Funnels are everywhere in your customer acquisition functions – email, software trials, the checkout sequence. Even in the use of your core product.
Here’s something that’s true about all of your funnels. They leak. A lot. Many people enter the funnel. Only some emerge from the other end.
If you can slow the leaks in your funnels, even a little bit, two good things happen.
You lower your customer acquisition costs. And you increase your profits.
Poor funnels kill companies. Because when you think of when you are trying to influence people’s behavior through multiple stages of steps, if you execute on this poorly your company will not succeed. You go home crying and your employees don’t have jobs anymore.
Who you gonna call?
Your software engineers can help you plug your leaks. Tasks that marketers find intimidating and insurmountable will be easy for your engineers.
But should you take engineers off product development to work on customer acquisition?
This is just an economic question. Look at projects that your engineers are working on. Now look at the projects that have the lowest expected return on investment. Are any of these projects “nice-to-haves” that will benefit a small proportion of customers and that could be postponed briefly?
Now look at all your marketing funnels and their conversion rates. Are there any opportunities to improve these funnels that would require small amounts of engineering time and that would have an immediate impact on your bottom line?
In most companies there is so much opportunity to improve their funnels that it’s like gold nuggets lying around on the ground. All you have to do is walk around and pick them up.
How do you go about improving your funnels?
How exactly do you go about plugging leaks? McKenzie suggests four steps.
- Describe the funnel. What are the main steps in the sequence?
- Measure the conversion rates of each step. Software products that people use to measure conversions in funnels include KISSmetrics, Mixpanel, Hubspot, and Google Analytics.
- Optimize it. Look for the steps where the conversion rates are lower than you think they should be. What is the user experiencing that could be improved? Is it too complicated? Are the instructions vague? Too little value to cause the customer to go to the next step in the funnel?
- Profits — When you increase the number of people who flow through the funnel, profits immediately follow. 🙂
What if you don’t know which change will improve conversions?
You have a lot of elements on your site where you have no way of knowing whether they are optimal.
Which message will people respond to most favorably? What positioning will produce the best outcome? How do people respond to your pricing?
The most common response to these questions in online marketing today is a combination of your best guess plus the opinion of pundits with a little “let’s look at what everyone else does” thrown in for good measure.
But these are all testable questions. At McKenzie says, “Try two things. Test them. See which one actually works.”
You can use the A/B library in your programming language or build your own A/B testing tool. Or go with one of the visual tools like Visual Website Optimizer or Optimizely.
Where should you run A/B tests?
Put most of your A/B testing efforts in places that are early in the funnel, before people feel strongly committed to continuing. These include your Home Page, Landing Pages, your Pricing Page, and the Shopping Cart.
You will often be surprised at which option wins the A/B test. Your predictions of user behavior will usually be wrong. Don’t even try; just look at their actual behavior.
Wait a minute. How will the engineers feel about working on marketing tasks?
Good point. McKenzie is an engineer and talks about the engineer’s view of the world:
One of the reasons I got into engineering is this simple and powerful idea, that math and science always work. Many engineers have emotional attachment to this one. We don’t have an emotional attachment to marketing.
For you marketers that haven’t cottoned onto it yet, the stereotype of marketing in the engineering field is that marketing is basically witchcraft. Eye of newt and toe of dog that goes into the pot. We have no clue how anything works. And it’s like vaguely evil. I mean, you are selling stuff to people. It’s okay when people buy our stuff, but when they buy other stuff, that’s kind of unnatural. We’re just creeped out by it.
What I’ve come to realize after running my own business is that marketing is not so much witchcraft as it is about changing people’s behavior. This is not something that is totally unknowable. We know we can change people’s behavior, we do it all the time. It can be as simple as changing our tone of voice.
We change people’s behavior in engineering too. That’s why we build products. I want you to change from a manual process to this better process that we have. That’s why Steve Jobs can turn around the whole world’s idea of how we use a cellphone.
As we get into this, we think, wow, marketing is really sort of an engineering discipline.
So, I think the answer is that if you ask engineers to work on marketing tasks that are based on math and science, you are likely to be successful. Fortunately, many online marketing problems are well-suited to measuring and testing. If you turn your engineers loose on them, you’ll start plugging those leaky funnels.