Subscribe to RSS News Feed

Drop Us a Line

blog@demandware.com

eCommerce Innovations Blog

Current Articles | RSS Feed RSS Feed

The Upside Down Pyramid

Digg digg it | Reddit reddit | del.icio.us del.icio.us | StumbleUpon StumbleUpon 

If there's any upside to a commute it's that it provides at least some opportunity for idle thinking. So on the way home last night, breezing North on 93 and beat down by Red Sox doom-and-gloom sports talk, I snapped off the radio and started to idle the old squash. Free form association kicked in and when I blew through the yield sign (nearly T-boning a lime-green hybrid in the process), I thought, "Wow! Now that's a much better model for ecommerce!"

Ok, so maybe I should back up and explain that a bit.

 

Over years of depicting the costs and effort allocation of a typical ecommerce operation, I had evolved into using a tried-and-true pyramid diagram. You know the one. At the bottom was the hardware and infrastructure. It was wide and thick, signifying gobs of energy and cost. Stacked on it were databases, app servers, IT professional support, point applications and other such things-each showing diminishing time and effort. Finally, perched at the very top, with the smallest land mass, was merchandising, marketing and site experience. It always seemed wrong that such important activities would get such a relatively small piece of the pie, but hey that's how things had to be. In fact, analysts point out that up to 80 percent of ecommerce resources (dollars, time and people) are spent simply maintaining the existing infrastructure and processes.

So wrong.

Effectively this means that 80% of an ecommerce operation is going to activities that yield absolutely NO competitive differentiation. For instance, when was the last time a customer went to a site and said, "Kudos. It's up!" or "Hooray! It has secure checkout!"? Likely never and yet that's exactly the type of activities that are getting the most resources.

And that's where the upside down pyramid comes in.

Wouldn't it make more sense if the basic table-stakes portions of the ecommerce operation took up the minority of effort? All the behind-the-scenes work that goes into a professional operation-security, compliance, hardware and database performance, application upgrades etc.-was just there? And instead, retailers could spend more time in the areas that will really drive top line growth. Price management, promotions management, presentation of the assortment, UI innovation... You know the list. This is the stuff that consumers really care about.

So I challenge you commuters out there to make your mindless travel time more productive. Ponder the tough questions and voice your opinions. While no one has the answer, everyone has an opinion-and stringing together enough of these thoughts and opinions can lead to a breakthrough big enough to change the game. Heck, it worked for the Sox, why not us?

What would revenue look like for most operations if these were the areas that received not 20% of the effort, but 80%?

Are there really people out there who would prefer to spend their time worrying about site uptime and secure checkouts rather than marketing, merchandising and selling their products?

What steps are you taking to turn the ecommerce status quo on its head?

Tags: , ,

COMMENTS

Great article Jamus. I suspect etailers were spending disproportionate time maintaining infrastructure five years ago. With commoditisation of most components more time is now being spent running real differentiators like customer service and inventory. Those that are still spending most resources on tasks with the lowest revenue yield are clearly doing something wrong.

posted @ Sunday, October 19, 2008 11:03 AM by BetterRetail


Thanks for the comments. Yeah, you're right and we've seen this trend--big and small--in the evolution of every industry. Cottage industry gives way to industrialization. Commercial databases like Oracle take over for home-grown databases. Now it's ecommerce's turn. Should be an interesting couple of years

posted @ Friday, October 24, 2008 3:52 PM by jamus


Post Comment
Name
 *
Email
 *
Website (optional)
Comment
 *

Allowed tags: <a> link, <b> bold, <i> italics

Receive email when someone replies.