Tactics of Microtesting
Brief Introduction to Pay-Per-Click (PPC) Advertising
For the purposes of all of our discussions and to avoid having to address the nuances of multiple ad platforms, "PPC" will refer to Google AdWords, as it controls about 70% of the PPC advertising market.
One of the most useful byproducts of our use of PPC for microtesting is that there is a stunning amount of information and tutorials available for all levels of experience, and essentially anything you need to accomplish. You will quickly see as you microtest that there are many, many PPC consultants and experts out there who do nothing but test and refine campaigns for ecommerce conversion and sales. While our application is a bit different, know that if you use this technique for testing, external resources are ample and easily engaged.
So, please know that there is a pretty significant difference between riding a bike and riding in the Tour de France in regard to the art and science of PPC... but for our purposes, I hope to demonstrate that someone with little experience will be able to set up initial microtesting quite quickly. Please watch the following 3:53 video.
Video: What is AdWords? (3:53)
A Note on AdWords Keyword Tool
There are myriad short, step-by-step videos on how to get started in setting up a Google AdWords account when you are ready. It takes about three minutes to get started.
While we will be seeing quite a bit of AdWords and we will be mocking up keywords and test designs for this week's assignment, we will not be setting up live AdWords accounts in class.
Here's why:
The AdWords Keyword Tool and other research tools used to be freely available online, until Google required you to create an account to access them. That is usually no problem, but Google no longer allows you to create that account without entering a valid Credit Card. Although you can set the account to not make any charges, I am not comfortable asking you to do so for class.
Happily, for our discussion, we can emulate about 90% of the core function of the Keyword Tool (and more) with SEMRush, an excellent package of research tools with fantastic analytics and trend data to help in decision making. Most importantly, it also offers a freely accessible trial. So while it will not provide the direct tap into Google ad pricing and search volumes AdWords would, it is more than ample for our purposes.
I just wanted to be clear there about the disconnect of talking about AdWords, but using SEMRush for research. We will each set up trial accounts for this week's assignment, but the Pro trial is only active for 24 hours, so you will want to delay setting that up until you are ready to begin your assignment.
Framing the Experiment
For the sake of this example, let's imagine that we have decided to explore one of the more straightforward strategic paths we proposed in Chapter 9, "The Lean Operation." Here is how we defined that path:
The map features three central ideas representing positive aspects of using native grass X.:
- “I mow less”
- “No fertilizer needed”
- “Grass grows slowly”
In this case, our goal is to understand if we can dip our net into the stream of people interested in and currently searching related topics/keywords to see what our conversion model could look like. In essence, in this microtest, our first step is to see if the market is interested in our most simplified proposition, and part of the beauty of microtesting is that we may have many tests of different executions on the same path and different paths running simultaneously.
The PPC Ad: Our initial "Escalator Pitch" to Test the Proposition
If you have ever heard of crafting a 30-second "elevator pitch" to effectively pitch a new offering to a prospect, you could think of what we are creating as closer to an "escalator pitch." Having 30 seconds to lay out our proposition on the web is a luxury we do not have, and we are realistically closer to the time we would have to talk to someone passing us on the down escalator while we were riding the up escalator.
At the highest level, we have to condense the most important "hook" of the strategic path into an ad totaling 130 characters. 35 characters of that is the URL you are linking to, so, as for usable message space, we are looking at a scant 95 characters to depict our proposition.
This may sound intimidating, but here are a few elements playing significantly into your favor:
- When your ad displays, people are already searching for information on a related topic, so they are already primed.
- You do not have to get it even close to "right" the first time, which is part of our testing.
- Because it is not an old media "Launch," you can change the ad in 30 seconds, 24x7x365.
- There are no high cost/risk factors.
- 95 characters is more than you think.
For the sake of testing "The Lean Operation" strategic path, let us suppose we want to test the initial viability of three test propositions.
Test Proposition 1: "Tired of Mowing?" In this test, we will actively pursue people shopping for more conventional lawn supplies and attempt to "intercept" them and gauge interest around the "Grass grows slowly" and "I mow less" concepts in the path.
Test Proposition 2: "Savings/spend calculator" In this test, we will again intercept those searching for more conventional lawn supplies. This time, we will call out how much the average home spends on lawn products annually, and what they can save by converting to Native Seed X. We will personalize the message by creating a calculator that will allow the homeowner to enter some basic inputs and get a realistic savings number. This proposition is centered around "No fertilizer needed" and "I mow less."
Test Proposition 3: "Better seed" As a bit of a control, instead of intercepting those with "conventional lawn" interests, in this proposition, we will attempt to sell the prospects of Native Seed X to those already actively searching for native seed. We could consider this as a bit of a counter-strategy to the other two, as the size of the market actively searching for native lawn seed is likely minuscule as compared to conventional lawn products (we will be able to quantify this in a moment). This proposition is centered around "Native seed."
From here, we would go about writing the actual PPC ads for each of the three test propositions in AdWords. Now, of course, we are not going to be the only advertiser in the space, which is also exactly what we desire for the test: to gauge how our proposition performs not in a lab setting, but in the real world, alongside competition.
Defining Keywords
The ads themselves are static, and so we must select those keywords which are related to the content of our ads to determine when they will appear. Almost in the sense of the Cognitive Map itself, we want our ads to essentially parallel when someone is searching for information related to the selected path (i.e., staying on our strategy). This, in essence, is what provides the revealed preference testing. We are not performing a mall intercept survey, or asking random groups of people online... we are placing our proposition in front of those who are actively engaging in the topic and who may be actively looking to purchase products with *real* money.
We would select our keywords based on both our learnings through research and tools to help us make informed keyword decisions in regard to quality and traffic, which we will examine in the next topic.
For "Tired of Mowing," our keywords could be centered around high traffic terms we would want to intercept like "lawn fertilizer," and perhaps we would test lower traffic terms like "mow less" or "low growth lawn."
The keywords for "Savings/spend calculator" could also be similar, but could also perhaps extend to "lawn savings" or "fertilizer coupon" to try to appeal to those who already show a desire to spend less on lawn products.
"Better seed" keywords could be more closely related to the seed itself, as this test is for those already searching for native seeds. "Native lawn seed," "North American grasses," and the like would be our keywords here.
Creating the Landing Page
A landing page, by definition, is the page someone "lands" on after taking an action. Overwhelmingly, that action is clicking on an ad.
The goal of the landing page is to "continue the thought" of the ad, and to quickly express the proposition and urge the visitor to take the desired action. If the PPC ad itself was the "escalator proposition," the landing page is the "elevator proposition," as we may be designing for 30 seconds of attention as opposed to 6 seconds.
Landing page design and high-level optimization is, in and of itself, a science. There are literally thousands of people who do nothing but shift elements of a webpage around, test colors, and revise messages to gauge how it may change response and purchase behavior. In our case, because we are simply looking for "signs of life" in our propositions and to begin to understand which may rise to the top, we do not need stunning levels of landing page refinement like an Amazon would.
What we do need is a landing page which we believe expresses the proposition, and has a measurable call to action clearly on the page. Whether that call to action is a pre-order, a catalog request, a sample request, or an order of the product itself, we want the prospect to take some "next step." Ideally, the next step is indeed purchasing the product in question, but given that we may be in pre-release, an "email me when this product is available" may be a logical replacement.
The proposition itself may be expressed in video, image, text or a mix of all, or, in the case of a concept like the "Savings/spend calculator," a very simple and straightforward calculator. Again, all we are looking to do is to provide that 30 seconds of proposition and interest to engage the visitor and make them take the next step.
A Brief Example of How the Pieces Work Together
Please watch the following 6:02 video.