Near Me Pages v2

Yelp has several types of pages built to improve its SEO game, but the "Near Me Pages" receive the most traffic today and number to over 2,000. If you were to Google, "Fast food near me", this page type would be one of the top results. Traffic to these pages has been stagnant, so I was asked to look into how we might improve them. Even slight improvements could boost traffic to Yelp considerably –– to the tune of ~100,000 sessions a month. That could be many people's first experience of Yelp. Fortunately for designers and our users, Google rewards better experiences.

Since I was new to the Growth team and to the very existence of these pages, I started by documenting visual/UI issues and picking the Traffic PM's brain. The original design was optimized to include a ton of content (the more, the merrier, or so I was told) but not optimized to convert (my opinion). Analyzing our Google Analytics and Splunk data has a decent learning curve, but the engineers on the team were more than willing to help. The data appeared to be validating assumptions I had already made – mobile users weren't finding what they were looking for. This discovery process helped me define my principles for the rest of the project.

Principles for the next iteration:

• Design mobile-first
  Mobile accounts for 78% of traffic
• Give users a clear ability to change their location
  IP-based geolocation isn't always accurate
• Guide users to a quick decision
  The UI is not optimized for scanning
• Make search more prominent
  It's our core functionality and it's being downplayed
• Show hours and/or elevate open businesses
  ‍Most people want to go somewhere near them sooner, not later
• More focused templates will emphasize key information
  These pages are well-categorized, let's use that to our advantage

Even though these pages are responsive, because I designed them from a mobile-first perspective, I'll focus on my proposed changes for each section as they would appear on mobile.

• Exposes our core functionality (search)
• Exposes location so users can update if needed
• Aligns more closely with our home page design on desktop and mobile
• Explains what Yelp can do for users who may not be familiar

• Consolidates redundant information
• Left-aligns content for scannability/legibility
• Removes confusing and unnecessary intra-page navigation

• Carousel content responds to screen size
• Exposes hours for users who want to go somewhere near them soon (research has proven this is more often than not)
• Use neighborhood instead of address when available to reduce cognitive load
• Expose categories to instill confidence in results and to help the user refine their search if needed
• Remove review snippets to save valuable real estate on mobile devices

Proposing this call to action is a compromise. In an ideal world, we would know precisely where the user is and be able to show where they are in relation to the businesses near them. But we have to ask for location permission in order to do that accurately, and asking for that up-front seems too heavy-handed. Let's have them explicitly state they'd like to search on a map, and ask for permission to use their location then.

• Since we now show hours info, show a different set of results based on our review content

• Keeps the SEO benefit of including review snippets and better attributes the snippet to its respective user
• Reuses an already-built component

• Leverages our (really good) search data to offer related categories to help users refine intent if they haven’t found what they’re looking for

• (Food-specific) Adds price point info for all food and restaurant businesses

• (Food-specific) Removes these sections.
• Users rarely make reservations in this use case, plus the section is irrelevant on one of our most frequented Near Me pages – Fast Food Near Me
• Popular Cuisines takes up excessive real estate and isn't as dynamic and therefore relevant as the Related Categories section

• (Services-specific) Show response-time info instead of hours. We're making a huge push on our Request-a-Quote feature

A couple of the proposed changes just ran as experiments over the past month – bounce rate is down and the changes will be shipping to 100%. The bigger bets haven't started being built yet. Unfortunately I did not have the time to explore my last principle to its full extent (Emphasize key information with more focused templates), nor run task scenario testing. For a third version (or redesign), those things would be crucial. Overall, this project was a relatively small investment, but one I believe Yelp will profit from when fully fleshed out. Time will tell! Onward!

Other Projects