This Is the Modem World: Review Sites Are Broken

Each week Joshua Fruhlinger contributes This is the Modem World, a column dedicated to exploring the culture of consumer technology.

I'm about to walk into my favorite restaurant in the world. The food is yummy, healthy and fits my budget. The staff knows my name and rarely has to ask how I'd like something served. And the location couldn't be better. It's the perfect place.

But before I can pass the threshold to nosh nirvana, a stranger on his way out gives me a sideways glance and whispers, "This place is horrible. Don't bother."

This place? My place? The best restaurant on earth? How can this be? Perhaps something has changed. Did the chef leave? Is the cute hostess gone? Have I been wrong this whole time?

But I have faith. I follow through, and the best restaurant in the world is still just that, just as I remembered.

And yet that other guy had a rough go at it. Maybe he doesn't like cumin. Maybe his grandfather used to make him clean his plate full of shishito peppers, all of them the hot ones. Sounds stupid, but it turns out people are different and have divergent opinions; I love the place but he won't be back. Humans disagree, and that's what fuels the raging online communities, political systems, wars and sporting events. It's our difference potential and it creates social energy. Yay.

It's also why I am done with community review sites like TripAdvisor and Yelp. They simply don't work for me. They mute disagreement. I've realized that I don't care what random people with enough time on their hands to write online reviews have to say about my favorite restaurant because I already know I love it and they're stupid heads.

Indulge me on this next sentence.

Most review sites are engines of social indifference and the net results are based on controversy and business interest rather than real reviews.

I recently returned from a honeymoon in Bali. It was a lovely time, but only because we dodged a bullet days before we left. When I was booking our accommodations six months ago, I relied heavily on TripAdvisor to point us at good choices. We decided on a new place that was listed as the "No. 3 Hotel" in the area. It had recently opened, already had about 20 positive reviews and the price was right. We booked a room and moved on.

Cut to the night before our rehearsal dinner when I was printing out our itinerary and confirmation emails. When I double-checked the hotel's address on TripAdvisor (their Google results are often ahead of the first-party site), the hotel had dropped to No. 34 and was offering a 50 percent off special. Turned out the place was still under construction, the pool was tiny and the staff were still learning how to run things. I made a quick move, found another place and we ended up having a wonderful time.

We stopped by the original hotel in question, and, indeed, it was still under construction. And yes, the tiny pool was jammed right next to a small cafe that served as the breakfast nook. We walked back to our spacious villa, relieved.

So what happened? Seems that the early reviews were juiced, probably by owners and others with some skin in the game. Because the hotel was new, TripAdvisor's algorithm saw strong upward momentum and gave them a Top 10 spot (I'm guessing here -- I don't have access to their code). Then, just as quickly, when the bad reviews came in, the math dropped the hotel back where it belonged.

This is arguably a fair way to handle community reviews -- weight reviews over time and look for velocity. But the problem is that it encourages businesses to play the game and juice reviews. Meanwhile, those of us who stay at places and have a decent time probably aren't writing reviews. Instead, we're left with hyperbolic review headlines that declare the place the best thing since free shampoo. Ask yourself: how many times have you written reviews about places you've stayed or eaten at? I'd bet it's either "never" or "all the time." If I'm right, that means we're left with polar information.

It's probably unfair of me to criticize review results and yet admit at the same time that I don't submit reviews. But that's just the thing: those who are motivated to review a hotel (or restaurant for that matter) typically do so because something wonderful or terrible happened or they were paid to do so. It's also true that had I not seen the negative reviews before I left, I would have ended up at the place, miserable and mistrusting the site even more.

But here's where things get weird. All those negative reviews from six weeks ago for this hotel are now gone. Somehow they were deleted, probably as a way to give the new hotel a mulligan and a chance to get its act together. That's fair, I guess, but I hope that some fresh traveler isn't stuck with construction and a tiny pool in paradise.

Either way, community review sites have failed me. And, to be quite honest, I'm not sure I care what someone I don't know thinks about a hotel. Maybe someone else doesn't care about the things I do, and I don't care about the things that impress that person about a hotel. Maybe that person owns the place. In the end, I don't trust the results.

For now, I'll be asking friends rather than chancing my meals and vacations on ninja marketers and broken algorithms. It just doesn't work. For me.

Joshua Fruhlinger is the former Editorial Director for Engadget and current contributor to both Engadget and the Wall Street Journal. You can find him on Twitter at @fruhlinger.