As part of The Drum’s special report on analytics, we spoke to a number of agencies operating in the space to gain an insight into the key issues surrounding this heady topic.
In a series of features, we’ll be looking at the industry’s responses to the questions we posed, to determine the challenges and trends facing analytics in 2012.
What is the main challenge in using web analytics?
Ron Person, director of analytics, Sitecore
One of the main challenges in web analytics is measuring cross-channel effectiveness and Marketing Return on Investment. With more than 12 online marketing channels and more being added, it takes at least 47 traditional web analytic metrics to accurately measure performance across these different channels. With that many metrics it's impossible to build a mental model of how to drive performance.
Avinash Kaushik, who wrote the bible of traditional web analytics, “Web Analytics 2.0”, identified eight key metrics for web sites. Five of these eight metrics are used to infer the eighth metric, Visitor Engagement. Avinash even proclaims that whoever finds an accurate way to measure Visitor Engagement has found the Holy Grail of web analytics. Using traditional web analytics, an experienced analyst has to infer the level of visitor engagement from metrics like Time on Page. Not only are there analytical flaws in this method, but it requires an experienced analyst to guess the level of visitor engagement.
The method we advocate is to measure visitor engagement directly by measuring actions the visitor takes. And the results have proven themselves statistically. There are specific actions in every human relationship, in every engagement between people. The process of developing a relationship where there is engagement goes through the steps of two-way communication, trust building and finally commitment. By measuring these actions we get one metric, Engagement Value, which works across for all marketing channels and marketing efforts.
An Engagement Value can be put on every transaction point between an online visitor and a company, whether that point is registering for a newsletter, participating in a forum or responding to an embedded link in a Twitter message. Once you have this single metric you can measure the effectiveness and impact across different channels, different messages and different marketing assets.
Guido Fambach, VP professional services, comScore
With the rising volumes of data available to a variety of people within organisations, one of the main challenges is to translate the data into actionable reports and insights. Another important factor for web analytics is to make sure that the right people across the organisation have access to the insights and act based on this newly gained knowledge.
Conrad Bennett, VP technical services, EMEA, Webtrends
One of the biggest challenges in using web analytics is not the analytic tools, but the lack of expertise within businesses to leverage the data. No matter how much money and resource you spend buying and implementing an analytics system, it’s only as good as the people using it.
Businesses get too easily wrapped up in their internal histories figuring out what worked or what didn’t long after the fact. However, nothing in digital stays static for very long because we deal with people, be they buyers, customers, readers etc. We don’t analyse websites, we analyse people’s behaviour with our content. Analysts focus on the future by helping businesses fail faster allowing them to understand and react to what people want now. Get that right and you will improve the bottom line.
The main challenge is ensuring that your site analytics package is set up correctly Regardless of the package you use it is generally best practice to get a recognised expert to perform this for you. Although many companies would prefer to avoid costs associated with this, they are simply storing up problems for themselves later down the optimisation pathway. If you’re not tracking the key metrics for your business, site analytics – even the enterprise-grade packages with enterprise grade prices, struggle to deliver basic reporting, let alone providing the “analytics engine” essential for conversion optimisation.
Carl Fernandes, head of analytics and conversion optimisation, iProspect
Data consistency is the biggest challenge we have faced working with web analytics across different types of clients.
We frequently work with clients who have not setup their web analytics to track data accurately. A common example is where e-commerce websites use subdomains (e.g. http://cart.example.com) or third party shopping carts to process transactions. Web analytics solutions generally do not cater for cross-domain tracking which means that sales referral information can be lost unless an advanced implementation is in place.
Another challenge is that web analytics tools by default do not differentiate between Paid Search and Natural Search traffic. This can be a big issue for marketing teams when trying to assess the value of each channel and understand how to set budgets accordingly.
Recent changes to Google’s privacy policies have resulted in search query data for natural search no longer being exposed for visitors who are logged into Google services. This has impacted all web analytics tools, including Google Analytics.
Digitally savvy industries will be more affected by this change than others but it is reasonable to assume that over time, as Google builds its user base across Google+ and gmail this will have a huge impact on all websites, regardless of vertical. It’s also believed that Firefox is considering using Google’s HTTPS search as the default option within the browser.
Seth Richardson, CEO, DC Storm
The main challenge for web analytics is reporting across the multiple channels that influence the users’ path to purchase, or conversion. An example of this is web research leading to an in-store purchase. Currently data is often held in several disparate systems, meaning that linking and consolidating this data to produce a complete reporting framework poses a significant challenge to analysts. Once achieved, this can change their “view of the world” with regard to the relative performance of channels.
Another major challenge with analytics is knowing the questions that you want answered and maintaining the discipline to use the tool to answer those specific questions. All too often users log in, start looking, find lots of interesting information but fail to find anything that drives any action.
Subscribers can download their copy of The Drum’s analytics supplement here.
Jigsaw image via Shutterstock