seatgeek open sourced seatgeek/SGImageCache

A flexible image caching library for image rich iOS applications

Our iOS app is image rich. To create appealing views we rely heavily on performer images, all of which must first be fetched from a remote server. If each image needed to be fetched from the server again every time you opened the app, the experience wouldn’t be great, so local caching of remote images is a must.

Version 1 - Ask for an image, get it from disk

Our first image cache was simple but effective. For each image view we’d ask for an image from cache, using its remote URL as the cache key. If it was available in the local disk cache a UIImage would be created from the file on disk, and returned immediately. If it wasn’t found on disk it would be fetched async from the remote URL, cached to disk, then a new UIImage returned.

For our purposes at the time this was perfectly adequate. But it had one point of unnecessary weakness: each cache request required the image to be loaded again from disk, which comes with the performance cost of disk access and image data decoding.

Version 2 - Memory caching

Thankfully Apple’s UIImage has a built in memory cache. So by changing a single line of code our image cache could go from being a disk only cache to a disk and memory cache.

When you ask UIImage for an image via imageNamed: it first checks its own memory cache to see if the image has been loaded recently. If so, you get a new UIImage at zero cost. So instead of something like this:

1

return [UIImage imageWithContentsOfFile:[self absolutePathForURL:url]];

We could get memory caching for free, simply by doing this:

1

return [UIImage imageNamed:[self relativePathForURL:url]];

UIImage will search its memory cache and, if found, return the image at no cost. If it isn’t in the memory cache it will be loaded from disk, with the usual performance penalty.

Version 3 - Fetch queues, prefetching, and variable urgency

As the design of our app evolved we became increasingly image greedy, wanting to show richer, larger images, and more of them.

Getting these larger images on screen as quickly as possible is critical to the experience, and simply asking the cache for each image at display time wasn’t going to cut it. Larger images take longer to load over the network, and asking for too many at once will result in none of them loading until it’s too late. Careful consideration of when the image cache is checked and when images are fetched from remote was needed. We wanted precaching and fetch queues.

fastQueue and slowQueue

We settled on two queues, one serial and one parallel. Images that are required on screen urgently go into the parallel queue (fastQueue), and images that we’ll probably need later go into the serial queue (slowQueue).

In terms of a UITableView implementation, this means that a table cell appearing on screen asks for its image from fastQueue, and every off screen row’s image is prefetched by adding it to slowQueue.

We’ll need it later

Assuming we request a page of 30 new events from the server, once those results arrive we can queue up prefetching for each of their images.

1
2
3
4
5

- (void)pageLoaded:(NSArray *)newEvents {
    for (SGEvent *event in newEvents) {
        [SGImageCache slowGetImageForURL:event.imageURL thenDo:nil];
    }
}

The slowGetImageForURL: method adds the image fetch to slowQueue, allowing them to be fetched one by one, without bogging down the network.

The thenDo: completion block is empty in this case because we don’t need to do anything with the image yet. All we want is to make sure it’s in the local disk cache, ready for immediate use once its table cell scrolls onto screen.

We need it now

Cells that are appearing on screen want their images immediately. So in the table cell subclass:

1
2
3
4
5
6

- (void)setEvent:(SGEvent *)event {
    __weak SGEventCell *me = self;
    [SGImageCache getImageForURL:event.imageURL thenDo:^(UIImage *image) {
        me.imageView.image = image;
    }];
}

The getImageForURL: method adds the image fetch to fastQueue, which means it will be done in parallel, as soon as iOS allows. If the image was already in slowQueue it will be moved to fastQueue, to avoid wasteful duplicate requests.

Always async

But wait, isn’t getImageForURL: an async method? If you know the image is already in cache, don’t you want to use it immediately, on the main thread? Turns out the intuitive answer to that is wrong.

Loading images from disk is expensive, and so is image decompression. Table cells are configured and added while the user is scrolling the table, and the last thing you want to do while scrolling is risk blocking the main thread. Stutters will happen.

Using getImageForURL: takes the disk loading off the main thread, so that when the thenDo: block fires it has a UIImage instance all ready to go, without risk of scroll stutters. If the image was already in the local cache then the completion block will fire on the next run cycle, and the user won’t notice the difference. What they will notice is that scrolling didn’t stutter.

Thought we needed it but now we don’t

If the user scrolls quickly down a table, tens or hundreds of cells will appear on screen, ask for an image from fastQueue, then disappear off screen. Suddenly the parallel queue is flooding the network with requests for images that are no longer needed. When the user finally stops scrolling, the cells that settle into view will have their image requests backed up behind tens of other non urgent requests and the network will be choked. The user will be staring at a screen full of placeholders while the cache diligently fetches a backlog of images that no one is looking at.

