My Photo

Clicky

  • Clicky Web Analytics

Your email address:


Powered by FeedBlitz

« What is ‘Big Data’ – Shortening the Path to Enlightenment | Main | Meeting at the Adobe Summit »

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d83454a6d169e2017d411b2868970c

Listed below are links to weblogs that reference Why Measurement Doesn't Matter : A Guide for Confused, Annoyed, or Disappointed Decision-Makers:

Comments

Great post Gary, I feel that organizational challenges and best practices within web analytics is a topic not discussed enough.

I recognize your points on centralised/de-centralised settings, but two organizational questions I'm currently spending much thought on are:

1. Web analytics in relation to Business Intelligence (BI). I guess most companies have started with web analytics separate from the BI-team. But when your digital distribution has grown from 10% to 70%, then suddenly web analytics is essential business intelligence and you find yourself wanting to analyze data sets from both sources. So how to organize the two teams and technologies (merge teams or not) to maximise insights and operational efficiency?

2. Centralised web analytics in relation to the more technical work of providing the web analytics platform (i.e. implementation, configuration, data quality, support, etc.). I think of it as 2 different roles, analyst vs system developer. Initially one or two centralised web analysts usually take on both roles. But when more (decentralised) analysts join and the requirements increase on the platform and the original analysts see the developer role grow at the expense of analysis. Adding more resources in the centralised team raise the question, should every team member have both roles or should we split the team into analysts and developers? Should the developers join the DW/BI-team mentioned earlier and let centralised analysts constitute their own team focusing on analysis? What are pros/cons, best practices?

I like how you divide web analytics skills into four different functions, web analytics is truly multi-facetted. I have made a similar categorisation by dividing web analytics into four perspectives, each with it's own challenges and tools: Customer, Product, Marketing and Distribution (i.e. site or app). For example, conversion rate means totally different things if you look att it from the different perspectives. Is it the conversion rate of a specific customer segment, a product or product category, a campaign or a buying journey on your website?

As always on your blog, a post which I truly enjoyed. It coincides with the restructuring of our e-commerce team of 10 people. We used to have a central e-commerce analyst who supplied reports to all business seniors and supported them in the interpretation and subsequent decision making processes. This e-commerce analyst was me.

Last week, I told my boss (the e-commerce director) that I would not want to go on like that. I felt that much of my work has not been used in a really valuable way: Some people seemed to not thoroughly read the stuff I provided them, others obviously felt that rather than supporting them I was interfering with their business. So last week we agreed on moving the controlling of high-level metrics / business targets to the director level (I wonder if he will have the time for that though) and moving the measurement of lower-level metrics to those people who actually make use of the data. That way, I hope that the overall measurement efforts will be more efficient since everybody will (hopefully) measure the most relevant metrics which perfectly help to achieve their personal targets. And if somebody thinks, time is e.g. better invested in the acquisition of new partners than in partner performance measurement - fine. Why tell people what and how much they need to measure in order to be successful? And moreoever, how could a central analyst like me ever give useful recommendations to a SEO expert on which keywords to optimise for? This guy has a range of analytics tools which I don't even understand. So let's rather bring some measurement knowledge to business people than business knowledge in really different disciplines (category management, performance marketing, CRM, SEO, CRO, etc.) to a geeky analyst.

Anyway, to be honest I still have some mixed feelings: Will everybody have (or achieve) the analytical expertise to optimise their own business? Will they be able to take the time it really needs? Will they finally adopt a data-driven mindset? Let's see.

Michael,

Thanks for the great comment. I think you've hit on what really is one of the central challenges in analytics from an organizational standpoint. We've often been in the same position when it comes to annotation and reporting - it's really tough. I don't think there is one right answer and, as I tried to get at, I don't think every Web analytics problem should get the same centralization/decentralization treatment.

Gary

Robert,

Great comment - Thanks!

Regarding your points, I'm very much in favor of blending BI resources and digital resources. But that doesn't always imply blending the organizations. I'm not sure I have a one-size fits all recommendation except that some level of resource type blending is absolutely critical to effective digital analytics.

In most cases, I think I'm in favor of merging those teams. But perhaps even more important to me would be ensuring that the digital team has some BI expertise working inside it.

Your point/question on technical team roles - particularly in a de-centralized environment probably deserves a full discussion. Though I'm a generalist at heart, I find that most large enterprise functions need to be fairly specific (going back, I suppose, to the point about how different big-company and entrepreneurial organizations and resources need to be). That means I'm generally in favor of not trying to have folks split roles - especially across technical and non-technical boundaries. It makes resourcing much harder, though when you can get those folks they can certainly make life a lot easier. Usually, that means that de-centralized analysts need to interface into centralized analysts who have direct relationships with technical folks or directly to the technical teams.

Anyway, your questions certainly deserve deeper thought and I'll see if I can think more about it!

Gary

Thanks Gary,

I really enjoyed this post and completely agree with the idea of having both centralized and decentralized analytics. In my experience, you often here people talking about making a choice between the two options but in reality that approach doesn't work in large organizations. You need a mix of the two.

The centralized team is needed to configure the reports, define metrics and set standards for the organization. The centralized team should be the authoritative voice for analytics and insight within your company.

When you get down to the granular detail of specific areas of the business you really need the people who are living and breathing that project or area to be using the data. This is really important as the centralized analytics team will never know everything about every area of the business and if you don't empower people to self serve with the data then your efforts will likely be seen as interference.

Thanks again for a great post Gary.

Best regards,

Billy Dixon

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Comments are moderated, and will not appear until the author has approved them.