Spray and Pray Doesn’t Work: How to Identify and Target Micro Segments

Spray and Pray Doesn’t Work: How to Identify and Target Micro Segments

As marketers, we’re all trying to improve the customer experience and increase conversions. We have these things in common.

However, some marketers are much better at understanding their customer personas and doing the right kind of research than others.

What is comes down to is that delivering a single message to your entire customer base is an inherently flawed strategy. High-value customers, frequent browsers, seldom purchasers, brand enthusiasts and first-time visitors are all differently characterized and must be engaged uniquely.

This is where customer micro-segmentation comes into play.

The way to reach micro-customer-segments with resonating messages is by taking a two-pronged approach:

  1. Developing deep customer segments
  2. Tailoring personalized or customized messages to each segment to guide them through their decision-making

Segmentation and the Big Data Problem

There’s a challenge to targeting micro-segments: taking vast piles of behavioral data garnered from customer interactions and coaxing meaning out of it is hard.

If viewed holistically, this data is just a generalized aggregation of disparate data points. In other words, it’s a bunch of useless, out-of-context numbers.

For example, let’s say you’re an eCommerce marketer, and the following CRM report just came across your desk:

Unique users: 13,020,403
Revenue impact: $260,408
Average revenue per user: $0.02

Are these numbers useful? Hardly. There’s no context here. You have no idea who your main buyers are, where they came from, which products they were most interested in, what their purchase journey looked like, or why they ended up buying.

We know that a spray-and-pray approach to marketing won’t help your business remain competitive in an age where customers expect website experiences tailor-made just for them.

We also know that personalization can deliver five to eight times the ROI on marketing spend, and can lift sales by 10% or more.

To start with robust targeting, however, we must first learn about our audience – their unique preferences, interests, desires, behaviors.

That’s where segmentation comes in.

What is Customer Segmentation?

Segmentation is the process of dividing visitors into subgroups based on a common set of characteristics. These characteristics can include behavior, psychographics, demographics, geo-location and more.

Segmentation ensures you’re treating different visitors how they should be treated — differently. It’s the data-driven answer to spray-and-pray, the blueprint behind creating targeted website experiences and not ones based on how any visitor might interact with your site.

segmentation illustration

Segmentation begins to answer questions like:

  • How did this visitor arrive onsite?
  • How many pages did they look at?
  • Which elements did they interact with?
  • How much time elapsed between initial page view and purchase?
  • Did they place high-ticket items in a shopping cart to determine total cost, with tax and shipping?

…and more. If we take a deeper look at interactional data from a structured, segmented point-of-view, we can map out the journey’s customers are taking, define interesting and actionable segments and identify high-value visitors within those segments.

Dissecting the Customer Journey

Google’s Customer Journey to Online Purchase tool shows how different marketing channels lead to online purchases. To illustrate, if you’re a large US eCommerce brand, your customers’ conversion journey usually begins via engagement with social or display ads, with paid and organic search following soon after.

The Customer Journey to Purchase

If you know, for example, that in 83% of the cases, Search helps customers gain awareness and increases interest in the beginning to middle phases of the purchase path — you can create a segment to engage search traffic differently.

The question we all want answered is: how do you do that? The answer is by creating micro-segments.

With many visitors exhibiting different behaviors, there is a lack of clarity in understanding which visitors and which behaviors warrant segmentation. However, by developing ‘micro-segments’— granular and precise segmentations based on real-time behavior and onsite interactions— marketers can unveil valuable insights.

Choosing the Right Micro-Segments

A critical question when it comes to creating micro-segments is: ‘how many segments should you actually begin to target?’ We already know that each new targeting rule brings increased complexity, so how do you know which segments to prioritize over others?

Let us be first be clear: micro-segmentation is especially effective for high-traffic sites. For sites without an excess of traffic, spending time and resources on micro-segmentation could be a waste of time.

If you have a high-traffic site, however, understanding which segments to prioritize over others is a matter of ongoing analysis, taking all available analytics and drilling as deep down as the data will allow to understand the behavior on your site (as you’ll see further below).

Start by thinking about pain points…

  • What causes some visitors to buy over others?
  • At what point are visitors falling out of the funnel?
  • What common characteristics do they share?

With these questions in mind you can begin to explore further and both define the amount of micro-segments that will be valuable to you and how you can create a better process to satisfy their needs.

