Are You Getting The Most Out Of Your Customer Data?

Are You Getting The Most Out Of Your Customer Data?

Photo Credit = ElasticComputeFarm@pixabay

What are you doing with your raw customer data in the data revolution?

Obviously, there is no shortage of data. Many brands are becoming more data-driven nowadays.Some have been collecting an enormous amount of data without really figuring out how to best use these data. Some are still lagging in both data collection and data utilisation. Some are dashboard-driven or metric-driven but not truly data-driven.

The leading and innovative businesses are well ahead of the game. They understand how to leverage their ever increasing raw customer data from different sources to gain a competitive advantage and to fuel exponential growth.

Let’s go through some of the areas with examples of how you could potentially utilise your raw customer data.

#1, Customer Segmentation

Customer segmentation is the process to divide a customer base into meaningful groups or sub-groups or even micro-groups who share similar characteristics and behaviours in the data. So that, you can understand your customers with a multi-dimensional view.

The sky is the limit! You can have hundreds or even thousands of segments depending on your marketing goals.

We all know that it is close impossible to get to know your customers at the personal level when you have millions of customers or even thousands of customers.

Essentially, the goal of customer segmentation is to gain a better understanding of smaller segments rather than getting to know each individual customer. So that, data-driven marketers can target and measure each segment more effectively.

Marketers have noted a 760% increase in revenue from segmented campaigns. – Campaign Monitor

And, there are many ways to slice and cut segments in leveraging advanced segmentation for targeting.

Let’s start with a simple example.

A very simple example is segmenting a customer base into Buyers and Non-buyers. Obviously, both segments should get different treatments.

A more advanced segmentation technique could be targeting customers who are aged between 21 and 32, female, purchased items X and Y but not Z, spent on average $200 in each transaction in the past 6 months, opened or clicked on your email campaigns featuring item Z. In addition, they look alike customers who purchased item Z based on the propensity model.

The map below is a simple customer lifetime value segmentation by post code.

Do you want to acquire more high value customers? Like loyal customers who stay for a very long time and who don’t complain. #wishlisted

Customer acquisition becomes easy when you have a well-defined segment that you would like to target. You can effectively acquire more lookalike customers with a crystal clear picture by knowing who they are and how to find more. It simply doesn’t make sense to acquire “more” customers. You want the high value ones and only.

You can attempt to change purchasing behaviour with segmentation to drive anticipated outcomes.

Here is an example.

You can create an incremental spend stretch campaign to get the mid value customers to spend a % more than what they normally spend. So that, you can strategically move this mid value customer segment into a higher tier.

Another example is to utilise a product hierarchy model to cross sell or up sell other products by strategically driving your customers in the lowest tier to upgrade or purchase additional products or services in the purchasing cycle. 

And, there are also other external data which can be linked to your existing customers for greater insights. External data sources like Mosaic provides a richer insight with socio-demographics, lifestyles, behaviours and culture information of your customers at the household level.

#2, Personalisation

Personalisation has been around for many years now. Many have already mastered the basic personalisation tactic. One of the most common personalisation techniques in email marketing is to include the [FIRST_NAME] in the subject line and or in the body of the email.

Are you already doing this? You can probably do better than this!

Customers are demanding higher expectation from brands nowadays. The primary goal of personalisation is to deliver an exceptional customer experience in relevance and engagement throughout the customer journey.

Personalisation can be far more personal and sophisticated than simply displaying a customer’s first name.

How about targeting your customers with an offer based on products or service that they are interested in? And, what about providing engaging and relevant information for your users? Brands like fitbit and Grammarly are doing some really good work in personalisation.

Batch and blast email campaigns like “Save now! 50% off!” and “BOGO sale” will probably train your customers to become bargain hunters who only shop when the sale is on. The more frequent the more anticipated.

A lack of actionable insights often struggles to have personalisation to reach its full potential.

74% of marketers say targeted personalisation increases customer engagement. – eConsultancy

Here is a simple example.

