Set yourself up for online fundraising success this year-end with a free year-end fundraising course.LEARN MORE »

A Thanksgiving Lesson on Donation Page Optimization

Published by Nathan Hill

Thanksgiving Optimization Blog image

The centerpiece of the Thanksgiving meal is the turkey. And if you’re anything like me, you’ve probably Googled “how to cook a turkey” in hopes of finding a step-by-step guide on how to cook the best turkey you’ve ever eaten in your life.

With searches like this, you find all sorts of ideas and opinions that often give you conflicting information. And how are you supposed to know which “best practice” is right for you?

  • Ways to Cook a TurkeyShould I bake the turkey?
  • Should I brine the turkey?
  • Should I smoke the turkey?
  • Should I deep fry the whole thing?

After 10 minutes of being overwhelmed with articles like “25 Ways to Cook a Turkey” (yes, there are apparently 25 different ways), we’re left planning to cook the turkey the same way as always – like mom used to make it.

But what if there is truly a best way to cook your Thanksgiving turkey? How could we go about proving that one way is better than another?

The answer is optimization. Not only can it help you cook a better turkey, but it can also prove what works to convert more donors and raise more money on your donation page.

My Thanksgiving Hypothesis

To get started, I need a hypothesis. A hypothesis should be an idea you have about your donation page, email, advertisement, or turkey that could help improve performance. My hypothesis is this:

Hypothesis: A deep-fried turkey will be more enjoyable than an oven-baked turkey.

After defining my hypothesis, I need to convert it into a research question – something we can actually measure and answer with data. If you’re optimizing your donation page, you might look at total conversions. With a turkey, you might measure how many people say “Mmmm…”

But an “Mmmm…” could mean a lot of different things. So let’s go with something more concrete: the number of post-turkey-dinner-naps.

Research Question: Which turkey will cause more people to take a post-turkey-dinner nap? 

Next, I need to define my treatments. Which turkey cooking methods (or designs, copy, form fields, etc.) am I actually testing? In this case, I have my control and one treatment:

Control: Oven-Baked Turkey

Oven Baked Turkey

Treatment: Deep Fried Turkey

Deep Fried Turkey

Running A Valid Thanksgiving Test

Before I get ready to run my test, I need to make sure that I’ve considered any environmental factors that could skew my results.

If you’re trying to test too many variables at once (design changes, form fields, copy changes, etc), you’re going to have a hard time knowing what variable affected your results.

In this case, my results could be skewed by someone eating more mashed potatoes than anyone else. Or maybe having one too many glasses of wine. In the same way, if I change both the headline and the design of my donation form, how will I know which change caused more conversions?

To ensure I get a valid learning, I need to make sure that all turkey-eaters have the same Thanksgiving spread. My personal go-to dishes include:

  • Thanksgiving MealMashed potatoes (with brown gravy)
  • Real cranberry sauce (not the gelatin kind…)
  • Stuffing (more savory than sweet)
  • Green bean casserole (not because I like it, but because it’s tradition)
  • And pumpkin pie (made with Libby’s pumpkin)

You’ll also want to make sure you’re collecting your data properly. Be sure to define what constitutes a nap before hand. Is it 10 minutes, 30 minutes, an hour?

On a donation page test, you’ll want to make sure your analytics tools are properly tracking donations for your various treatments, and that nothing is skewing your data.

Once you’ve considered and eliminated all validity threats, you’re ready to run your test.

Cook the turkeys. Set the table. Feast.

Determining the Champion Turkey

As the results come in, you’ll want to make sure they’re valid. You’ll need to have a large enough sample size (people eating your turkey, or visiting your donation page) and a statistical level of confidence of 95% or greater. If this is too much for you to calculate on a holiday, we have a free experiment validator tool you can use.

After plugging in your results, you may realize that your sample size is too low. In that case, you’ll want to grow your email file for next year so you can invite more people to your Thanksgiving meal. We have a resource for that as well called 6 Ways to Grow Your Email File.

Campaign Donation Page TemplateAnd if you wake up from your Thanksgiving coma realizing that you could use these same optimization principles on your donation page to grow your fundraising exponentially…we have just the tool to help you get started.

Inspire more generosity this year-end season by crafting a high-converting year-end campaign donation page. Download your free copy of the Campaign Donation Page guide with 21 ideas that you can test.

About the author:

Nathan Hill

Nathan is the Optimization Evangelist for NextAfter. He spends every day working to help nonprofit organizations discover how testing and optimization can transform their marketing and fundraising, leading to greater impact and organizational growth. He is also a giant Star Wars nerd.