In this article, we will dig into some Key Performance Indicators (KPIs) in depth, focusing on the metrics that my colleague, Julie Mason of Kellysearch.com, wrote about in her recent article ‘A Practical Guide to Key Performance Indicators’ :
- Visitors per conversion
- Average page views per visit
- Cost per lead
- Percentage of new visitors
These are all great starting points for website analysis, but sometimes they raise more questions:
Visitors per Conversion
All websites should have a goal, even if it’s just a PDF download, and so tracking your conversion rate in this way is an excellent measure of your site’s success. But what if you know your conversion rate is low, but you can’t identify the reason? The answer lies in the customer’s journey through the conversion process and by configuring your analytics tools’ funnel analysis you can quickly identify problems:
The checkout process on an ecommerce website is probably the most common example of a customer journey, typically catalogue -> product -> shopping cart -> shipping -> billing. In your own conversion process you may see 1,000 people visiting the catalogue page, but only one coming out the other end having made a purchase. This is probably because of a design or usability flaw that is creating a bottleneck at one of these steps. Funnel analysis can identify where that problem is by showing where users enter and leave the process at each step.
Funnels can be tricky to configure and analyze, but they are one of the most effective analytics tools for determining user interaction and generating genuinely actionable findings.
Average Page Views per Visit & ‘Stickiness’
These KPIs are typically for websites with content & advertising-based revenue models. The key here is to measure how well your visitors are engaging with your website. As Julie points out, you may not always want these numbers to be high – sometimes you want a user to get in and out of your site as quickly as possible, for example, when assessing the quality of an internal search function. One of our key metrics at Kellysearch.co.uk is how quickly users move on from our search results pages, i.e. how ‘unsticky’ these pages are. This is because we want our users to find what they want as quickly as possible, a common goal among search-centric websites. So we measure average page views per visit just for search results pages and we want to see low numbers.
There are also some new technologies out there that can disrupt these particular metrics. For example, Ajax is an increasingly popular scripting language that enables websites to, among other things, load new content on a page without refreshing that page. One page view of an Ajax page can actually mean the user has viewed multiple pages. Consider this when looking at your analytics data and you can use alternative measures of ‘stickiness’ such as time spent on the page.
Cost per Lead
Establishing the Return on Investment (ROI) of your website can be critical to establishing its success and if you are going to measure ROI then you need to understand what your traffic is costing you, i.e. what is the cost of your online marketing. Marketing is traditionally a difficult medium in which to measure ROI, but with the right analytics this can be much easier. For example, Google Analytics has the facility to import AdWords cost data into your analytics reports automatically. Add a monetary value to your conversion goals and you can have an automated calculation of ROI and Return on Advertising Spending (ROAS) for each visit.
Google Analytics’ integration with AdWords is seamless and ideal for the analytics novice, but most other analytics solutions can also import cost data for PPC campaigns in some way and therefore provide real world numbers that can justify a website’s existence.
Percentage of New Visitors
This is another great metric for measuring the success of your marketing, but there are two key points to remember in order to get the most from this KPI. First, make sure you are using your analytics solution’s tagging system to measure where these visitors are coming from and assess which elements of your marketing campaigns are successful.
The solution is to use first party cookies wherever possible and, as with all analytics data, study the trends rather than the absolute numbers.
With the multitude of possible metrics that analytics systems make available, it can be difficult to select the right ones for you and even more difficult to interpret the results. However, once the basics are in place you can start digging deeper – and the results can profoundly change the way your business works online. One final tip: If you’re starting out in analytics you could do a lot worse than take the advice of analytics guru Avinash Kaushik, spend 10% of your budget on the software, and 90% on your analyst.
Jim Newsome is Kellysearch.com’s Search Marketing Manager. Jim has more than 11 years experience in the business of building, optimizing and analyzing websites. He has worked for small agencies as well as large multinational companies and his particular specialties include Search Engine Marketing and Web Analytics.