Conversion Rate Optimization

5 Tricks for Getting Deeper Visitor Insight with Your Current CRO Tools

Conversion Rate Optimization is all about understanding your target audience and what makes them tick. CRO tools are essential to gaining that insight, but you don’t necessarily have to invest in a new tool every time you want to dig a little deeper. In fact, often all you have to do is take advantage of the more advanced features of your current tools.

Here are 5 tricks for how to get deeper visitor insight without investing in new tools.

1. Find the conversion and drop-out rates for every field in your funnel

Digital marketers should be aware of drop-out rates at every step of their funnel, and will have often worked to ensure that the conversion rates of each are as high as possible. For many sites, however, this misses a lot of very useful data.

Every time a user fills in a form, each individual question or field forces them to make a decision. Will they provide their name to get your product or service? How about their email? Phone number? If they decide to leave, what field caused the drop out? It is extremely valuable to know exactly which field is causing users to drop out of the funnel – not just what page.

In this example from an e-commerce client, we tracked the completion rates of each field. If the user left a form field after entering text, we tracked this as a successful completion:


[sociallocker]We can easily see that users were concerned about providing their telephone number, which is typical for many sites. We also tracked when users were shown a form error. Similar results were obtained, where errors from not completing the billing telephone number were the highest:


This showed that users were very concerned about providing their phone number, but it was only needed by the retailer in case of delivery issues. By adding a message telling customers this, we significantly reduced dropouts.

The Good Way

Several CRO tools offer funnel analysis. Clicktale offers form field tracking as part of their tracking solution, Formisimo focuses specifically on form tracking, and many analytics and insights packages include such options. Although these programs are specifically designed to track this data, it can be difficult to manage the additional onsite tags, multiple analytics dashboards, and added costs.

The Better Way

There is a simple, free, solution, which uses Google Analytics events to track this data: The first event tracks field completion and provides:

Category: Page where form was submitted

Action: field_completed

Label: Name of field

The second tracks any errors on the page and provides:

Category: Page where form was submitted

Action: field_completed

Label: Name of field The code to fire these events is simple and can be implemented site-wide.

Sample code is given below:

$(document).ready(function() {

var currPage = window.location.pathname;

$(‘:input’).blur(function () {

if($(this).val().length > 0){

_gaq.push([“_trackEvent”, “Form: ” + currPage, “field_completed”, $(this).attr(‘name’)]);



var errs = $(‘label[style=”color: red”]‘).contents();

for(var i=0; i<errs.size(); i++) {

_gaq.push([“_trackEvent”,“Form: ” + currPage, “form_error”, $(errs[i]).text()]);




The code will get the path name of the page (for the event category). Then, when the user leaves a field (it loses focus), it checks if the field contains a value, and will then fire an event with the name of the field.

This must be updated for your specific site, particularly the error tracking (highlighted yellow above). In this case, we use JQuery to get every label that has been styled in red, as on this particular site, errors are shown this way. You should be able to use similar code on your site.

Using this technique should generate large amounts of very valuable data, and allow a much greater insight into the usage of forms on your site.

2. Call tracking on a budget

An important metric for many sites is the volume of phone inquiries received. A call is often more valuable than a web query since call center staff are trained to convert users. This important metric is rarely tracked in tests, but it can be vital to find the best version. We used the following technique on a client’s site and found that a test with an insignificant web-conversions result had generated +150% increase in calls, a huge benefit to the business.

The Good Way

Tools such as ResponseTap, Infinity Tracking allow advanced call tracking on your site. They automatically show each user a unique phone number onsite, allowing them to be tracked across web and phone. This gives really detailed data, but can be very expensive for medium to large call volumes. They also require additional technical integration.

The Better Way

A practical alternative is to use several different numbers (all directed to the same call center) to track your split tests. For this technique, you just need to be able to track call volumes on each number. Then, assign a different phone number to each variation in the test.


With a few lines of JavaScript, you can automatically replace your normal phone number every time it appears with the number for that variation:


function walk(node) {

var child;

switch (node.nodeType) {

case 1: // Element

for (child = node.firstChild;


child = child.nextSibling) {




case 3: // Text node

node.nodeValue = node.nodeValue.replace(

/01632 960 000,“01632 960 001″





In the example above:

01632 960 000 is the Regular Expression which gives the number to replace on the page

01632 960 001 is the new number that you want to display on this version

This code changes the number everywhere on your site, so you can monitor call volumes on each number, and ideally the conversion rate of each.

This will give you top-level insights into the calls each version is generating. This gives you far more detailed insights with almost no additional work.

3. Track offline conversions through your split testing tool

For many businesses, tracking inquiries through A/B tests is only a very basic metric. In reality, customer journeys continue offline, whether it is through a call center or a visit to a physical location. Perhaps the customer has a long buying cycle which isn’t completed from an initial web inquiry. Monitoring the outcomes of these interactions are the true indicators of profitability, but this is very challenging when the split testing tool only tracks web conversions.

The Good Way

Often the only way to calculate the true conversion rate is to use the overall ‘offline conversion rate’ for the business. Well known metrics such as the percentage of inquiries that a call center will convert, or the number of appointments that result in sales are usually simply applied to the metric being monitored by the tool.

