Finally I disagree with Avinash!

It had to happen one day! I have been reading this excellent blog for some time now, but today finally I saw a post which I have issues with.

Avinash starts by saying that Engagement is not a metric, it’s an excuse. Then goes on to make comments which I think are well in a word wrong.

Engagement is not a metric that anyone understands and even when used it rarely drives the action / improvement on the website.

To me this depends on who you’re working with, how you’ve defined engagement and what your KPI’s are. MANY of my clients ARE using engagement KPI’s to optimize campaigns spend. Avinash’s point is that it’s used as an excuse by people who should be improving conversion rates of sales is pretty short sighted. Sales are rarely made on the first visit as he’s said many times in previous posts therefore engaging the visitors and answering their questions is not only needed, but imperative to improve sales.

But let¢‚¬„¢s try to understand why in the context of web analytics so many efforts at measuring ¢‚¬Å“engagement¢‚¬ have yielded almost no results…

Speak for yourself. Maybe your clients and colleagues are driving “no results” but mine have. I would suggest your engagement KPI’s are not well defined and therefore the information you have is not giving you the insights you require, not that engagement per se is at fault.

It is nearly impossible to define engagement in a standard way that can be applied across the board.

True, but who ever said it was possible to define an engagement standard? Who ever said that was a good idea?

At the heart of it engagement tries to measure something deeply qualitative

False. I have used quantitative engagement data to drive things like surveys and other qualitative studies, but never assumed the figures are trying to replace qualitative data. It’s simply wrong to do so, the data should compliment each other. Engagement metrics can be as useful as flags to take action on doing qualitative stuff like surveys, usability studies, heuristic tests and such like. But to simply say “replace engagement metrics with qualitative data” is naive. Why then do web analytics at all? Why not just guess at when to run surveys and usability studies? Expensive testing needs justification to suggest that a test is required in my experience.

One of my personal golden rules is that a metric should be instantly useful. This one is not.

Again, i completely disagree. I have seen incredible gains (millions of euros) simply by acting on engagement data with some of my clients. Again it’s got to be down to how you have defined your engagement criteria.

Most of all engagement is a proxy for measuring an outcome from a website

I agree with this, but so is measuring effective direct sale reach sources, so is measuring the effects of nurturing existing customers, so I miss the point you’re trying to make. All business (ALL) is about activation at the end of the day, everything else supports it. Surely you’re not trying to say that all measurement other than activation is a waste of time?

If we are measuring page views divided by unique visitors as a proxy of engagement

Page views per visit is one form of engagement, one miniscule minute form of measuring effective engagement, but as Avinash knows is by no means the end of the story. Rahul (the second comment on his blog) hit on a couple of the metrics we’ve used in the past when he talks about scoring individuals. We’ve segmented audiences based on session time for instance more than 30 seconds but less than 10 minutes. There are a number of others we’ve used, keyword engagement based on the above sessions for instance driving changes to SEM strategy. All of it is good and actionable data when used correctly and when defined well.

Avinash then goes onto re-define engagement in his own terms. Doing exactly what he mocked earlier in the post;

Pundits have pontificated. Bloggers have blogged. Guru¢‚¬„¢s have spoken from their perches. Industry Analysts have given their brains to the cause. Vendors have¢‚¬¦. well tried. Hard.

What concerns me is that people have been agreeing left right and centre on Avinash’s blog that his article was right on the button. I disagree with those guys too. Sorry but I know for a fact if I asked most of my clients to do expensive qualitative data gathering with no justfification, no evidence to suggest a problem could exist then it is simply wont happen. What concerns me is the message this is giving to the masses. To suggest that engagement is a waste of time with powerful rhetoric like engagement is an excuse is in my view a bit off. We should be promoting the effective use of engagement metrics, not belittling their worth.

I for one know for an absolute FACT that engagement metrics when used well can lead to great results. I can think of three clients in the last couple of months in which engagement KPI’s have directly saved or made them money by taking action on the flag raised.

