Web analytics is a great solution for website authors and web administrators to know how well their site is selling and how to optimize it for maximum outreach. However, measuring raw data using web analytics solutions is not free from error and issues. Think about it, when was the last time you were 100% satisfied with the results you got?
There is always concern about the accuracy of data, reliability of the data collected and the everlasting first party/third party cookie argument.
Lets discuss some of the most popular web analytics issues and how to tackle each one of them.
1. Data Inaccuracy- Web analytics data is never hundred percent accurate. The data that streams through the dashboard of the analytics software is not free from error.
There can be several reason for data inaccuracy –
a. Installation– Error prone web analytics solution installation can be the number one cause of data inaccuracy. Each web analytics package comes with different installation mechanism. Some are tag based and others are based on java script and require different set of coding expertise.
Here are few suggestions to make sure your web analytics installation is free from installation flaws-
(i) Use free tools like wasp.com to debug your installation.
(ii) Install free solutions like Google Analytics on top of your enterprise installation. Google Analytics can be installed within minutes. Plus, two sources of data will help you compare the metrics and kpis more effectively.
(iii) Hire expert analytics programmers in your marketing department. [Hint: Web analytics should stay within the marketing department and not with your technical team]
b. Content issues – Depending on the programming language you use for your web pages, analytics installation should be installed accordingly. Your installation should also adapt to your cms system.
c. Cookie issues – Study conducted by Comscore showed that 30% of the visitors delete their first and third party cookies. This means that you can lose campaign attribution data and upto 30% of your paid efforts could result in direct traffic (clean entry traffic) or unaccountable traffic.
To overcome this I suggest using the keyword brand identification technique. Here is how it works –
(i) Identify percentage of brand vs non brand keywords (paid and non paid) using advanced segments.
(ii) Deduct your branded keyword % from your % direct traffic
True direct traffic = % Brand keywords traffic – % direct traffic reported by your web analytics system
(iii) The number that you get will be your actual direct traffic originated from sources other than your paid or natural search. This is your actual branding effort and could be linked to referring sites, press release or other non paid sources.
2. Data Integration- There has been a big debate on the importance and reliability of data integration services. Enterprise analytics providers are a big player in promoting the heck out of data integration solutions as an upsell. Several new companies have their entire business model built on data integration.
Agreed, data integration is important and connecting the dots can provide you a complete picture from visits to the final sale. However, you have to be very careful in choosing the data integration service provider.
If you have resource then I would suggest you should integrate the web data to crm and other solutions in house using your datawarehouse team.
If you are resource constraints then ask your self these questions before paying money the data integration service provider –
a. Why should I integrate my web data?
Using simple coding hacks you could bring in your crm and other data sources into your web analytics reporting system.
b. Who are the stake holders involved?
Dealing with different departments who manage the software packages in your company can be a nightmare.
c. How long will it take?
This one is my favorite because all the promises about quick results are pure sales pitch. Make sure you put this on paper before your sign the contract.
d. Do I have to make changes to our software interface?
1 out of 5 times your integration will get delayed because the integration process is stuck due to variable mismatch between your web and other solutions.
Finally, I would highly recommend you read “Web Analytics Integration: Holy Grail or White Whale?” at webanalyticsdemystified.com.
3. Privacy- Almost all the analytics solutions in market including both free and paid capture personal site visitor data including their IP address, geo location, computer and internet configuration and time & frequency of visit. While most of the data is safe, the IP address tracking proves to be a sensitive issue especially when identity theft and spamming are concerned.
Consumers are getting frustrated because they don’t want their web actions to be tracked by companies. Companies on the other hand are paying big bucks to know everything they could about their prospects so they can sell more.
Hope you enjoyed this post. Please share your thoughts, insights and comments because it really helps me write more effectively. .
Here is a list of other helpful blog posts –
Bounce rate…Does it really matters?
Five Rocking Google Analytics Tips, Tricks & Hacks