To get a better sense of which micro-segments you could be creating, you can use this segmentation recipe template which highlights some interesting and useful segments and can be applied to almost any eCommerce site. As a rule of thumb, each segment you create must have a least a few hundred visitors in it.

You can access this spreadsheet of ~50 recipes by clicking here.

To understand how you would create a particular segment for eventual targeting, let’s take one very potent and relevant example from this list: unconverted recommendation engine clickers — users who clicked on a product page recommendation unit, added items to their carts, but eventually – did not make a purchase.

Using your web analytics solution, you could create a session-level segment using the following set of conditions:

Screen Shot 2016-08-30 at 09.23.23

Why Invest in Micro Segmentation?

Micro-segmentation is critical for three reasons:

  1. The process of defining smaller but more fine-tuned customer segments allows you to discover valuable information about the characteristics of your site’s most valuable visitors.
  2. Micro segments are extremely homogenous, compared to broader segments which are more difficult to target since they are heterogeneous by nature. Targeting homogenous segments with customized offers across touchpoints is relatively easy, since you already have a great deal of information about their buyer profiles, online behavior, and intent.
  3. Defining a highly profitable segment is the most important step towards achieving revenue growth. Essentially, companies who invest in micro-segmentation as a means to improve revenue, are able to provide better online experiences, improved engagement and increased conversions.

Take a look at an actual segmentation example from a major retail brand.

Micro Segmentation Example

The marketer behind this site knows that footwear is a best-selling product. To explore this further, they created a segment of visitors who browsed the footwear category.

A closer look at this data reveals that 15.8% of all users visit the footwear category, representing close to million unique users. Yet although these users constitute approximately 16% of the site’s entire visitor base, they’re actually contributing 39% of the sites overall revenue, $8.181 Million to be exact.

Segmenting visitors by browsing behavior is a great start. But if you were to stop there and begin creating messages to engage this segment as a whole, you’d be severely limiting your chances for conversion. In other words, by creating only one experience for an entire segment, you’re assuming that all visitors within this segment are the same and should be treated the same.

What would happen if you narrowed down this audience?

Notice that while the Average Order Value of this segment is $122, in reality, some of these users may have much higher average order values. Grouping all these visitors under the same umbrella would be akin to me saying that the average net worth of Warren Buffett and myself is $31.6 billion. See what I’m getting at?

Instead, let’s add another layer of filtering to this segment, to show only users with a lifetime purchase value of more than $300. The result: Less than 0.1% of the overall users, nearly 4,000 people, who contribute to more than $2.1M in revenue. That is a truly valuable segment with an AOV of almost $250. By applying this additional layer of segmentation, we’ve identified visitors with an increased ARPU of approximately 6,230%.

Micro Segmentation Example

Micro-Segmentation and The Pareto Principle

Italian economist Vilfredo Pareto famously implied that most of the wealth is generated by a minority of the population, which has been commonly understood as ‘80 percent of the effects come from 20 percent of the causes.’

While the 80/20 rule certainly has benefits in framing our minds to efficiently prioritize vital activities over more trivial ones, the truth is that marketers today find themselves in a post-Pareto world, where the causes and effects of consumer actions are driven by an even smaller, elite minority, just like these 4,000 visitors.

Pareto Principle in Marketing

When you look at the stats above, it is immediately apparent that the 80/20 rule has become a much narrower one. In today’s marketing environment, less than 1% of customers are driving 90% of the overall revenue.

Seeing results like this should make us ask: Shouldn’t we be determining which visitors bring us the most value, revamp and double down on our marketing efforts to them, while reducing efforts to convert the less-contributing ones?

One reason this doesn’t happen is that customer acquisition often takes precedent. In other words, we spend so much time and resources figuring out ways to bring in new customers when we should really be concentrating on enhancing the experiences of our most valuable, existing ones.

Taking Action on Micro-Segments: 5 Targeting Ideas

Let’s assume you’ve already invested significant time micro-segmenting your visitor base. Now what? How do you engage them? How do you convey that you understand their needs on an individual level?

The answer is by creating segmented experiences for individual visitors within a given micro-segment. Here are a 5 ideas to follow:

1. Tailor different homepage experiences for different segments

Affected Segments: “Multiple Purchases in Category” / “Highest Spenders in Category”

For many visitors, the Homepage is the first page they see of your website – and in many cases, if the homepage is not gripping enough, it is the last and only page they’ll see. Customizing your homepage with relevant content, imagery and messaging can go a very long way in reducing hard bounces and taking your visitors through an engaging journey onsite.

