Conversion Rate Optimization: The Small Business Owner’s Edition
It’s a mistake I see all too often…
A business owner has a page on their website advertising their services—but it’s not converting.
So what do they do?
They invest money into a pay per click campaign in the hope of remedying the problem.
They create a Google Adwords campaign, drive hyper-targeted traffic to the page and wait for inquiries to roll in.
And do you know what happens?
And not only are they left out of pocket; they’re left with a page that isn’t converting and costing them money.
The problem, though, wasn’t the traffic (or lack thereof). Rather, their page wasn’t optimized for conversions, to begin with.
Now I know what you’re thinking:
“But Josh, I’m a small-to-medium-sized business owner, I don’t have a multi-million dollar marketing budget to optimize all the pages on my site.”
Here’s the thing:
Neither do the majority of businesses who are crushing it.
The truth is, whether you’re in B2B or B2C; a plumber or a plaster; growing or downsizing, conversion rate optimization (CRO) is an essential requirement for any business.
And if you’re trying to maximize every visitor that’s visiting your site, it’s particularly important (especially if you want to become the type of business with a multi-million dollar budget).
While I’ve written about on-page conversion optimization before, in this article, I’m going to discuss how to optimize conversions for a product or services page for small-to-medium service-based business owner.
So, if you’re looking to convert warm traffic into more qualified inquiries, read on…
Step 1. Find failure points in your funnel
In order to optimize conversions, you need to know what to optimize. Specifically, what’s performing well and what isn’t. To do that, you need to gather data. Without it, you’ll make choices based on your gut, which in the long-term, will cost you time and money.
The best data to gather is a combination of quantitative data and qualitative data.
The former can be measured by numbers (e.g. the number of inquiries you receive). The latter can not be measured, and instead, can only be observed (e.g. the reason a prospect didn’t make an inquiry).
While quantitative data provides a good starting point, qualitative data offers more insights into why visitors aren’t buying your products or services.
Let’s discuss two affordable ways of gathering data:
- Google Analytics (for quantitative data)
- Qualaroo (for qualitative data)
1. Google Analytics
Google Analytics is one of the most effective tools for mining quantitative data.
Not only can you track how visitors are interacting with your site; you can identify which steps in your funnel they’re dropping off.
Fixing failure points in your funnel is crucial because it’s costing you money. Even a minor tweak can fix it (and potentially save you thousands of dollars).
Assuming you already have a conversion funnel set up in your Google Analytics, login and click ‘Conversions’ > ‘Funnel Visualization’.
This will show you what percentage of your site visitors convert and how many visitors are moving through your funnel (if you haven’t set up your conversion funnel, read this guide).
A typical funnel for a small-to-medium service-based business might look like this:
- Services page
- Thank you page
(Or if you’re using paid advertising, Services > Thank You page)
If there’s a rapid drop-off rate on your services page, for example, that obviously requires your attention, first.
Another factor to consider is average time on page.
If, for example, the average time on your services page is 1:09, you might want to set your qualitative survey to appear just before the visitor exits so you can get immediate feedback on why they’re leaving (see next section).
While Google Analytics tells you what visitors are doing, Qualaroo, tells you why they’re doing what they’re doing.
The goal, here, is to learn why visitors aren’t converting into inquiries by asking them directly.
For example, you might ask your visitors:
- “What’s your biggest concern with scheduling a free call?”
- “Why didn’t you complete your inquiry today?”
- “What else can we place on this page to convince you to schedule a free call?”
In his Definitive Guide to Conversion Optimization, Neil Patel writes:
It’s better to have too much data than not enough. It’s easier to identify patterns and find useful information with 100 survey responses than it is with 10.
As a rule of thumb, you want to aim for a minimum of 30 response (granted, if you’re running a high-traffic site, you want as many as possible).
As for what to do with your data, we’ll return to this in Step 3.
Step 2. Develop a hypothesis
Once you’ve gathered enough data, you need to identify any recurring patterns. For instance, if you’re surveying visitors on your services page, are there certain words or objections getting used over and over again? If so, you need to propose what might happen if you address those objections in either your copy or your page layout.
This is where your hypothesis comes in.
Similar to a scientist making an ‘educated guess’ as to what will happen in an experiment, you need to formulate a hypothesis based on your existing knowledge (i.e. what visitors are telling you) and observation (what’s worked for other business owners in your industry).
Before writing your hypothesis, you need to know
- What your conversion goal is
- What obstacle is preventing you from achieving that goal (your audience are telling you, remember?)
If you want to increase the number of inquiries on your services page, for example, you might have found (after surveying your visitors), a few problems with your page, such as:
- Your value proposition isn’t clear enough. Ask yourself, “Am I addressing my visitors’ pain points and goals as well as I could be?” If not, this needs your immediate attention.
