Tips for Experiments

• Make sure your experimental content is different from existing content: If your content is too similar, it is less likely to affect customer behaviour, and you may not be able to confidently determine a winner.
• Let your experiment run for the entire duration: Even if early results are promising, they can be misleading. If you end an experiment early based on those results, you increase the chance of overestimating the impact of your experiment or even selecting the wrong version.

Ideas for Experiments

Here are some things you may be interested in experimenting on. But do not restrict yourself to these ideas – use your own judgement to determine what to test.

• Reducing product title length to under 100 characters to reduce noise and encourage more customers to visit your detail page
• Revising your product title to include your Brand name
• Make your product title more descriptive to appeal to attributes of interest to your customers
• Add creative product imagery to help your product stand out from the competition
• Adding high-resolution imagery to your content or refreshing your existing imagery
• Using exclusively lifestyle imagery or standard product imagery
• Highlighting one set of features versus a different set

Experiment Results

By making the most of your results, you can maximise the total impact of each experiment.

• Do not check your results too frequently before an experiment ends: Interim results can be misleading.
• Run more experiments: You can continually improve your content strategy by testing on different ASINs, changing your content hypotheses for the same ASIN or testing during different seasonal periods to understand how different types of customers approach your product.