I am slightly surprised at Avinash’s post because it seems to be on the one hand mocking the idea of engagement and then on the other defining ways to measure it. So I don’t actually know whether I agree or disagree with him come to think of it. Maybe I should re-name this post, “Finally, I think I might be disagreeing with Avinash!” :)

Steve is a well known analytics specialist, author and speaker. A pioneer since 2002, he established one of the first European web analytics consultancies (Aboavista), later acquired by Satama (now Trainers’ House) in 2006. In 2008 he wrote his first book Cult Of Analytics published on May 14th 2009. He currently serves as CEO at Quru and has presented and keynoted web analytics topics across Europe. These include The Internet Marketing Conference (Stockholm), The Search Engine strategies (Stockholm), IIH (Copenhagen), the IAB Finland (Helsinki), Media Plaza (Amsterdam), Design For Conversion (Amsterdam) The eMetrics Summit (London, Munich, Stockholm), Divia (Helsinki) in addition to sitting on dozens of panels.

Posted in General, Web Analytics
5 comments on “Finally I disagree with Avinash!
  1. Weird feeling to be disagreeing with Avinash, eh? Don’t worry, he won’t hold it against you; the guy loves a debate.

  2. Steve: I have come to deeply appreciate the power of blogging in terms of having a conversation, with people you don’t know and might never get to meet. It is amazing that the medium empowers this so effectively.

    I have benefited tremendously from feedback from peers and thought leaders in the little web analytics fish bowl.

    I welcome your feedback and thank you for adding to the conversation.

    -Avinash.

  3. Steve,

    I agree with you and had many of the same questions you raise in your post. I was at SEMphonic X Change week before last and heard several great examples of how engagement metrics have been successfully used on both a very large and also a very small scale.

    I’ve thought a lot about measuring engagements using reasonable proxies and like you am absolutely convinced it can be done. Maybe not using Google Analytics, but using reasonably powerful tools it is possible to build a more informative metric than just page views per visit.

    Anyway, nice post!

    Eric T. Peterson
    http://www.webanalyticsdemystified.com

  4. I have added this after a day of thought around the subject (and also posted this as a direct response to Jim Novo’s post saying that it was a semantic issue)

    It probably is semantics.
    Engage to me is when you hit the first touch point and your actions after that. Measuring this is very important, and we’ve got quite sophisticated with it. Way beyond page views per visit. Nurturing the customer is what I see Jim Novo’s field as (simplistically speaking). Engagement to me is when the visitor engages with the creative and what he/she does.

    To most web analytics folk I’ve worked with this is what engagement is. Forrester agree, so do TNS, so do prominent analysts and customers I’ve worked with.

    So my point and hence my stance is the exact point Avinash was trying to make, why call engagement something it aint? At least according to everyone else?

    Except perhaps advanced BI folk that do in depth CRM, predictive data mining work like Jim Novo and Avinash. My customers don’t have too many of those people sitting in their offices. They understand engagement as point of contact.

    Avinash’s post seems to strongly disagree with this stance and belittles the approach (at least the approach using the name engagement) which I know works which is why I reacted to it. Nurture phases (measuring engagement with the customer and measuring what they don’t do) requires a lot of qualitative data which is typically much more expensive to do. Without measuring what I call engagement first I can’t justify that expense to my customers.

  5. Sharad says:

    Steve,
    A good blog and rightly said, one rarely gets to disagree with Avinash but this one, I do feel that there are gaps.

    I view engagement as the time a person came in contact for the first time with the business and till the time, this person becomes a customer to be an upsell candidate or repeat customer to be engaged in a different manner. It is a continous process.

    If we really do not measure engagement across channels then as a whole we might miss the complete implication of a channel. Very often, rather most often people do all their research about a product on a Web site and go to some local store to buy the product. In such cases though the sale happened through a store but Web site played a major role in engaging with the customer. The Page views/Product view for a particular product from a particular region and the sale from a store in that are can be corelated to identify the role that a web site played. I would say if we miss this piece of information then we are really not tying all our channels together.

Leave a Reply

Your email address will not be published. Required fields are marked *

*

Subscribe

Enter your email address:

Delivered by FeedBurner

Recent Tweets
Site Sponsors

Find out how we're trying to make our clients €1bn, click below!