Discounting 101: Designing Tests to Uncover Hidden Patterns

Discounting 101 is a series on all-things discounting strategy, based on the ebook, “The Ultimate Guide to Discounts” available for download now.

Discounting really works when done right. As we said in the introduction to this blog series, discounting can be your best friend or your worst enemy depending on how strategically you use it.

The secret sauce for discounting success has two main ingredients: the right data and the right tests to analyze the impact of your discounting strategies on various consumer segments.

Testing Is A Way of Life

Marketers who get the best results from testing know it’s not a one-time thing: Testing is a way of life, and it should be an intrinsic part of your company culture. And, much like brushing your teeth, it’s not really effective unless it’s done frequently and thoroughly.

Through careful testing you can see, for example, how different discounts perform across various platforms and determine which approaches garner the best response on mobile apps.

To get the most useful feedback, you’ll need to test for short-term results as well as long-term results. We’ll take a look at both types of testing here.

Short-Term and Long-Term Tests

Short-term tests are good for answering certain kinds of questions, such as, What kinds of discounts are my users responding to? Which channels are we reaching them on?  Short-term tests are also good for finding out what types of offers lead to immediate conversions and how much they increase average order value in the short term.  They’re also good for determining which techniques maximize your revenue per email sent (RPM).

For maximum impact, however, you absolutely must look at the downstream impact of discounting strategies. That’s where long-term testing is critical.

Let’s say you want to test the impact of a free-gift-with-purchase offer. Using cohort analysis, you can measure the efficacy of the strategy and determine if it’s more likely to drive a repeat purchase at full price, or lead users to make one purchase and churn. The idea is to test multiple discounts and/or offers to observe the long term impact on customer value. By segmenting your customers into individual discount cohorts, you can see which approaches – for example, purchase floors, mystery discounts or free shipping – work best in the long run.

Long-term tests are good for answering a number of questions, such as, which discount level results in the biggest boost to a customer’s lifetime value? Which type of discounts are most effective in different situations, such as Welcome series or a cart abandonment series? Which types of discounts increase revenue most – a welcome message series or a shopping cart abandonment message series? Long-term tests are also effective for determining which discounts encourage purchase frequency, how best to tie them in a message series or exchange, and the best time to introduce a discount.

Testing for short-term and long-term results will give you the best chance for improving your success today and insuring your viability in the future.

Coming up next week, we’ll discuss consumer-driven and market-driven discounts – and how to use each to best advantage.

Steve Wood, Optimization Analyst at Sailthru