This is where moveTaskToSlowQueueForURL: comes in.

1
2
3
4
5
6
7
8

// a table cell is going off screen
- (void)tableView:(UITableView *)table
        didEndDisplayingCell:(UITableViewCell *)cell
        forRowAtIndexPath:(NSIndexPath*)indexPath {

    // we don't need it right now, so move it to the slow queue         
    [SGImageCache moveTaskToSlowQueueForURL:[[(id)cell event] imageURL]];
}

This ensures that the only fetch tasks on fastQueue are ones that genuinely need to be fast. Anything that was urgent but now isn’t gets moved to slowQueue.

Priorities and Options

There are already quite a few iOS image cache libraries out there. Some of them are highly technical and many of them offer a range of flexible features. Ours is neither highly technical nor does it have many features. For our uses we had three basic priorities:

Priority 1: The best possible frame rate

Many libraries focus heavily on this, with some employing highly custom and complex approaches, though benchmarks don’t show conclusively that the efforts have paid off. We’ve found that getting the best frame rates is all about:

1. Moving disk access (and almost everything else) off the main thread.
2. Using UIImage’s memory cache to avoid unnecessary disk access and decompression.

Priority 2: Getting the most vital images on screen first

Most libraries consider queue management to be someone else’s concern. For our app it’s almost the most important detail.

Getting the right images on screen at the right time boils down to a simple question: “Do I need it now or later?” Images that are needed right now get loaded in parallel, and everything else is added to the serial queue. Anything that was urgent but now isn’t gets shunted from fastQueue to slowQueue. And while fastQueue is active, slowQueue is suspended.

This gives urgently required images exclusive access to the network, while also ensuring that when a non urgent image later becomes urgently needed, it’s already in the cache, ready to go.

Priority 3: An API that’s as simple as possible

Most libraries get this right. Many provide UIImageView categories for hiding away the gritty details, and most make the process of fetching an image as painless as possible. For our library we settled on three main methods, for the three things we’re regularly doing:

Get an image urgently

1
2
3
4

__weak SGEventCell *me = self;
[SGImageCache getImageForURL:event.imageURL thenDo:^(UIImage *image) {
    me.imageView.image = image;
}];

Queue a fetch for an image that we’ll need later

1

[SGImageCache slowGetImageForURL:event.imageURL thenDo:nil];

Inform the cache that an urgent image fetch is no longer urgent

1

[SGImageCache moveTaskToSlowQueueForURL:event.imageURL];

Conclusion

By focusing on prefetching, queue management, moving expensive tasks off the main thread, and relying on UIImage’s built in memory cache, we’ve managed to get great results in a simple package.

Links

• SGImageCache on GitHub
• SGImageCache on CocoaPods
• SGImageCache on CocoaDocs

seatgeek open sourced seatgeek/SGAPI

The SG Api SDK for iOS

The SeatGeek Platform provides a web service for our massive database of live events, venues, and performers. If you want to build live event information into your app or website the SeatGeek Platform is the best way to do it. Until now, if you wanted to use it in an iOS app you had to handle all of the awkward network requests and response processing yourself. With today’s release we’ve made that a whole lot easier.

Since the first release of our iOS app we’ve been gradually evolving a handful of libraries to manage communicating with our API, progressively abstracting away the messy details so we can focus on writing features. Today’s CocoaPod release is that code, in the same form as we use it ourselves. The first line in our app’s Podfile is:

pod 'SGAPI'

Fetching and Inspecting Results

The SeatGeek Platform is all about events, venues, and performers, so the same is true of the iOS SDK. Individual result items are encapsulated in SGEvent, SGVenue, and SGPerformer objects, and query result sets are fetched with SGEventSet, SGVenueSet, and SGPerformerSet objects.

Objective-C

1
2
3
4
5
6
7
8
9
10
11

// find all 'new york mets' events
SGEventSet *events = SGEventSet.eventsSet;
events.query.search = @"new york mets";

events.onPageLoaded = ^(NSOrderedSet *results) {
    for (SGEvent *event in results) {
        NSLog(@"event: %@", event.title);
    }
};

[events fetchNextPage];

Swift

1
2
3
4
5
6
7
8
9
10
11
12

// find all 'new york mets' events
let events = SGEventSet.eventsSet()
events.query.search = "new york mets"

events.onPageLoaded = { results in
    for i in 0..<results.count {
        let event = results.objectAtIndex(i) as SGEvent
        NSLog("%@", event.title())
    }
}

events.fetchNextPage()

Output

New York Mets at San Diego Padres
New York Mets at Seattle Mariners
... etc

Query Building

SGAPI uses SGQuery to build all its URLs. If you’d prefer to use your own data models or HTTP request classes and just want a tidy way to build API queries, then SGQuery is what you’re looking for.