For example, you could customize the hero banner per each segment. This means personalizing your homepage slider or hero banner according to past behavior, purchase history, or product affinity. For visitors with a demonstrated history of browsing or buying children’s clothing, a relevant ‘kids clothing’ variation can be automatically presented to these visitors as soon as they land on the homepage.

segmented homepage banners

How do you measure the efficacy of the targeting? How can you be sure that the resource cost will be worth the investment?

As a rule of thumb, you should be measuring the impact of your targeting on one KPI. For example, if you are targeting the hero banner to different segments and your goal is to increase revenue, then your chosen KPI should be the amount of purchases that are derived as a result. If, upon viewing the analytics, you see that the cost outweighs the benefit, then the targeting initiative isn’t worth it.

When it comes to the message itself, the only way to be certain that you are sending the most effective offering is by A/B testing each message against a control group over time. However, before you do that, you must ensure that you have a large enough sample size. When targeting micro-segments, a significant sample size has to be at least a few hundred visitors per segment. There’s no reason to focus on very small segments that will have no or very little impact on your business.

2. Dynamically change category pages with a product catalog that is tailored to the characteristics of each segment

Affected Segments: “Big Spenders” / “Maximizers”

The category and grid pages are your store’s aisles and shelves. Visitors are exposed to hundreds of products, and you must make it easy for them to find what interests them by pushing the most relevant products to the top of the page, and sorting the products according to their relevance for each and every visitor.

For a segment of top spending customers, for instance, you can show the highest-yielding products first, including new arrivals (which tend to be more expensive).

grid page customization

3. Drive revenue and build trust by leveraging social proof and demand messaging

Affected Segments: “Browsers” / “Researchers”

Visitors are easily distracted and often lose focus when shopping, leading to high abandonment rates and incomplete purchases. If you motivate them to make a decision, you can encourage them back into the purchase flow.

For example: Booking.com has perfected the art of surfacing real-time notifications based on vistor’s activity, including location-based view counters or trend alerts to validate purchase intent in a user’s mind.

booking.com example

Notifying customers that 24 people have booked a specific hotel for example, creates a correlation between urgency and buying in the mind of the customer, and induces a fear of missing out for not going through with the purchase. This method of using the power of social proof marketing could be adapted to any eCommerce site and can be extremely effective with browsing and researching users, e.g. users who spend a relatively great deal of time on the site, or a lot of time exploring a specific product page.

social-proof message

4. Deliver segmented, time-sensitive coupons to drive more engagement

Affected Segments: “Bargain hunters” / “Coupon Lovers”

Bring visitors back to context and encourage them to take action by serving targeted coupons with limited-time offers, tailored to different segments. Visitors that have a demonstrated history of purchasing with coupons can be automatically offered them.

time-sensitive coupons

5. Surface notifications to drive visitors to purchase forgotten items

Affected Segment: “Shopping Cart Abandoners”

Cart abandoners cause companies to lose $18 Billion in sales revenue each year. One way to engage returning visitors that fall into cart abandonment segments is to present a drop-down notification upon site arrival, encouraging them to purchase the item(s) they left behind.

forgotten item notifications


Marketers are living and operating in an aggressive Pareto era where the vast majority of revenue is generated by a small percentage of users. By identifying and nurturing your most valuable visitors using micro-segmentation, you will gain a sharper perspective of how to improve the online customer experience as a whole, and increase revenue and conversions.

While micro-segmentation will enable you to distill your customer base and discover your most valuable customers, keep in mind that it’s not as simple as plugging in and running on autopilot. Discovering valuable segments and creating segmented experiences is a continuous process that requires marketers to constantly be thinking about how visitors are interacting with their sites.

Feature image source

Related Posts

Join the conversation Add your comment

  1. Great post! Thanks for your valuable insights!

    1. Avatar photo

      Thanks Joris. I’m glad you find it insightful.

  2. What tool do you use to customize a web experience per segment? Optimizely has one, any others you’ve used?

    1. Avatar photo

      Dynamic Yield offers a full suite of capabilities to personalize and optimize the user experience for revenue growth. What sets Dynamic Yield apart is that we provide a single platform for advanced segmentation, real-time personalization, product/content recommendations, automated optimization, and 1:1 omni-channel messaging – across web, mobile and email.

      I highly recommend you check it out – http://www.dynamicyield.com

Comments are closed.

Current article:

Spray and Pray Doesn’t Work: How to Identify and Target Micro Segments