I have been thinking a lot lately about retail as everyone has.

The reaction by off-line retailers has been to create or up their online presence to compete with potential new threats coming their way. If I hear the word omni-channel marketing from another retailer I think I am going vomit.

Having said that Amazon is developing out a bricks strategy with the purchase of Whole Foods. Alibaba is rolling out the Hema model in China that combines off-line and on-line.

My feeling is that Jeff Bezos has a vague idea of how combining off-line and on-line might work but he will be driven by his blueprint of idea iteration-change-test-feedback-repeat loop. Making small changes, testing these changes seeing what works seems to be more his style. Lots of little failures and iterations and then some home runs thrown in. I would be surprised if it is not the same here with Whole Foods.

Online retailing in the US Q2 2017 represented 8.9% of total retail sales according to the US Census Bureau so there is a long runway if it tips to the 20% threshold of mainstream penetration. So it’s an important strategy for a company to get right.

So that takes me to the idea I am grappling with. Is it good strategy to build out an online presence if you are an existing offline retailer? For management it may seem good strategy as they can point to shareholders and say Hey look we are doing something to combat online retailing.

To me this strategy depends on the type of product and the market you are servicing. The key for a company to consider is whether their offline presence is servicing a vertical marketplace which cater for a small subset or specialised items or a horizontal marketplace which sell a broad range of items.

Josh Breinlinger sums it up nicely when he states:

“There are two factors that matter most when looking at whether a horizontal or vertical marketplace makes the most sense: the size of the transaction and the frequency of the purchase”.

The following graph gives an illustration of how certain online marketplaces are better suited to catering for purchase price and frequency:

So if you have a product that has a high ticket price and has low frequency purchase then it makes sense for a company to specialise and invest in this online vertical. Realestate.com can have a focus on selling houses and nothing else. Carsales.com can have a vertical on selling cars. Vertical marketplaces make sense if you live in the top left and the bottom right of this chart.

On the other hand would you have a vertical marketplace selling only shoes or tents? Do you as a consumer want to change platforms or websites every time you shop for multiple consumables? Probably not and this is why a horizontal platform is best suited as you can shop for multiple products on one website. Amazon is the best example of collecting these small vertical markets into one horizontal platform.

So I have found this chart potentially a good mental model to use. Next time I see a retailer talk about developing their online strategy to complement their offline business, and their product is low ticket size and they specialise in only this product, then I am wary as this product would make more sense being on a multi-sided horizontal platform.  It would not be a good defence strategy for this company to build a vertical online shop.

Maybe this may work if you have a great brand and customers will search you out and find your marketplace. But if you have a great brand due to your distribution advantage then think again.

Put it this way I do not want to keep track of multiple logins and go to 20 sites to buy cheap items every now and then. I will move towards one horizontal platform that does it all. Especially if that horizontal platform has scale to offer me free shipping. But for an expensive low frequency purchase then I am happy to search out a specialised market place that caters to this niche.