Sitting here talking with Tara about her new VRM business. Lots of helpful ideas bouncing around. So we’re both pausing in the midst to write stuff down. Here’s my brain dump of the moment, with some actionable ideas toward the end…

For retailing, the Net changes everything. But it’s still new. It’s three seconds after the Big Bang and all we have are a few light elements, a lot of heat and no galaxies. Yet $billions are already being made in online retailing, and$billions more are being spent and saved by retailers and shoppers using the Net to advantage. And because of those $billions, and the successes of companies like Amazon and Zappos, and services like Google Checkout and Orbitz, we’re inclined to think this stuff is mature. It isn’t. It’s still embryonic and protean, compared to what it will become.

In the meantime, consider this thesis: Amazon and other excellent online retailers have improved the online shopping experience as far as a retailer can. Yes, there is always room for improvement, but there is only so much improvement you can carry out only on the sell side, even if you’re equipping buyers to do a better and better job. At a certain point the improvements need to happen on the buy side. You need better buyers, not just better sellers. You need to improve the tools available to buyers — tools that help buyers with all sellers, and not just within each seller’s walled garden or silo.

Therefore… At a certain point the problem is no longer scale but scope.

Amazon and its competitors are pushing out the envelope of sell-side scope. On the buy side we’re just getting started. The envelope is still mostly empty. The job of VRM is to start pushing out its walls, starting at close to zero.

Another way of putting it. There is only so much any retailer can do, because they are sellers and not buyers.

Still another: It doesn’t matter how big you make a walled garden if it’s still a walled garden. At a certain point you reach the Multiple Walled Garden Problem. (Shall we call it MWGP? Or SO for Silo Overload?)

If you’re carrying a pile of retailer loyalty cards, you have a silo overload problem. It may not be a huge one for you, but it’s still a problem. It’s friction, and not just for you. Loyalty cards can be a PITA for the seller as well. They require multiple pricings, slow things down at the cash register, and involve piles of often wrong and irrelevant data. But I don’t want to go into how good or bad loyalty cards are here, because they’re beside a larger point: that we need to start solving market problems from customer’s side, by improving the scope of what the customer can do in the same way with multiple vendors.

For example, take affiliate programs, or affliliate marketing. Tara has been schooling me about these things, which are a huge part of how online retailing works. Hell, I didn’t even know that when I clicked on an a Head Butler link such as this one, Jesse Kornbluth gets a kickback (or at least puts himself in a position for one) from He’s not just pointing to a book. He’s part of a new retail system in which commissions or kickbacks (or whatever you want to call them) are silo’d. Amazon has one kind and other retialers have other kinds. Some have none. Whether he means to or not, Jesse discriminates against those, and does so for financial reasons.

From the sellers’ side this is all fine in the sense that it’s a free (and fee, I suppose) marketplace. Every retailer is at liberty to compete by providing the best kickback system.

But what if the customer wants, say, purchasing guidance that’s uncontaminated by bias toward one kickback system over another? What if the intermediary guides the customer to a seller that doesn’t have a kickback program, and the intermediary gets nothing from the sale while the seller gets everything? Wouldn’t it be better to have the buyer (or the intermediary, on behalf of their buyers, and for the good of the marketplace) assert a single form of commissioning that’s fair and helpful to all sellers and all intermediaries — even while respecting the kickback (or commissioning) systems that are already in place?

This is a greenfield here. Let’s think and talk about it.

Also, let’s think about what kind of research project this might make — or that the theses presented here might make — for a business school student or class (at HBS or elsewhere). Because that’s one of the things I’d like to do in the next school year, which is just getting started.