Why online marketing and web analytics often become stuck
Many companies fail to exploit the full potential of web analytics and online marketing. They usually have a moderately adapted standard web analytics script on the pages that produces moderately useful data in standard reports – and sometimes with even quite a lot of reporting. But rarely is the reporting data used to properly support the decision-making process.
Organisational integration of digital analysis
The main reason is the insufficient strategic and organisational incorporation of digital analysis within the company (to articles: “Successful e-business thanks to web analytics”). To exaggerate a little, the company management has not fully realised that one should invest in the processes, personnel and technical development of digital analysis – tools such as Google Analytics, Webtrends and Adobe SiteCatalyst do not merely count visitors, but can also earn or save the company money.
Figures that count are often obtained only with advanced tracking
However, in order to accurately collect the data that enables revenue-increasing decisions to be made, advanced reliable tracking is required. Not visits or page views, but Swiss francs, registrations, form error tracking, A/B testing, customer segmenting, etc – and this is only the beginning. Without advanced tracking, no figures that count. Without figures that count, less acceptance at management level of a strategic role for digital analysis. Without a strategic role, no investment in processes, people and technology.
Why progress is so slow
So why is advanced tracking such a slow process for many companies? The reasons are usually:
- An overworked IT department
- Inflexible release cycles (usually every two to three months, sometimes only twice a year)
- Data quality problems (why does tool A say 2,000 when tool B says 2,500?)
Data quality problems engender mistrust of web analysis tools. Normally, one doesn’t invest in technology that one doesn’t trust. Here, I would like to use a few examples to illustrate why IT availability and the dependency on release cycles are particularly critical for online marketing and web analytics.
“God created the internet so we could fail faster”
In principle, the advantage of being online as opposed to offline is that you can obtain data with little effort in a very short time. You can learn from your mistakes faster – or as analytics guru Avinash Kaushik puts it, “I believe God created the internet so we could fail faster“. The only problem is that ‘faster’ usually still means ‘when the developers find time to resolve the error, and then only with the next release’. In most companies, the next release (website update) appears about once a month to twice a year.
Example of online marketing: gift certificate campaign
First, an example from the world of online marketing. You are running a campaign. You are using a limited gift certificate campaign to attract new users; i.e. to generate new customer accounts. The campaign is not going very well so far and your target is at risk. Luckily, some of the budget is still left. So you decide to try out your banner on a new ad network.
However, even if the developer is available immediately, they cannot add anything without you having to wait another month – because, unfortunately, the last release was just the day before. So you complete 95% of the work in two days and then have to wait 30 days for the remaining 5%. You will not be able to see how your new marketing partner is performing until you have potentially paid too much money or the campaign is already finished.
This is strange in a world in which marketing agility is increasingly important.
Examples from other fields: form tracking, A/B testing, etc.
A couple more typical examples: you want to know how well the new navigation structure is working. Or you are interested in how many users from which campaign have sent a booking form. You want to use the new Google Analytics ‘demographic data’ feature. You want to know what percentage of your visitors is logging on. How often users switch from the mobile to the desktop view, and vice versa. Or you want to try out two variants of your landing page. For all this, you usually require a new tracking code on your website; for example, Google Analytics or Webtrends code. And to implement this, you require, of course, the developer. And a release. Too bad if your boss expects reliable data on the use of the navigation at the start of next week.
It will be even more expensive and time-consuming if you make an error in the specification, or the developer overlooks something in the first installation attempt. Then you go through loop after loop with the developer on the test system. With every loop, the costs increase. And if you discover a non system-critical error only after the release, you will have to wait until the next release for this to be removed.
The crucial question: “Do I need a release for this?”
It’s no wonder that for the next tracking request, you think long and hard about whether you really want to invest the budget and time in it. “Do I need a release for this?” has therefore become the crucial question for more than a few marketing people.
For this reason, it often remains an expensive dream even for ambitious web analysts – advanced tracking that provides them with the figures that count.
Tag management systems can help. They can be used to significantly reduce the implementation time for new tags; i.e. tracking codes.
What is a tag management system?
Tag management systems (TMS) have been around for a number of years. However, the term is still unfamiliar to many online marketing people in this country. Basically, a TMS is a type of content management system for web analytics and other tracking codes (also known as tags).
One universal tag instead of many individual tags
The main advantage: independence from IT
From then on, you yourself define in the tag management system – in a browser-based user interface – which tags (tracking codes) are to be executed under which conditions (‘rules’ or ‘filters’). This universal tag now loads all the tracking codes to the page that IT would otherwise have had to install – for example, Webtrends, AdWords, Affiliate, A/B testing, social sharing, content recommendation and Netmetrix tags. In 90% of cases, the installation of new tags is performed without a release.
Figure: most frequent application cases for a tag management system
In the next article in this series, you will find out how to use tag management systems to reduce the implementation time for new tags; i.e. tracking codes.