Think big, get results

Katherine Boliek
Typed Out
Published in
4 min readAug 20, 2019

Test and learn: How to take big swings and still get results

Failure is scary. People avoid it. So it’s not surprising that at work, we want to do things right the first time. The challenge though, is sometimes we can rely on our experience or bet too much on our creativity. Today, the playing field is changing too rapidly to assume the way we did it last time will work for this campaign or this audience. So how do you take big swings while providing the best results for your clients?

At Archetype we believe that data-driven programs drive the best results.

We apply a test and learn mindset to guide how we make decisions. We act as scientists running experiments. Ask a question, set aside a control group, make a hypothesis and see what happens.

We’re more open to new opportunities when we embrace some failure, and we develop more effective campaigns for our clients when we base our decisions on real data.

Here’s our formula for a successful test and learn mindset:

  1. Identify your goals, audience, and how you’ll measure success.

First and foremost, focus your test on what will help you achieve your goals. Do you want to see what will drive the most downloads or are you looking to get an audience engaged?

Next, consider what you want to know about your audience. Where do they like to get their information? What matters to them?

It’s also important to know what you’ll be looking for in the results. If you want to drive downloads, do you care about engagement rates? This will keep you focused when reviewing the data and help you better understand your project success (or failure).

2. Make observations about audience behavior.

Use whatever you already think you know about your audience to build your test. If it’s an audience you’re familiar with, try a test that compares how you typically communicate (e.g., on Instagram, through emojis, etc.) with something brand new like a move to Twitter or a change in tone of voice. You may be surprised.

When you have a new audience you can look at industry or platform trends to see if those make sense for your audience. Assume nothing, test your gut feeling rather than relying on it blindly.

3. Keep your experiment simple.

Answer one question per test. Only change one variable per test. You won’t be left wondering “Was it the different call-to-action, copy, or the creative?” You’ll know that “Download Now” was more compelling than “Learn More.”

4. Give the experiment space to be successful.

Depending on what you’re testing, you may need to consider the length of the test or the sample size. If you’re testing a social media campaign, allow the platform time to optimize. If you’re running an email campaign, how many people received the email?

Also, make sure that nothing changes during the testing period that could impact the results. A clean test will give you clear-cut, actionable results.

5. Put the data to work.

This is really the learn part of test and learn. What insights did you gain with the test? If you’ve set up your test properly, you should know what you’re looking for. Was shorter or longer copy more effective? Longer? Let’s apply that moving forward.

Use your insights to inform your next steps. Think critically about where these insights apply. Did longer copy perform best across all channels or just this one?

6. Rinse and repeat.

Even though we typically designate a test and learn phase at the beginning of a campaign, we continue testing and learning afterward. If your first test proved that your audience is most engaged on Twitter, test what format performs best on Twitter next.

This can be applied across campaigns as well as ongoing efforts. We often take a flighted approach. If the campaign is eight weeks long, we’ll reassess our content every two weeks, optimizing based off our tests each flight.

We extend our test and learn mindset to how we approach all of our work at Archetype. Our goal is always to provide the best work for our clients in order to help them succeed. Test and learn allows us to experiment and let our team think big, like really big, so we can deliver bold, results-driven campaigns for our clients.

How do you inspire your team to think big and how do you leave room for failure?

Typed Out is an Archetype publication.

--

--