6: Landing Page Experiments
Now we're going to test your value proposition and channels in the wild with a landing page experiment.
Don't worry if you have no technical expertise. There are so many solutions already out there that make it incredibly easy to run landing page tests that require little or no coding expertise.
Before we jump in, let's make just look again at what we're doing here and why.
Recap, so far you've:
- Checked your assumptions
- Interviewed some target customers
- Analyzed the competition
- Researched the market
If any of these four steps stopped you cold in your tracks... well, good! That's the point, right? We're trying to save time eliminating bad ideas before you get too much into building a product.
And if you're idea didn't seem like it would work, don't take it personally. There are so many ideas, and most of them are awful! It's not a signal that you're a bad person or anything. Just grab your next idea, and go around the cycle again. Have fun with it!
Now, we are going to test whether or not your value proposition resonates with your target customers. Your landing page experiment is going to target two specific sections of your business model; the value proposition and the channels. That's sort of a compound experiment, because your testing your value proposition in context.
A hypothesis that you might use for this experiment might be stated as:
"When users land on this page and see this value proposition, they will click the call to action button at a rate of X% of total page visitors."
What threshold you set depends on your business, specifically your cost of acquisition and your unit economics.
The structure of the experiment goes like this. We will look at each step in more detail below.
- Establish a clear call to action that you can measure, such as emails captured in a signup form.
- Design a landing page that clearly articulates the core value proposition.
- Get traffic to the page in sufficient amounts to get a statistically relevant result.
First, we need to come up with a clear call to action. Since we haven't built our app yet, it is not possible to use signup for the app as our metric. For now, we are just going to use sign ups on an email list (or click to call) as our metric.
To make this work, you will need to ensure that you have a tool like google analytics set up and that you are measuring those clicks. You'll also need an actual email list tool, such as MailChimp, that you can actually send them into. Both of these are very easy to set up. But if you get stuck, email me, and I'll talk you through it.
Second, you need a landing page. You can hire a designer to build one, or you can use a tool like Unbounce (or one of their many competitors) to build it. Whatever you're comfortable with. Here I'm focused more on the validation process than the technology. But if you need help with this part specifically, let me know.
Third, you need to push enough traffic to the page to validate your hypothesis statistically.
About how much traffic is that?
Well... it depends. Typically you need at least a hundred actual conversions to get a relevant signal. That means that you need 100 people to actually click your call to action. But also keep in mind, this test isn't going to be all about statistical significance. You're just testing whether or not there is a strong enough signal to develop your idea just a little more.
The quickest way to get traffic, is probably just to buy it.
I spent $200 on an Google Adwords for a campaign recently and averaged $.15 per click. That got us about 1300 site visits. That was plenty for a test like this. You can also try Facebook or Twitter ads if that's your target channel. The clicks are more expensive, but the targeting for the ads is good enough that the quality of visitors may be higher.
Once you have your landing page built, with your trackable call to action, launch your campaign in your channels and see what you get. In many cases, you won't get any conversions at first. That doesn't mean that your idea is bad, just that the landing page experiment did not work. Remember, according to the scientific method, you can't prove a hypothesis--only disprove it. If you're landing page got hundreds of views, and only 1 or 2 signups, you may not have articulated the message properly, or maybe the channel was all wrong, or maybe the page.
Of course you want to attract customers. But you also want to stop wasting time and money on a bogus value proposition if it isn't going to resonate. If a failed landing page experiment can do that for you, well great.
In the next and final step of the series, we're going to review what we've done so far, and I'll give you some concrete methods for evaluating whether to kill, pivot, or persevere.
Next: Kill, pivot, or persevere?
I am here to help.
Don't feel ready to go it alone? Give me a call so I can help you.
Prefer email? That's fine. I am reachable at mcafee . sam at gmail . com