Amy is one of your top spenders who frequently purchased size S playsuits, short skirts and heels from your online store. You know there are thousands of Amys in your customer base who fall into the same segment. Obviously, you don’t want to launch another batch and blast newsletter to everyone in your base with irrelevant offers.

How would you deliver an engaging and relevant customer experience for Amys?

With a basic product recommendations model, you can create a personalised email campaign based on data such as transactional data and behavioral data.

You can also recommend other products which are closely correlated with products already purchased with a targeted and personalised offer.

Here is another example.

To offer a superb customer experience with personalisation, you can even become an online personal stylist to recommend products which are likely to suit her style and her body size by targeting the right offers at the right time with the right products.

You know she is a size S girl and loves wearing heels from the data. She would probably be interested in high waist hot pants with sandals as well. She is probably not interested in long dresses which make her short. #mybestguessonly #justanexample

In this example, you can offer a table for one experience with personalisation in scale otherwise not possible without personally interacting with the customers. The more data the more accurate the recommendations can be.

Personalised emails deliver 6x higher transaction rates. – Experian

Here is another example.

For international online retailers, personalisation can help factoring in seasonality and time zone differences by location when sending emails to international subscribers and customers.

Look, it doesn’t make much sense to send a winter collection email campaign to subscribers and customers in countries which do NOT have winter at all or they are currently having some great summer time! Well, they might be planning to travel overseas with your new winter collection. But, the chances are quite slim. This is not a great customer experience. The same applies to festivals. Be culturally sensitive!

Many savvy email marketers focus on email optimisation by send time. It doesn’t make sense having your subscribers to receive an email campaign at 3:00 AM in their local time zone. Again, this is not a great customer experience after all.

#3, Predictive Analytics

Predictive modeling is a commonly used statistical technique in predictive analytics to predict the probability and trend of future events by data mining and analysing associations between variables in historical data.

In data-driven marketing, predictive analytics has gained in popularity as a vital marketing tool. Predictive analytics can be used for a variety of marketing strategies such as campaign optimisation, customer lifetime value, propensity modeling, offer proposition engines and product recommendation engines. Many household brands such as Netflix, Google and Amazon are the leaders in this space.

Amazon provides site visitors with real-time product recommendations based on browsing data to offers discounted pricing and or package deals in order to entice upsell as well as premium pricing when demand is high and inventory is low.

Amazon has also begun to ship products in advance of customer orders based on the results of its predictive models.

Amazon Knows What You Want Before You Buy It. – Mashable

Would you proactively engage with your customers who are likely to churn or who have already churned? They are gone. They are gone.

Since new customer acquisitions are very high, predicting customer churn can help businesses to retain customers with targeted and personalised retention programs or offers.

Plus, it is more unlikely to retain the customers who have already churned. It is more likely to engage and to retain the customers who are likely to churn in the near future. #thepowerofanalytics

Your customers have always been telling you how they are interacting with your brand. There are signals which indicate that they are about to become disengaged with your brand. Some of these signals can be used to help predicting the likelihood of churn. Are you simply letting your customers leave? Hope not!

Here is an example.

A customer who used to engage with your promotional emails a lot and shopped regularly. Now, the customer is engaging less with your email campaigns (lesser opens and fewer clicks) and shop less frequently (-15% every period). Yet, the customer is still buying from you. But, you know this customer is likely to spend less and eventually stop buying from you.

What can you do when knowing this is happening? Act now not later!

Predictive modeling can help identifying these signals in preventing undesired outcomes like predicting customers who are likely to churn, to become disloyal, to complain and to shop at your sales events.

The effect of combining predictive analytics with personalisation is that product recommendations can be highly personalised based on the probability of purchasing other items i.e. purchases have not been made yet but very likely. “How do you know I like this?” #creepy #but #smart

There are many other ways to get the most out of your customer data. What other ways are you using your customer data? Tell us in the comment below.

Related Posts
what is data driven marketing