On Digital Fundraising: By Failing to Experiment You Are Experimenting with Failure

Experimenting has long been part of direct mail fundraising. Whether it’s testing different ask strategies, theme, copy, design, font size, freebie, personalisation, envelope colour, picture, prompts, non-financial ask, fundraising target or mail sequence, tests give invaluable insight to your donors and inform the strategy of future appeals. Now, with the rise of digital, experimenting is becoming increasingly commonplace in digital fundraising.

Implementing a Testing Platform

Digital fundraisers need to implement a testing platform to experiment effectively. This can be daunting and expensive. But it doesn’t have to be. Google’s product, Google Optimize, has a freemium model, which is more than adequate for the majority of organisations. And, it’s easy to implement, by adding just a couple of lines of code to web pages. It also seamlessly connects to Google Analytics, a web analytics platform that you’re likely to already have on your website.

Once you have a platform like Google Optimize installed it’s time to get experimental. Good experiments usually contain a strong hypothesis, measurable primary metric and are likely to gain statistically significant, actionable results. An example of a good experiment outline is below.

Hypothesis Visitors from social media are more likely to donate if the asks are lower.
Primary Metric Increase the number of donations from social media, improving overall channel revenue.
Actionable Results Yes. On average the donation page is viewed 1,000 times per week from social. Therefore, this test should reach statistical significance in a couple of weeks.

Recently at On Agency we ran a small test on our homepage navigation bar. The test’s objective was to increase visits to the blog, which we currently have labelled as “opinion”. After 23 days of testing, the results show, with 90% confidence, that visits to the blog increased by 50% when the title read blog. Therefore, changing the title to blog will increase website engagement.

Creating a Test Plan

Once you’ve successfully run your first test it’s likely that you and your team will collate a long list of digital tests that could boost the performance of your website. But with finite resources and time, you need a way of prioritising these tests. This is where a test plan comes in. A test plan outlines the test hypothesis, objective and includes four variables that help rank the tests; potential impact, ease of implementation, ease of testing and impact confidence. Each variable is scored from 1-10 and, by adding the scores together, the tests are ranked and prioritised. An example of a digital fundraising test plan is below.

charity digital fundraising test plan

In summary, testing is as important in digital fundraising as direct mail. After the initial setup and implementation of an experimentation platform it’s a good idea to develop a test plan to prioritise tests. This test plan doesn’t need to be static, it can be a living document that evolves with the outcome of other tests and your overall fundraising strategy.

If you need help with digital testing, or have a fundraising challenge, why not contact On today?

This article was written by Peter Longton, Head of Data & Insight at On Agency. Email, peter@onagency.co.uk. Tel, 01273 208 311.


Notice: ob_end_flush(): failed to send buffer of zlib output compression (0) in /home/onagency/public_html/cms/wp-includes/functions.php on line 4613

Notice: ob_end_flush(): failed to send buffer of zlib output compression (0) in /home/onagency/public_html/cms/wp-content/plugins/really-simple-ssl/class-mixed-content-fixer.php on line 111