Establish a baseline for A/B testing goals

1. List your business objectives in the first column of a spreadsheet to ensure they form the basis of your A/B testing goals.

For example, this could be an increase in sales for an ecommerce store or an increase in content consumption and social sharing for a publishing website.

2. Deduce key website goals for each stage of your conversion funnel from your business objectives and add them to the second column of the spreadsheet.

For example, for an eCommerce business, the funnel stages could be awareness, consideration, preference, intent, and purchase. The corresponding goals that you might aim to improve through A/B testing could be store traffic, product page views, add to cart rate, checkout rate, and purchase, respectively.

3. Define specific KPIs by identifying what you want to measure for each website goal and add these to the third column.

These will guide your A/B testing goals. For example, the KPIs for checkout rate would include bounce rate of the cart page, time spent on the cart page, and click-through rate of the Add to Cart button for an ecommerce store.

4. Choose a historical date range for baselining conversion goals and KPIs based on your average sales cycle and accounting for unprecedented and seasonal influences.

For example, if you own a SaaS business that has an average sales cycle of 3 months, to set a realistic baseline you would want to consider data for at least the past 3 months. Ensure the same date range is available across all the platforms and tools which capture your key business goals, conversion goals, and KPIs.  Use a date range of more than 30 days, as it would give you a clearer picture of your baseline metrics.  Make sure the data you use is not corrupted because of any past tests.

5. Export the data required for the finalized KPIs over the decided date range and add it to the fourth column.

Use analytics, marketing automation, CRM, or tools like VWO  to track your key metrics.

6. Collect all the data and add it to a single dashboard on a data visualization platform.

Ensure data sanity by removing outliers and accounting for randomness.

7. Use a tool like VWO to repeat the above steps for key segments based on customer parameters important to your business, like demographics or psychographics.

Tools like VWO help you to establish the baseline for your A/B testing goals while running the test, and compare it with the performance of the variation in real-time.  Calculating the baseline beforehand isn’t necessary, but is great to have so that you can prioritize tests.

8. Use a data visualization software such as Tableau to analyze historical trends in your data for each of your key segments.

Add the data to the fifth column of your spreadsheet.

9. Research industry benchmarks for your KPIs and add them to column six of the spreadsheet.

Conduct secondary research from market research reports or work closely with a consultant who can help you collect and analyze data from thousands of industry leaders for the metrics you’re tracking.

10. Condense all data into a single baseline for each of the KPIs and segments.