How to Divide & Conquer Your Sales Funnel

Before making a purchase, a potential customer has to learn about your product or service, decide it’s right for them, and finally decide to pull the trigger. Your job as a marketer is to help guide potential customers through this process, commonly referred to as the sales funnel because it’s wide at the top and narrow at the bottom.

The more people you bring in at the top of your funnel, the more chances you’ll have for successful conversions at the bottom of it. And the more you can optimize each step a lead takes along the way, the higher percentage of conversions you’ll get.

We’re going to look at how to think of your sales funnel in stages, and how to boost traffic at each one of these stages.

There are a couple different approaches you can take to divvying up your marketing funnel, but the simplest way is to break it into three sections - top, middle, and bottom. You can further divide each of these sections if you’d like, which we’ll discuss as we go.

It’s also important to think of each step of your funnel as relating directly to your website. So we’ll tie each section of the funnel to a particular page of your site, as well as relating it to the customer mindset at each stage. We’ll also suggest experiments to try boosting traffic in each step of your funnel.

Let’s Talk About Testing

Before we get into dividing and optimizing your sales funnel, let’s go over some basic testing principles.

The first thing you’ll need is a tool for creating the test on your site - something like Optimizely or Visual Website Optimizer. This will make it easy to set up your test and split your traffic correctly.

‍

NOW FOR A HIGH-LEVEL OVERVIEW OF THE RULES OF A/B TESTING:

1. Only test one variable at a time — This is the golden rule of testing - never change more than one element at a time. For example, if you’re changing a CTA, that’s the only change you want to make. Messing with headlines or images at the same time is going to make it impossible to tell what effect your new CTA is having. It’s a great idea to test every element of your site - just do it one piece at a time!
2. Always have a hypothesis before testing — When you install your A/B testing tool and get everything set up, it may be tempting to start making changes based on feeling or personal preference. . Resist this temptation. You should always head into a test with a hypothesis in mind - do your research, find things that have worked for other sites, and test them on yours. It can be as simple as: “I believe using green buttons for all of my CTA’s will increase my conversion rates, because I just read an article that suggested it would.”
3. Test towards a specific goal — Plan your tests to hit a specific goal, such as increasing revenue by 10%. This will help guide your experiments and keep you working towards overall business objectives. You can even string together multiple tests in order to hit bigger targets - like raising conversions 20% overall by finding four areas to raise it by at least 5%.
4. Pay attention to sample size — Sample size is vital to the validity of your test. Most A/B testing software will have some sort of built-in algorithm for determining when you’ve hit a substantial sample size, so pay attention to those. If you’re winging it, there’s no hard and fast rule for what constitutes a valid sample size other than “more is always better,” but testing at least 1,000 respondents is generally a good starting point.
5. Record your results — This may seem obvious, but it’s worth saying: keep track of all your experiments and results! You’ll be amazed at what you can learn from past test results, and being able to reference all of your tests to date.

‍

Top Funnel

This is the widest part of your funnel, where your main goals are generating leads and building awareness of your brand.

The top of your funnel ties directly to your homepage, and also to your main blog landing page. You can improve traffic to the top of your funnel through content strategy, SEO tactics, and paid advertising. Basically, anything that is going to get people to visit your website.

At this stage, potential customers are just finding out who you are and getting a rough idea about your product. Your job here is to get your name out there and to bring in as much new traffic as possible.

‍

A TEST FOR THE TOP FUNNEL

We’re going to assume you have some sort of online ad strategy working for you, whether it’s banner ads, social media advertising, or what have you. If not, no worries - you can just start one for this experiment and run two different variations as your test.

For this experiment, we’re going to play around with ad copy. You’ll create two nearly identical banner ads, the only difference being the copy. This will let you see exactly how much impact the copy alone is having (because you’re only testing one variable at a time). One of the ads will run a standard value prop statement, and the other will use customer testimonial.

Here’s a table showing all of the elements of our test. We recommend setting up a table like this whenever you run an A/B test, so you can refer back to it when devising future experiments.

Hypothesis: Customer testimonials in our ads will help drive more traffic to our site. (See Tip #5 in this article.)

Variable to Test: Ad copy

Variable A: Current ad copy: “See what makes Alpha Brand the best!”

Variable B: New ad copy with testimonial: “Alpha Brand saved me hours of work each week. It’s incredible!” -C. Jones, customer

Goal: 10% increase in click-through rate of ads.

Minimum Sample Size: 2500 ads served

‍

Middle Funnel

Depending on whose articles you read, you’ll find a few different definitions of the middle of the funnel. We like to think of it as someone who has heard of your brand and is now trying to learn more about it.

In terms of your website, this means a visitor who, at the very least, has progressed beyond your homepage or blog landing page. Maybe they’ve browsed your store page, read a blog article, or even signed up for your mailing list.

That’s right - by reading this article right now, you’re in the middle step of our sales funnel. Hope you’re enjoying it!

It may help your marketing efforts to break the middle of your funnel into two pieces - creatively called the top-middle and bottom-middle. As a simple distinction, think of users entering the top-middle when they leave your homepage for a deeper piece of content, but not reaching bottom-middle until they complete some sort of action - providing an email address, putting something in their cart, requesting a demo, etc.

Experiments for improving conversions in the middle of your funnel generally relate to tinkering with the content on your landing pages.

‍

TESTING THE MIDDLE OF YOUR FUNNEL:

For testing the middle of our funnel, we’re going to play around with the images on our hypothetical product page. We suggest using custom photos from a professional photographer for this test, but if your budget or timeline make that impossible, simply try updating your stock photography.

Here, we’re changing images on a page and measuring success by the percentage of visitors who click the CTA on that page.

Hypothesis: Custom images will outperform stock photos. (See this study)

Variable to Test: Images on Product Landing Page

Variable A: Current Image: Stock Photos

Variable B: New Images from custom photo shoot

Goal: 20% increase in conversions

Minimum Sample Size : 5000 total conversions

Results: TBD

‍

Bottom Funnel

This is it - the end goal - getting a customer’s money. The bottom of your funnel is your purchase page, though a customer isn’t technically out of the funnel until they’ve completed the purchase.

Everything you do, starting at the top of your funnel, is leading up to this event. At this point, the customer has discovered your brand, researched your product, and is ready to make a purchase decision. So how do you help them do it?

One experiment for the bottom of your funnel is to try offering a Free Trial. Depending upon your product, billing platform, and other factors, this experiment could prove to be a bit more complicated to implement than the others in this article. If that’s the case, simply use this as a template and try testing a new variation of your call to action.

Hypothesis: A free trial will boost overall conversions. (See Experiment #3 in this article)

Variable to Test: CTA / Presence of Free Trial

Variable A: Current CTA: “Sign Up Now”

Variable B: New CTA: “Start Your Free Trial”

Goal: 10% increase in conversions.

Minimum Sample Size: 1000 accounts created

‍

Back to the Top

After a customer completes a purchase, he or she goes back to the top of your funnel. Ideally, they bring a few more people there with them through word of mouth referrals. You can help encourage this by keeping customers in a mailing list, and incentivizing them to return to your site, leave reviews of your product, or send discount codes to friends. All of this will help bring new traffic back to the top of your funnel, where the whole process can begin again.

How are you currently testing your sales funnel? We’d love to hear the types of experiments you run, and the results you see. Share your progress here in our comments!

‍