The number of new customers is a popular KPI request of a database provider. Believe it or not, there is more than one way to answer the question and key factors that can impact measurement.
1. Determining Original Purchase
The most common definition of a new buyer is the date of first purchase ever, with a given company. This is usually determined by reviewing all purchases and isolating the first one. If you haven’t provided all purchase history going back to brand inception, your first purchase date will be limited to the time period captured. If you or your database provider performs archiving, it is important to “freeze” the first purchase date. Maintaining the first 5 purchases is even better as it allows you to analyze a path of purchase trail for 1st time buyers.
2. The Impact of Data Aggregation
Customers can be analyzed at different levels of aggregation: email level, customer level, household level, address level or site. If you are determining new buyers by counting the number of email addresses in your ESP, you may be overstating the count. One customer, as defined by first & last name and address, can be associated with multiple email addresses. Aggregation is key to truly understanding the difference between additional or new contact information vs. a new relationship.
If you are defining a customer at the household level, the first purchase is based upon last name and address. The first purchase can change over time with the ebb and flow of people moving in and out of a household for various reasons. When NCOA and other address changes are applied to the database, its composition changes. A buyer may leave a household which could result in any of the following changes: 1) the creation of a new buyer record which only looks at that person’s history 2) the aggregation of that buyer into a pre-existing household which adjusts the overall first purchase date given his/her history and 3) a change in the original purchase date for the household that he/she left based upon the remaining orders. As a result, you may see changes in historical acquisition counts (for prior years) based upon the latest householding of the customer file.
3. The Impact on Prospecting
Many marketers are members of a co-operative database (i.e. Epsilon, Path2Response) and use these sources for direct mail prospecting. Some of the co-ops send prospect names net of the entire brand’s client file. Others send names net of a smaller segment of the buyer file, i.e. those who purchased in the last 3 years. In evaluating acquisition performance, reporting may show that a given list delivered x number of new buyers when in reality some are re-activated older buyers who were not suppressed from the rental. There are pros and cons to either approach. An argument can be made that re-activation of a former customer is acceptable in that “a sale is a sale”. However, it does cost less to run a re-activation model than to pay for the name as a prospect.
4. New to Year
There are other ways to talk about acquisition: some consider a buyer to be “new” for the entire fiscal or calendar year of acquisition. This is helpful for understanding not just the number of new buyers acquired but what can be expected from this group in a given year with regards to repeat buying activity. Note: This is different than LTV because not all will have an equal number of months to repurchases (i.e. Dec new buyers will have less than 30 days on file prior to the end of the calendar year). Each metric has its own purpose. Analyzing new buyers for the calendar year will allow a marketer to understand what proportion of their business is coming from new buyers and their repeat activity in that time. This can provide a health assessment of your file. If 60%+ of your business is coming from first time buyers, you are hopefully a new brand else you may be in replacement mode. This may signal a need to focus on customer retention to ensure you are cultivating your existing buyer relationships.
5. New to the Season
The apparel industry is interested in knowing who is buying their Fall and Spring product lines. While they track new customers to the brand, they may also want to understand the number of customers who are buying for the first time in a given season as this is another measure of “stickiness” or increased customer relationship/value. There may be very good customers who shop in one season exclusively vs. those who shop in more than one. Both are important to measure.
6. New to the Corporation
For companies with multiple brands, you can measure new customers for a given brand and/or new buyers across all brands. Marketers may cross promote one sister company to another to expand total customer value. This is a cost effective means of brand level new customer acquisition. However, it does not change the total number of customer relationships for the corporate entity as a whole. Both metrics can be viewed as measures of success. The lifetime value of each new customer acquired for a given brand can have a larger impact when total spend is tallied across all brands.
So how many new customers do I have? It seems like a simple question. A short discussion is warranted to ensure all are in agreement as to what is to be measured and how.