- Your call-to-action is unclear. Only 22% of visitors scroll all the way to the bottom of a page. With that in mind, you need a web form visible above the fold, a large bold phone number and a compelling offer that gives your visitor a reason to enquire.
- You have conflicting calls for attention. Each page needs to have one clear purpose and one clear action that can be taken.
- You lack visible credibility. Have you been featured on an authority site in your industry? If so, you need to demonstrate this to your prospects. Authority is a persuasion trigger and more importantly builds trust in your brand.
Whatever qualitative feedback you received in Step 1 will provide a good starting point for your hypothesis (that said, your feedback won’t always be conclusive and often, you’ll need to make educated guesses about what to change by split testing. We’ll explore this in detail in Step 3.)
When it comes to formulating the hypothesis, itself, Dale Cudmore recommends putting your hypothesis in writing by completing the following framework:
“By [making this change (or these changes)], the conversion rate will increase because [problem it fixes].”
If the majority of your feedback tells you your readers didn’t know what to do on the services page, for example, you might hypothesize rewriting your value proposition will increase conversions.
Sometimes, a hypothesis can be as simple as tweaking a line of copy. In fact, when Michael Aagaard surveyed his audience and found they were too busy to read his free eBook, he added how long it would take to read the treatment copy and increased his conversions by 18.59%:
The more data you mine in Step 1, the better informed your decisions will be in Step 2 and the more likely you’ll increase your conversions in Step 3.
Step 3. Run an A/B test
A/B testing, or split testing, is when you run a simultaneous experiment between two pages to see which performs or converts better.
Once upon a time, A/B testing was reserved only for multi-billion dollar companies with multi-million dollar budgets.
Today, not only is A/B testing affordable for everyone; it’s one of the most effective ways to validate a change to a page and improve conversion rates.
To run an A/B, effectively, you need to create an alternative version of a specific page (e.g. a services page) and show, both, the existing design (the “control”) and the new design (the “variation”) to a predetermined percentage of your visitors.
Let’s suppose, for a moment, you’re running a pest control service specializing in termites and want to increase the number of visitors requesting a quote.
If your feedback shows your visitors aren’t clicking on your call to action (through a click map or otherwise), you could A/B a number of options.
In fact, that’s exactly what one Australian pest control service (Pink Pest Control) tested.
In Version 1, they offered a free quote and a free report:
In Version 2, they offered only a free quote (and replaced the image of the free report with that of a smiling man):
Guess which performed better?
If you chose Version 2, you’re right.
Not bad for a slight tweak in optin copy, huh?
While impressive, outcomes like this are not uncommon when you know exactly what to optimize.
(In fact, when clients work with Authority Factory, many are able to achieve similar increases in conversion—if not more.)
Examples like the above emphasize an important reality about rigorous testing: empirical data bridges the gap between what we assume customers want (e.g. a free quote and a free report) and what they actually want (e.g. a free quote).
It’s important to mention here as well the importance of split testing every step of your funnel. Just because you increase conversions on your services page, for example, doesn’t mean you increase conversions on your home page (i.e. visitors moving from your home page to your services page).
Ultimately, the more A/B tests you run, the more data you’ll have and the better your chances will be for optimizing conversions at each step of the process (providing they’re relevant to your conversion goal, of course).
Step 4. Analyze your results
So far, you’ve gathered data on your customers to determine why visitors aren’t buying from you and ran a series of A/B tests to identify what converts best. The final step is to analyze your results.
Your results need to be statistically significant. This means the margin of error, or likelihood that your results are merely chance, is low. In general, the larger the sample size, the smaller the margin of error.
This is why many experts recommend running all your tests until there is a 95% chance of beating the original version (like the above example) and, depending on the level of traffic your site receives, running your test for a minimum of seven days (to account for traffic variations).
Once you have a clear winner, that becomes your control page which you can then continue to improve on. Returning to our previous example, if removing “free report” from your opt-in copy increases conversions, you might experiment with where the opt-in is displayed to further improve conversions (e.g. above the fold, against, under the fold, against, the sidebar, and so on).
While tweaks like these appear minor (and perhaps even insignificant), over time, they will aggregate to produce noticeable results in your conversions—and eventually revenue.
To ensure you haven’t made any mistakes that might have skewed your results, make sure you check it against Google Analytics, too. If there has been a change there as well, you can rest in the knowledge your results are accurate.
Unless you’re making a living through referrals, the lifeblood of your business comes from inbound inquiries generated through a contact or services page on your site.
And while it’s easy to believe a set-it-and-forget-it approach works fine, it’s imperative you continue to optimize your page as your industry changes (your competitors are doing it, after all, so why wouldn’t you?)
With services like Optimizely, it’s never been easier to create split tests, analyze your results and make improvements. That said, you need to carve out time, even if it’s once a month, and make it a priority.
When you approach CRO like a scientist, there are no failures—there is only feedback.
What have your experiences with CRO been? Leave a comment below.