it requires both a great deal of data and processing power to identify them.

Big Data Makes a Big Difference

Most of the tactics in this section will be familiar to any marketer. We’re always looking for new opportunities, testing which audience segments are likely to respond to a particular messaging, and seeking to optimize our efforts. The difference is one of scale and volume. It’s the sheer amount of data, the depth of the analysis, and the quality of insight you can glean — the difference between a backyard garden and a 1,000-acre farm.

Of course, generating these insights isn’t the end of the process. The next step is to put these insights to work, applying them to your marketing efforts, measuring success, and optimizing further.

Derive Customer Insights