A Web Analytics Intervention, Part 1

• Tweet
• Buffer

I received the following letter earlier this month. The names and company info have been changed to protect the innocent.

Shane,

It was great to meet you last week. I'll cut to the chase: we need an intervention. I joined Company X 16 months ago. I had hoped to make an immediate impact by leveraging Web analytics and site optimization, skills I honed at a previous company and something I'm particularly passionate about.

Unfortunately, I haven't been able to create the same culture of analysis in this larger environment. Either I'm completely naÏve, politically ineffective, or just haven't found the right way to gain support and resources for marketing and site optimization. Hopefully, it's the last one.

My department is filled with brilliant marketers with the right intentions, but every day I witness things that make me cringe:


• Misuse of data, usually after project completion
  
  
• Inconsistent KPIs [define]
  
  
• Endless rounds of debate over visual design and copy with every project
  
  
• Ego-driven decision-making: the most senior person always decides
  
  
• Haphazard project prioritization
  
  
• Incomplete Web analytics implementation
  
  
• Data overload
  
  
• Lack of A/B testing to verify concepts
  
  
Here's the good news. There have been a lot of recent changes at the company. When you line them all up, I think it's the perfect storm to gain the support I've been missing.

To start with, we just hired a VP of marketing with a background in measured marketing and a creative director who is eager to break new ground. We also keep hearing from the executive staff, asking us to increase our focus on ROI [define] and make decisions based on that data.

I can tell there's a bigger appetite for performance dashboards, too. (Most of our business is transacted online.) We recently made a significant investment in Analytics Software Vendor Y, but the implementation is incomplete for marketing purposes. We're eager to tailor our online experience based on segmentation and behavior. When you add in a curiosity about multivariate testing and a new CMS [define] due to launch this quarter...well, you get the idea.

I know I'm not alone. I work with smart people who realize we should be making data-driven decisions. But [I've been] waving the flag and evangelizing actionable analytics for some time now; the application is falling short. I'm looking for a different angle to get the organizational support to bridge this gap. I see the potential and the impact all these things could make.

Shane, do you have any tips for me? How would you go about tackling the situation?

Much appreciated.
(name withheld)

In Part 2, I'll respond to these issues.