How many times have you run an experiment and not gotten the insights that you wanted? Or not known what to do with the pile of data you captured? How many times have you looked back at test results and wondered what you were thinking at the time? “What were we trying to do again?” “Why did we decide to go with X?” Well you aren’t alone. One of the biggest pain points business creators experience is not being able to capture, store and share all of the information that surrounds a test. The context for the experiment, the decision making process and the “why” can often get lost in a sea of data. This pain point gets exponentially worse when you try to scale testing across teams and over time. Startups and entrepreneurs are moving so fast they don’t often stop to capture their thinking. Corporate teams are often so big and siloed, that the context of their experimentation isn’t standardized or centralized. The loss of institutional knowledge for both startups and corporate teams is extremely costly and more importantly, the wins discovered aren’t able to be compounded.
Last year I created a canvas on Miro that I could use to capture all of my thinking around each experiment. I’ve been using it ever since and have found it really useful in a few different ways so I thought I would share it and get some feedback. Using this tool, I can go back and remind myself of what I was thinking when I ran a certain test. Or, on longer tests, I can revisit and remind myself of what the goal actually was or what business outcome I wanted to achieve. Secondly, it has helped me standardize the information I use to create tests and helps me think through everything a little bit more without slowing me down. It’s broken into three sections; Context, Experimentation and Analysis.
This section is fairly straightforward. It starts with labeling the date and team working on the project. Then moves into the top section that captures why we are running the experiment.
First you can capture your initial thoughts under “Our Insights”. Maybe you saw a piece of data that was interesting, maybe there was a request made or maybe you just want to learn something. Capture it all under insights, one thought at a time.
In this example the Events team wants to convert more of the social media traffic into event registrations.
Next is “Our Vision”. If your Insights answer the “Why?” Vision answers the questions “Who cares?” “So what?”. This can be an incredibly powerful tool to help communicate to people the impact that the idea could have. It could be to support a business objective, it could be to advance a product or it could be trying to relieve an internal pain point. Regardless, capture the impact here.
The Events Team here knows that if they get more people into their webinars they can drive more leads to sales and help the company grow.
The final section of the Context section is “Our Customer”. This is where you can stop for a minute and think through who you are testing this with. Be as specific as you can because this section can really help as you design out your test.
The Events Team has decided to target their enterprise buyer persona and is explicit about ones that come from social media.
Once you have the Context Section filled out you can move onto the experiment itself, which is where most people start. First the test design starts with a simple hypothesis broken into 5 sections.
The first step is to simply capture what you believe, your hunch or idea.
The Events Team believes that custom landing pages for traffic from Instagram will convert at a higher rate than their generic webinar signup page.
The second section is your Leap of Faith. This can be a bit trickier but adding a “because” after your belief is a powerful way to capture your key assumption. Think about your Leap of Faith as “if this isn’t true then the whole thing falls apart” and capture it.
The Events Team thinks that people will convert at a higher rate because the content on the page will be more relevant. If that’s not true then it might not work.
Next is the test itself. How, when and where you want to run the test, what you will measure and what the barrier for success will be.
The Events Team is going to build 2 landing pages on Unbounce and drive traffic using paid Instagram ads. They will run it for 2 weeks or until they get 1,000 people to the pages. They will measure conversion rates for the pages and set a goal of 5% or higher.
Once you get to this point you can stop and go run your experiment. Refer back to the canvas when you get lost along the way or you start to get overwhelmed with options and data as the test is in progress.
When the test is over, return to the Business Testing Canvas and capture the results in the Measure section. Start with the metric you outlined in the design phase but also capture any interesting results or data points you also collected. Don’t ignore valuable information just because it wasn’t your primary goal of the test. Remember the purpose of this canvas is to capture a 360 degree view of what happened so someone who has no knowledge of the experiment could pick it up and see the whole story.
The Events team captures the conversion rates for the control page and their new Instagram focused page as well as some info they didn’t expect. They capture the performance of various ad types as well as a nice bonus of newsletter signups that they didn’t expect.
Next Evaluate. There are some thought starter questions on the canvas but use this section to capture the “so what” for the data you collected. What do those numbers tell you? Did anything go wrong? Was there anything unforeseen that happened to capture? Sometimes you realize after a test that what you measured didn’t actually answer the question. That’s okay, just capture all of your observations and post-test thoughts.
So the Events Team didn’t reach their goal however they realized after that 5% increase may have been a bit too ambitious. They saw more than a 2% increase which resulted in enough extra webinar signups that they consider it a success. They also captured that video ads worked the best.
Finally, maybe the most important section, Action. “What are we going to do next?” is a question that often stalls teams and projects. This is where you can think through if you want to proceed with the project, go back and test again or even kill the idea. If you decide you want to re-run the test, no need to make a second canvas. You can just colour code your stickies and keep it all in one place.
Lastly, the decision making of the Events Team is captured. They have decided to start creating channel specific landing pages going forward and are going to pass the insight along to marketing about the success they saw with video ads.
Finally, they are going to re-run the same test on Linkedin to see if what they learned from Instagram can be exploited elsewhere, compounding their wins.
The Business Testing Canvas has been a great tool for me to organize what I am working on and if you’d like a PDF version feel free to DM or reach out to me directly.