If version B generates 20% more inquiries than version A, most businesses assume that this translates through to 20% more sales.

This is often a flawed assumption as different versions of a page can generate more (or less) qualified traffic who have a different probability of converting offline. This can be hidden by the split test results. That means that your business could be making incorrect decisions.

The Better Way

Many split testing tools including Optimizely offer the ability to track offline conversions within the tool. This means that any action people take after a conversion on your website can be tracked within your split test results. This could be someone phoning your call center (and converting) or maybe even going into a store or making an appointment at a clinic.

Optimizely gives the details for this service here. In order to track these, however, you need to include the variation (and ideally user session) data in the customer record in your backend systems. One way to do this is to pass these details as hidden fields in a form, and store them as data in your database.

Then, when an offline conversion event happens, your system can request this URL to record the conversion. If this automation is difficult to implement, a call center agent can simply click the link, which would record the conversion.

For many businesses, however, this major IT integration is very expensive. There are still ways to provide this data, however. If there is a free text field on your form, for example a comments field (or if you can add one and hide it), you can use simple JavaScript to append the URL to any details the user enters.

That means that when your call center agent sees the web form submission, they will also see the link (and can click it). The user sees nothing different and no major IT integration is necessary.

Tracking your conversions offline can generate very valuable data for your businesses and should be considered as part of any conversion optimisation project to ensure the full impact of testing is known.

4. Get Qualitative results from your split testing

Qualaroo and other qualitative feedback tools have become vital parts of a typical CRO toolkit. Being able to get instant feedback from your users, often in their own words is extremely valuable, but once you start running A/B tests on your site it can be difficult to interpret the data. Some users will mention things which have changed, or may refer to new features you are testing.

The Good Way

For most free text responses, it’s pretty obvious which version the user is looking at. As long as someone has the time to process them all, you can typically identify which responses relate to your split tests. When it comes to more detailed insights, most marketers will ignore this data completely during a split test – using qualitative results for insights only. This risks missing some key understandings about your tests.

The Better Way

Luckily there is a useful feature in Qualaroo (and similar platforms), which allows you to combine the two techniques. Qualaroo offers a simple JavaScript call which allows you to pass in a name/value pair (other tools will have similar features):


You can also take this further by including other metrics in your feedback. One client uses the Net Promoter Score as an important assessment of the performance of the website. This standard metric asks users “On a scale of 0 to 10 how likely would you be to recommend us to a friend?”


The percentage of detractors (scoring 0-6) are taken from the percentage of promoters (scoring 9-10) to give a Net Promoter Score (NPS). This ranges from -100 to +100, with higher scores being better.

By asking this question through Qualaroo, and combining this with the technique given above, we were able to find that a split test that gave equal conversion rates for the two versions actually caused an 89% increase in the Net Promoter Score – a massive win for the company and one that wouldn’t have been discovered without this technique.

Whenever you run split tests, pass the data into your other tools, so that you can analyse the split of results.

5. Discover the real “fold” on your site

It is generally agreed that important information should be “above the fold” and, wherever possible, the information and actions which the user needs at any point, should be visible on screen.

In the past, the fold was generally predictable. There were only a few browser sizes and it could be taken to be one of just a few positions. Today there are so many browsers and devices that the best judgement is to work out what percentage of users can see an element above the fold.

The Good Way

Tools such as FoldTester allow you to enter a URL and show approximate percentages for a typical breakdown of browsers, but this data is aggregated across many sites, and becomes outdated quickly.


The actual breakdown may be very different for your site if your traffic pattern is different. For example, if you have many mobile and tablet users, the fold will be higher on the page, whereas if you attract many designers and marketers with huge screens, much more will be visible.

The Better Way

Google Analytics can give you live, accurate data for your specific traffic breakdown and this is a much more useful place to look. In the Behavior section you will find the In Page Analytics report. Open this, and chose “Browser Size” on the top right. Then choose “show percentiles”:


As you can see from this example, the site has a lot of mobile users (the sections at the top), but generally most visitors see a large area of the page. You can browse around the pages of your site using this report, and it gives very accurate and useful information on the location of the fold.

If you have a separate mobile site, you will see correct fold data for this (based on the viewport sizes of the users), and for a responsive design, you will see all traffic overlaid on the same page.

By using this report, you can see very accurate fold data for your site and ensure that you’re optimising for your visitor profile rather than someone else’s.


Your CRO tools are very powerful, but it is only by trying out their more advanced features and integrating them together that you can unlock their full potential. That means that you’ll understand your visitors better and ultimately, you can increase your conversion rate and revenue.[/sociallocker]

Anil Valvi
Founder of Dynastic Tech & Digital Marketing Consultant, Product Designer, Technologist & Inventor. Passionate about Online Advertising and Social Media.
You may also like
Validation Boosts Website Lead Production
10 CRO Experts Reveal Their Biggest A/B Testing Mistakes

Leave Your Comment

Your Comment*

Your Name*
Your Webpage

How to Get 25,000 Visitors Per Month?