We have seen tonnes of KPI’s filling up our dashboards over the years. From conversion rates to bounce rates and from time on site to depth of visit. Now, I know that these are not the best examples of complex KPI’s but they (should) form the foundation of every company’s decision making as far as online activities are concerned.
Over the last several months, while working on some extensive Voice of the Customer projects, I started to think about how to integrate data from the two sources. I wanted to gain more insights from qualitative responses (attitude) from our visitors and correlate this with quantitative data (behavior). In other words:
Were people really doing what they say they wanted to do?
or even better:
Could the people do what they say they wanted to do?
For this reason I have come up with ‘True Intent Conversion Rate’. Intention should allow us to better segment and serve our visitors. Like the Beastie Boys sing in their song That’s It That’s All “Intention leads to action”.
Back to the business at hand… soon after starting work at my new employer, I was able to push through a minor Voice of the Customer project that would involve the implementation of iPerceptions‘ 4Q survey tool.
My experience as an in-house analyst has been to start out small and cheap when it comes down to tools. Spending loads of money on research tools is an easy thing to do, but it won’t guarantee you anything. By limiting your initial investment you, your colleagues, your company can slowly get to grips with the new insights. Hence our choice for 4Q. Free, good basic analysis suite, easy to implement and unlimited number of responses.
After implementing 4Q I quickly realized that these basic insights were quite powerful. Not only were visitors relatively happy to provide us feedback, but they were quite often able to pinpoint the real issues bothering them during their visit. Now, although these insights are great, my attention shifted to one of 4Q’s default questions in the survey ‘What was the purpose of your visit?’.
Although this is one of the standard questions in the survey, the answers can be edited in the back-end. Before implementing the code via our CMS onto the pages we wanted to invite users on, I edited the questions to match the context of our travel website. The 5 main questions being:
These 5 questions reflect the 5 basic stages of our visitors that we have identified within our travel, packaged holiday, vertical:
** can be pre- or post purchase or even visitors wanting to continue via our offline channel
Below an example of responses to this question (only showing 2). ‘I want to book a holiday’ is shown here as ‘Ik wil een vakantie boeken’, the Dutch equivalent.
I think that I can say with quite some certainty that there are many companies out there who thrive on Conversion Rate. It is most likely the most often used metric/KPI among chiefs and managers. Any analyst out there will hopefully agree with me that the old school conversion rate metric is extremely prone to irregularities such as (poor quality) traffic spikes, website errors and the list goes on. It will basically tell you how well you converted the visitors to your website, including the people who:
With True Intent Conversion Rate, this does not change much, but it will help me filter out traffic who’s purpose it is not to book/purchase a holiday. It will allow me to segment my visitors based on purpose of visit, giving me a clearer image of how well or poor we are converting the visitors who want to make a purchase.
For the mean time, since 4Q’s Google Analytics integration has not been completely deployed, the calculating of True Intent Conversion Rate is pretty much a manual chore.
iPerceptions’ Product Community Support Coordinator Fiona Peter has confirmed that with the 4Q Suite you will be able to fully integrate 4Q with Google Analytics and create segments based on responses given to the Purpose of Visit question. The metrics (excluding date ranges) involved are:
The calculation would then be:
O / (V x I) = True Intent Conversion Rate
4Q, as Google Analytics, allows you to download the data in .csv format making it easy to build a sheet in Excel that can do the calculations for you.
Although 4Q measures Task Completion through one of the four questions and is in actual fact not an exit survey tool, I have found this data to be unreliable since the question is often asked and responded to at the beginning of a visit and does not reflect the answer of a visitor who has actually ‘completed’ their visit or task.
What can we do with this metric? How can we grab the attention of the chiefs, the managers, the budgetholders? Like many wise guru’s have said many times in the past… monetize the metric. Add a monetary value to the metric and quickly show the potential income gain that can be made.
Imagine a website, maybe your website, where you get close to 200.000 visits a month. With a tool like 4Q, you have found out that 5% of those visitors (sampled of course) came with the intention to make a purchase, that’s 10.000 visitors. With 1000 orders, you regular conversion rate would be 0,5%.
Based on True Intent Conversion Rate however, you quickly realize that you only converted 10% of the visitors who came to your website holding their wallets or credit cards. With an average order value of $100 and 9000 visitors who did not convert, even though they had the intention, this means you potentially missed out on (9000 x $100) $900.000 in revenue.
Hey Mr. Budgetholder, can you hear me now?
Like web analytics, calculating True Intent Conversion Rate is not and never will be 100% accurate. The insights gained from this exercise should help you and the website stakeholders become aware of how well your website is performing for those who have the intention to purchase goods from you during that specific visit. It should give you enough ammo to further investigate what it is that is keeping visitors from purchasing. Maybe it is a usability issue, or a web performance issue, either way it should be treated as a start off point for further research.
A visitor’s attitude and behavior are difficult to match at an individual level, let alone at a group level. Sample based results should also be handled delicately for that matter. Qualitative research tools are great but will never tell the full story. Maybe the visitor was just about to enter his/hers credit card details when: the phone rang a distracted from the purchase, dinner was ready and being served etc.
The metric helped me leverage the importance of the many aspects of performance of our website. By calculating how much money we were potentially losing, give and take a few percentages on either end of the said calculation, more people inside the organization became aware of the leakages.