Ask HN: I want to build an alternative to Yelp.com | Hacker News Reader

Ask HN: I want to build an alternative to Yelp.com

8 pointspalidanx9 comments

Recently, I feel that Yelp is good as a directory listing, but getting worse for reviews.

Some of the problems: + The quality Yelp Elites (full disclaimer I am one) are getting worse and worse. To let you guys know, Yelp Elites get priority sorting whenever reviews are displayed.

I've met several elites who review things who aren't that knowledgeable or who even haven't eaten at a given restaurant.

+ Review average is skewed

I feel that the reviews should be based on a rolling 6 month average to give businesses a chance. Those who start off with bad reviews, but eventually get better should not get penalized with their initial launch.

+ Yelp's 'premium service' are a bit dodgy

I know that this area is a bit vague, but I have several restaurant friends who have been called by Yelp to get a 'premium service', so they can choose what reviews to highlight. But what people say is what Yelp can do is help 'raise' your review average through a premium service.

What I am thinking I would like to launch is the follows

+ Provide a structured review format

I would provide a review template that hopefully businesses would find more useful. For example, 3 text boxes with: "the good", "the bad", "and what can be improved." The goal would to actually provide input some your local business can get better.

+ Change average display system

I would introduce the ability for users to view the averages on things like a rolling 6 month window (as indicated before), or other metrics to show how the businesses is doing through time

+ Business Analytics

For businesses, I would rather work with them so they can extract their daily reputation easier so they can track what is going on. They would get a weekly report on things they can improve on, and their ratings.

Ideas/thoughts/comments?

Yelp's biggest problem is that it sucks for the businesses but not for the consumers who use the site. A review site which caters primarily to businesses as yours would has no chance of competing with Yelp because it offers nothing to the consumer that Yelp doesn't already provide. Without those consumers, you offer businesses very little worth paying for except yet another review site.

You need to come up with some compelling ideas for the consumer-user experience before you think about the business-user side. Urbanspoon went with UI, and their mobile app was/is awesome (but their coverage of restaurants is spotty outside of major metropolitan areas).

[By the way codegeek, your comment got deaded because of the profanity filter, so no one can see it without showdead.]

Good points!

Social review sites require a critical mass of users to be successful. Nobody will want to write reviews for a site that doesn't have a significant user base, and conversely no-one will visit a site that doesn't have many reviews. Be prepared to spend a lot on user acquisition. It's going to be cost-prohibitive.

Also, think about the following: What prevents Yelp from implementing some or all of the changes you outlined? Do you think your tweaks would sufficiently differentiate your service from Yelp?

Yelp is still losing money. The Premium Service you mentioned sounds like another attempt at monetizing the site. What would you do to be more profitable than Yelp?

In my opinion, I think Yelp can't really implement some of the algorithm changes because it would radically affect all of the ratings across the board. Also for changing the review structure (one text box), that probably most likely wouldn't change.

I think the tweaks would differentiate the service from Yelp, but you do have a super valid point about the cost of user base acquisition.

As for monetizing the site, I'm hoping (again laying out a business assumption that may be faulty), that businesses would want to subscribe to a service for a low cost say like $5/month to get customer feedback to help improve their bottom line. It does kind of assume businesses would listen to the customer though.

One thing that is annoying with all these review sites is that there is no way to verify if the reviewer has actually used the service she is reviewing. Amazon differentiates between the reviews from people who have actually bought the product, from those who may not have. If you could figure out a way to do it on your site, it would be cool (may be put a unique code in the bill, the customer has to enter it to add her review). Even though the quantity of reviews would be much lesser this way, the quality would be very high.

That's a really intriguing concept. I wonder if there would be some way of doing metrics on Yelp to see how many are obviously faking the review.

Start with something drastically different, much simpler and even slightly controversial. "Yet another yelp" or "yet another social sharing site" is not going to cut it.

Instead of food, let people rate quality of restaurant toilets. There is a wisdom that you can judge quality of establishment by the condition of their restrooms. I found it to be true in vast majority of cases.

Then - migrate this review model to hotels and other businesses.

Be the first, attract attention and be useful and unique.

Gleb

Most my local Asian eateries have probably the scariest restrooms ever...

Codegeek's comment (which can be seen with showdead), brings up some really interesting points. Maybe when you join the site, you can enter your value system such as liking ambiance, service, or food. A good example for myself is I really don't care about ambiance, but I really care about the quality of the food.

Reply on news.ycombinator.com