The Algorithm Economy Heads To Amazon

Holidays are a time for families to come together, catch up over great food and drinks, and determine all the technical problems that need solving throughout the house. Indeed, for children growing up in the digital age, the holidays ultimately boil down to free (or more accurately, meal-subsidized) technical support for our most cherished loved ones.

Kids: Amazon has you covered.

Well, almost. This past week, Amazon publicly introduced an early release of Selling Services, which we had previously mentioned the company was working on a few months ago. Amazon is developing a marketplace that offers after-sale services such as car alarm installation, iPhone repair, and computer hardware setup to consumers buying relevant products. Today, the marketplace is available in 15 early rollout cities, including New York City and Lexington, Kentucky.

After-sale services are among the highest profit margin revenue streams for retailers, so it is little surprise that Amazon is jumping into the fray. Geek Squad, which was founded by Robert Stephens in 1994 and sold to Best Buy in 2002, is perhaps the most prominent example. As the Minneapolis Star Tribune wrote last year about the hometown retailer, “Over the past decade, Geek Squad has been a cash cow for Best Buy. […A]nalysts estimate Geek Squad generates a gross profit margin of 40 to 50 percent based on a minimum annual revenue of $2 billion, or about 4 percent of Best Buy’s total revenue of $50 billion.”

That profitability of after-sale services has also attracted the attention of startups. One particularly notable example is Geekatoo, which is building a marketplace for technical support and raised $1.7 million this past July. And in a certain kind of disruption irony, Stephens, the founder of GeekSquad who moved to Silicon Valley after leaving Best Buy in 2012, recently recommended the service.

But Geek Squad still has one critical advantage that many of these startups lack: point-of-sale access. After customers have purchased a new flat-panel television set or computer, there is a precise window of time to offer them installation services where the rate of acceptance is significantly higher than it otherwise would be. Few customers later decide to search for a service specialist, hence the need for children to head home for the holidays to install and fix technology.

This is why Amazon’s announcement is so important. Given its dominant position in online commerce, the company’s decision to offer a services marketplace could literally create thousands of jobs across the United States, or at the very least, improve the prospects for specialists already working in the field. For each product, Amazon will list the available services next to the listing, guaranteeing visibility and even potentially increasing sales among customers who are unsure if they can install or use a product.

Perhaps even more importantly, the prices of these point-of-sale services can be much higher than they would otherwise cost, thanks to framing effects. After someone has just spent thousands of dollars on a new device, adding a few hundred more for installation service doesn’t sound like such a bad deal. That’s what made Geek Squad such a profitable division for Best Buy.

That’s also why Amazon’s decision to run Selling Services as an open marketplace is so surprising.

In my research, I split labor marketplaces into two groups: versions 1 and 2. Version 1 marketplaces, which include the newly combined Elance-oDesk, Craigslist, and the original TaskRabbit, are open marketplaces where people can request services or sell their services at any price, and consumers wade through pages of relevant listings to find exactly the level of service quality and price they want.

These marketplaces have many challenges, not least of which is that consumers are terrible about judging quality from online postings. In many marketplace verticals, there can be a race to the bottom as customers rank services by price and then select the cheapest options. Ironically, this behavior is bad for the marketplace as well, since many consumers would be willing to pay more if they had fewer options.

These models compare to version 2 marketplaces like Uber, Rev, Scripted, and the new TaskRabbit, where prices are set flat or with a formula, and the quality of service is guaranteed by the marketplace itself. There is no haggling over price or being overwhelmed by the sheer volume of options, but often only a single option or a small handful to choose from. On the whole, these version 2 marketplaces tend to have better customer satisfaction and arguably higher profits.

Interestingly, Amazon’s marketplace doesn’t seem to follow these more recent version 2 marketplaces, but is instead more in line with the version 1 model. Amazon is not prescribing prices for its different services, instead offering an open marketplace where service providers can charge what they want and compete. The choice is no doubt familiar to the company given its other marketplaces.

That’s unfortunate, given that Amazon already has a tendency to cause buyer fatigue due to their massive product catalog. Given the number of customers using Amazon, I imagine many service providers will join the platform, meaning that there will be a surfeit of options for every service in each geography. Now, buyers will be expected to make another purchasing decision as part of their checkout process. I have little doubt that many potential consumers of these services will be left on the table.

The good news is that this is a beta product, and Amazon has a lot of time to get the mechanics of the process right. Unlike startups in this space who have to seek both sides of this market, Amazon has a guaranteed demand-side in its marketplace, so it has plenty of time to figure out how to best develop the supply. That won’t just make after-sale services easier, but save the holidays for millions of children across the country.

Now, if you excuse me, I need to get my mother’s new Chromecast working. Update: We got it working.