Troubleshooting Landing Pages
In this chapter:
• Understanding why users convert and why they don’t
• A/B testing to troubleshoot landing page problems
• Tracking anonymous users to identify content problems
• Segmenting specific users for retargeting
• Ensuring clean data
You can create a great landing page, but the important question is how visitors to your site will use it. Let’s look at understanding user behavior on your landing page.
Visitors to your landing page will likely do one of the following:
• Scan the page, see it’s not for them, and leave within several seconds. If a visitor exits after viewing just one page, their visit will be counted as part of the bounce rate (the number of visitors leaving the site after one page per the total number of visitors) on Google Analytics.
• Spend up to 30 seconds on the page and read the first paragraph or two, view any graphics, decide the page is not relevant, and leave.
• Engage with the page, scroll down, and look at various parts, but ultimately decide not to fill out the form.
• Start completing the form, have trouble with it or doubt why they should give their information, and abandon it.
• Actually complete the form and download the content.
If you are not receiving the number of content downloads expected, you need to troubleshoot your landing page. Why don’t users fill out the form and download the content? If the page is truly not relevant to the user, that is fine. But if users aren’t filling out the form, find out why. Is the form too daunting? Too many fields? Is it easy to spot amid all the other images? Does the text make it clear how your company will use their contact data? Is the value proposition clear?
A/B testing
One way to answer these questions empirically is through an A/B version test. Construct a few versions (no more than three or four to keep it manageable) on your automated marketing tool. Change only one variable. Choose it carefully. A good candidate for a test variable would be changing the placement of the form on the landing page. Another would be to test different text describing the content.
If you are sending users to the landing page from an email, make sure that the email list is randomly split into separate groups that will be led to different forms. This is crucial, because a nonrandom sample can create biased results that reflect the kind of user rather than the creative elements you are testing.
Other variables to test include:
• Headline: optimal length and content
• Offer: incentives, different kinds of content
• Images: graphics of contents, varying colors
• Form length: Is usage higher with fewer fields?
Diagnosing form problems
If users are starting to complete the form and abandoning it, this is a clue that something is wrong with the form process. As I said above, you may have too many required fields, and the user experience may be too daunting. Investigate the error messages as well. When the user puts the wrong information in a field, what happens? Does the user have to figure out the problem or is the wrong field highlighted? Does the user have to reenter the information?
If users are abandoning the form, something is wrong with the form process.
These sound like small technical issues, but they can impede form downloads. Remember, only the most motivated users will persist through a difficult form. For most people, even small hindrances can cause them to become frustrated and abandon a form. I recommend that you display error messages right after the field is filled out incorrectly rather than after the user submits the form. Use language like “Please use a state abbreviation like VA.” Using drop-down menus to permit only correct inputs will minimize user errors.
Garnering information about anonymous visitors
Most of this discussion has focused on maximizing downloads of content where users give their information. But the reality is that many users will not enter their information and will remain anonymous. Contrary to popular belief, you can still garner important information about anonymous users.
Why is tracking anonymous users important? Because these are the bulk of users. Understanding their behavior to the greatest extent possible will help inform marketing sales decisions. While you don’t know the individuals, you can still garner some information about users’ companies, at least some of the time.
Even information from anonymous visitors can help you determine if you have the right content.
One of the key pieces of information is the IP (Internet Protocol) address. In some cases, IPs can identify the company, particularly in larger organizations that have their own servers. Another key field is hostname/referrer. This tells you the visitor’s origin, including its IP address and the name under which it is registered.
From this information, you can determine which industries and company sizes are visiting your site. It can help you determine whether you have the right content for your site. If your sales team is doing business with a select group of companies, you can see if any are visiting your site and, if so, with what frequency.
On Google Analytics, you can analyze engagement metrics such as bounce rate, page views, and time on site. These are indications of whether your site is engaging or not. You may also notice that users are spending far more time on certain pages. This may mean there is something to be fixed on the low-engagement pages. This could be a creative, content, or navigation issue. Try changing up one of these elements and see whether there is a corresponding increase in engagement rates.
Still another clue about anonymous users lies in the keywords. By viewing their search term keywords, you can determine which keywords are effective in driving users to your site. You can also analyze keywords used relative to user intent. Is the keyword search generally showing that the user is simply trying to gain information? Or does the search use your company’s product names showing a more dedicated interest?
There are also solutions to help analyze anonymous user data. One is Demandbase, which employs account-based marketing. Its purpose is to identify, target, and retain valuable accounts. According to Demand Metric, “71% of B2B organizations are interested in adopting Account-Based Marketing.”1 Essentially, you can see if your target companies are visiting your site. The data is at the company level, not the individual level.
Targeting specific prospects
To determine the target list of prospects or clients, pick the companies with highest value to your business. You may want to confer with your sales leadership on identifying high-value targets. Another tactic is to identify what type of business is a priority for your company, perhaps by industry or by size of firm. Still another way is to use predictive analytics to identify the characteristics of valued clients or prospects and derive your list of companies that best fit this model.
You can use Demandbase for both net new prospects and existing clients. You can execute an advertising campaign that personalizes ads by industry types, for example, or tailor the content on a landing page to speak to a certain category of prospects. You can also retarget users who visited your landing page but did not download—you can serve up messaging to them on other sites to encourage them to return to your page and to download content this time. It’s like a second chance for the user to receive your content.
You can also target customers for a nurture campaign whereby you can retarget them on sites and offer them specialized content, ungated content, or invitations to industry events. It’s a way of differentiating the experience for your most sought-after clients.
Clean data
Finally, a word about the importance of clean data. A big issue in collecting prospect information is dirty data. If you don’t maintain a clean data set, your CRM will have duplicated and clunky information, which will; impede robust analysis and use by salespeople.
Here are some tips to combat common problems in your data:
• Automatically check for duplicates. This is a very common problem, because most companies do not check their CRM data for duplicated names. The problem often arises when a user fills out different forms and varies his or her entry slightly, so the system thinks one user is two different people. See if your marketing automation tool can use a unique identifier to eliminate duplication.
• Ensure the validity of email addresses. This can be accomplished through setting up the form system to ping the email domain in real time. This averts invalid email addresses. As I said earlier, you may want to insist on work email addresses only. This is preferable because it shows higher engagement from the user than inputting a free email domain such as AOL, Gmail, or Yahoo. You should label the field as “work email address.” That said, be aware of company spam filters that may block your future emails, even if that is not the user’s intent.
• Standardize responses through drop-down menus. If users are allowed to enter whatever response they choose, one user may enter “VP” while another may enter “Vice President,” thus making future analyses more cumbersome.
• Stop bots. Bots are automated responders that will create lots of noise in your data. To avoid bots, find a spam trap. The other option is the less user-friendly Captchas, which ask users to duplicate numbers presented on the screen to prove they are human.