This isn’t the first time a blog post has carried the title “Is OpenTable Bad for ____?”. The debate over the real-time reservation management platform and the lift it provides restaurants has been raging for well over 5 years. I’m not a restaurateur but I understand the importance of maximizing revenue, especially in the restaurant industry.
As a user, I appreciate its usability and how it’s prevented me from never being denied a table by a snooty maître d’ over the phone.
As an online marketer, it has been one of the most frustrating platforms I’ve ever used.
A client in New York City recently opened a beautiful new restaurant in Midtown Manhattan. It’s a great space that’s receiving wonderful reviews and visitors are flocking to it. After some internal debate, our client settled on OpenTable to manage their online reservations. Having had little experience with OT, I jumped at the chance to attribute our ad spend dollars to booked reservations. As our dev team began the process of adding the OpenTable widget to the website, I dove into all the OT resources I could find to ensure accurate, closed-loop reporting. This was easier said then done.
The first obstacle I encountered was that OpenTable doesn’t allow third-party code on its pages. This prevents the implementation of AdWords conversion codes, Google Analytics event tracking, or DoubleClick Floodlight tags to track completed reservations. These pieces of code are incredibly important to online marketers and without them we’re left more frustrated than Leonardo DiCaprio at the Academy Awards.
I combed through tutorials, searched for answers, and finally submitted my issue to the AdWords community. If ever in doubt, go to the community. From the 592 views it received, it was clear that I wasn’t the first person to be stung. With some direction from my peers, I felt confident I’d found a work-around.
Luckily, it was decided we’d build our own on-page widget within the website that users would engage prior to being redirected to the OpenTable system. That meant I could implement all the event tracking and conversion codes my heart desired. This might actually work better than I would have hoped.
In the end, we decided to track users that completed all of the parameters of our on-page widget (Day of the Week, Number of Guests, Time of Reservation) or what we call “initiated reservations” and then we could use OpenTable’s own reporting interface to attribute booked covers to our initiated reservations. We’d have conversion percentages to give us even more insight into user behavior and drop-off from the website. Unfortunately, we overestimated OpenTable’s reporting.
OpenTable reports on covers that originate from either its website or your own. That’s it. Not by traffic source, medium or campaign. I can’t tell if a user booked from an email, paid ad, or organically. Those campaigns complete with A/B testing and confidence intervals will go unattributed. With my hopes dashed I reached out to the OpenTable community and I did get a response —
“Unfortunately at this time, we do not have a method in place for getting that sort of traffic conversion data over to our users. We absolutely understand the importance of this and it is something we are working towards for the future.”
I did appreciate that they responded to my concerns. The future described by the OpenTable Community Administrator is bright and full of possibilities: where conversions are tracked at (dare I say) the keyword level. In the meantime, I’ll swallow my pride and report on “initiated reservations” through gritted teeth. It leaves a bad taste in my mouth that will only be eliminated by direct attribution.
So, is OpenTable bad for online marketers? I won’t go that far. It is certainly frustrating. It appears they recognize the limitations of their platform and are working on resolving those issues. I’m also sure online marketers aren’t their top priority, but I look forward to the day when we can all come together and in unison say, “Our March e-blast drove 20 booked reservations.” That’s the world I want.