Eating Opentable's lunch

A recent post by a disaffected San Francisco restauranteur complains eloquently that Opentable is sucking the whole margin out of the restaurant business just for processing the bookings - about $10 on a four-cover booking, which assuming that table spends $200 at a normal 5% margin means Opentable is keeping all of the profit on the booking leaving the restaurant with, err, nothing. If that's true it's bad for restaurants but bad for Opentable too - a parasite that kills all of its hosts doesn't have much in the way of long-term prospects.

The bad news for the restaurant industry is that the problem of Opentable carving out a position for itself as a new middleman in the eating-out value chain arose because of a basic strategic error made by the restaurants themselves. The good news is that the error is relatively simple for the restaurants themselves to reverse.

Restaurants charge for meals but give away reservations. Since the reservations also have value, this creates two problems for the industry. First, pre-digital restaurant reservations were broken. To book a table you rang round some restaurants, you found one that had a vacancy on the day you wanted and you gave the guy on the other end your name. That's what the restaurants themselves decided was the way to allocate tables - a first come, first served queue, with the exclusivity or scarcity of specific establishments merely expressed as the need to book months in advance (or have appeared in OK magazine sufficiently recently for the doorman to recognise your face).

This was a wearisomely inefficient customer experience, and more to the point the economic surplus the restaurants created by giving potentially valuable reservations away was simply handed over to whoever happened to ring them up first (in other words a combination of people in the know, people with good PAs and lucky idiots - almost pure social waste).This created an opportunity for Opentable - along with a number of very similar competitors such as Toptable and Urbanspoon - to build a central marketplace where none previously existed by aggregating the bookings system and transferring the surplus that the restaurants had left on the table to themselves.

More importantly, there was - and still is - no secondary market in these valuable reservations. Partly because they have no face value and partly because the coordination problem has simply never been solved, a reservation made three months ago for a table tonight at the Ivy might be worth ten or fifty or a thousand pounds to someone else today but if it turns out I'd rather have the cash than the chance of dining four feet from a B-list celebrity there is no way of realising that value. Unless I happen to value my reservation at more than the (currently-unknown) market rate, it's simply a waste that my only option is to turn up to dinner rather than sell my place on. By the same token, I might be willing to pay a hundred pounds to dine at Hakkasan tonight, but because the reservations have been given away weeks ago the book is full and I have no way to convey my willingness to buy to any of the hypothetical sellers whose names appear therein. 

Opentable solves one of these problems but not the other - it has a created a sort of primary market for reservations in which the restaurants pay it for diners. If the restaurants want to take back the economic surplus they once handed over to diners and have now handed over to Opentable, they need a pricing strategy for reservations that goes beyond merely giving those reservations away to the first person in the queue. Instead of giving away their valuable reservations, they should make those reservations a transferable currency (rather than a name scribbled in a book) and then distribute them via an online auction platform. Then they should withdraw, preferably collectively, from Opentable. A real (primary and secondary) market in restaurant reservations would make bookings more of a commitment to turn up; it would satisfy customers by making their reservations potentially more valuable as the due date drew near; it would incentivise satisfied customers to book again and then widely promote the excellence of the place at which they'd eaten; and of course the restaurants could take as large a cut of the new market as they wished. Better still, it would let them leave Opentable out of the feast. 

 Update: more discussion of the idea by Citywire's Rich Harris is here:  Rich comments

"it would make the experience of restaurant booking similar to that of booking planes and hotels - you have to book far in advance to get the best deal and even then you feel anxious that you could have done better. Or worse, you'd have to deal with the restaurant equivalent of ticket touts."

No disagreement here - as I've argued lots of times before, tickets are another product suffering from massive underpricing in the primary market.

(Photo from informatique on Flickr)