The Truly Pathetic State of Mobile Commerce

Do you know why Internet Retailer’s Mobile 400 List isn’t “The Mobile 500”? Because of all the companies Internet Retailer reviews to compile its famous lists, there weren’t 500 retailers with mobile sales worth mentioning. The last company in the list, #400, did just a little over $17,000 in mobile revenue in all of 2012.  If that’s not pathetic, I don’t know what is.

Don’t get me wrong, I don’t want my money back. I think this was particularly insightful to see, and glad they published it.  It was also interesting to see that most of the big winners in mobile commerce are in travel and entertainment. If you’re in that industry, I would take mobile commerce seriously. Hotels, airlines, rail travel, vacations, car rental, movie tickets, concert tickets. Basically ticketing and reservations. That’s not to say that the top of the list isn’t peppered with huge retailers like Walmart and the like, but if you were to remove the retailers who are just doing 1% to 2% of their total online revenue on mobile, the top of the list would be almost exclusively ticketing and reservations. Think about what this says.

While Internet use is obviously trending away from desktops to mobile devices, mobile Internet usage is fundamentally different than desktop. People seem most comfortable using their mobile phones to transact when they don’t really have to “shop” or closely look at or evaluate what they’re buying.

They just know they want to go to a particular place at a particular time, and are fine closing the transaction on mobile. What drove them to decide to stay at that hotel or watch that movie or take that flight did not originate online or from a mobile search engine.

If you’re not in ticketing or reservations, here’s my generalizing takeaway. Large brick and mortar brands that launch an e-commerce site to sell online without any particular effort, do generally about 2% of the company’s overall revenue online.  It would be zero for most but because of the brand equity from the brick and mortar, about 2% trickles down to their existing customer base who would prefer to transact online. These companies are so big to begin with that even 2% of their revenue translates to hundreds of millions of dollars.  Next time you’re impressed hearing that Walmart.com is on track to do $9 billion online, remember that Walmart’s total revenue is $447 billion, so they’re right on track to claim their default sales of 2%.

Mobile seems to have the same relationship to desktop sites. Mobile revenue seems to be about 2% of online revenue. That tells me that just like big brick and mortar brands who launch an e-commerce site and make only 2% of their revenue online don’t understand to run an online business, online businesses who launch a mobile site and make only 2% of their online revenue via mobile don’t understand mobile.

Take a look at this example with 2012 revenue numbers:

In store revenue vs. online revenue vs. mobile revenue, 2012

From seeing this pattern, you might joke that if the future Internet enabled watches have special apps to shop on them, that you might then take 2% of the already small numbers being done in mobile and attribute that to watch device sales.  It’s almost like the smaller the device, the smaller the sales.

Desktop Geolocation – How valuable is ‘sort of’ knowing a users location?

I think Geolocation services are hugely important.  Since we’re all now used to the idea that our smartphones know our exact location and can immediately serve up content based on that, it almost seems prehistoric to have to enter your ZIP code into a web form for a website on your desktop computer to know where you are and serve content based on your location.  If my little phone can know exactly where I am, why can’t my big expensive desktop?

So sites are clearly doing geolocation so you don’t have to enter ZIP code.  But they’re doing a fairly bad job of it.  Desktop based geolocation services have 99.9% accuracy within a 25 mile radius.  While certain providers will have different accuracy guarantees and there are different data sources out there, all of them share the 25 mile radius limit.  Think about what that means for a moment.  The desktop knows where you within a 25 mile radius.  Not very accurate.  This is how we serve up “Results from Pleasantville” when you are really 20 miles away from Pleasantville.  Better than having no idea where you’re at but not terribly useful.  There are at least a couple important implications of this to consider.

Implications on Advertising by Geography

If you’ve ever tried to limit your AdWords, display, or other advertising to a geographical location, if you’re trying to limit your reach to areas where your conversion rate is highest or only to areas where you have brick and mortar locations, consider that it’s only going to be accurate within 25 miles.  So you can forget about being accurate down to the city in smaller cities.  You’re accepting that you’re going to allow in some traffic that is not actually in the geographical area you think you are targeting, making it less efficient than it should be.

Implications on Store Location Services

Many e-commerce sites from brick and mortar chains are challenged by having different pricing in different stores across the country.  In most cases before these sites can display a retail price they need to know your location, involving the user to have to enter their ZIP code and select their store, a few extra clicks and user interactions just to be able to see a price.  This can hurt engagement rate and conversion rate.  So a growing trend is to use geolocation services to automatically select the store closest to the user, and present this allowing the user to confirm the location or pick another so prices can immediately be shown.  This is smart but has drawbacks for retailers for lots of stores within a given area.  Remembering that desktop services are only accurate within 25 miles, it is likely that the “wrong” store may be automatically chosen.  While the UI can prompt the user to confirm or pick another closer store, users don’t know about the 25 mile limitation and may assume their desktop computer is as good at knowing their location as their cell phone is and assume that must in fact be the closest store to their home and not bother checking.  This concern can be mitigated with messaging however it is a key concept to keep in mind when employing this tactic.

Will the future be brighter?

Given the way desktop geolocation works to find your location, it is unlikely that the service will improve much unless desktop computers start shipping with new hardware with a cell signal so that location can be triangulated.  In the meantime, more and more Internet use is going to mobile devices including tablets.