Objective-C

1
2
3
4
5

SGQuery *query = SGQuery.eventsQuery;
[query addFilter:@"taxonomies.name" value:@"sports"];
query.search = @"new york";

NSLog(@"%@", query.URL);

Swift

1
2
3
4
5

let query = SGQuery.eventsQuery()
query.addFilter("taxonomies.name", value: "sports")
query.search = "new york"

NSLog("%@", query.URL())

Output

https://api.seatgeek.com/2/events?q=new+york&taxonomies.name=sports

Additionally, every item set (SGEventSet etc) has a query property which you can modify directly to add filters and parameters, change perPage and page values, etc.

Conclusion

See the documentation on GitHub and CocoaDocs for more details. If anything doesn’t make sense or could be improved, let us know. We’ll be evolving the SDK over time, and are looking forward to seeing how you make use of it!

Links

• SGAPI on GitHub
• SGAPI on CocoaPods
• SGAPI on CocoaDocs

seatgeek open sourced seatgeek/react-slider

A slider for React

A few months ago we launched the Omnibox, a single reconfigurable ticket-buying interface that replaced our old static listings and a thicket of popup windows. The Omnibox is written entirely in React, Facebook’s new user interface framework, and in doing this we had to come up with our own solutions to user interface elements and innovate where we could.

One of the products of our work on the Omnibox was a price slider component, which allows users to filter tickets by price:

But for an event with large price ranges - the Super Bowl, for example - a simple linear slider would be unwieldy. Tickets are likely sparsely populated across the full domain of prices and, more importantly, users are far more interested in lower-priced tickets than the exorbitantly priced ones.

We solved this problem with two features of the slider: firstly, the upper limit of the price slider was truncated to the 90th percentile of ticket prices, and only dragging the slider handle to its right end will reveal all tickets greater than that price:

Secondly, the slider’s scale is no longer assumed to be linear. The implementation currently deployed on the SeatGeek site positions the slider on the horizontal axis using the square root function, making lower prices take up more space than the less-desirable higher-priced tickets.

Today we’re happy to open source this two-handled slider implementation written in React; it has no dependencies other than React itself.

seatgeek open sourced seatgeek/dev-challenge-v1

The first version of the SeatGeek Dev Challenge. Crack open the beers.

In a land before time, SeatGeek created an extremely hackable admin panel. Its primary purpose was to stir the curiosity of developers who might be looking for a new job. You can read more about it in this previous post.

While the response over the years to the dev challenge was great, we retired the dev challenge over a year ago. Please stop trying to hack into our backend! (If you do happen to be hacking our website, we’d appreciate a heads-up on any vulnerabilities you find at hi@seatgeek.com Responsible disclosure and whatnot.

In order to cater to the curious, I took the opportunity to open source the dev challenge. It’s a small Sinatra app that you can run locally and hack to your heart’s content.

A few notes about the panel:

• Configuration is done using environment variables with a few sane defaults.
• You’ll need a local SMTP server to send email notifications of new applicants. We used postfix at the time, but you can use whatever you’d like.
• Applicant resumes are stored on disk. Yes, we know, there are probably better ways than what we did, but since the box was in a DMZ, it was probably okay. Not like we weren’t trying to have you hack us anyhow.
• Ruby 1.9.3 is what we used to deploy—actually 1.9.1 at the time, but it works with 1.9.3—but no guarantees that it will work with a newer Ruby. Pull requests welcome!

We’d like to thank all the thousands of developers who have hacked our backend over the years. Don’t worry, we’ll have a new challenge soon.

In the meantime, we’re still hiring engineers.

Last week we launched a feature that even the most die-hard SeatGeek fans probably didn’t notice. However, we think that this feature makes a huge difference in usability and overall user experience, even if only at a subconscious level. You can now interact with the section and row shapes themselves, rather than just section/row markers.

For anyone not familar with our maps, here is an example of what one looks like:

Each of those markers represents a ticket or a group of tickets in that section. Until recently, all of the map interactions revolved around those markers. In order to find out more about the tickets in a section or row, the user would have to hover or click on the marker itself.

Fitts’s Law

One major concept in human-computer interaction is Fitts’s Law. Fitts’s law models the time it takes for a user to move a pointing device (e.g. cursor) over an object. In order to decrease the time to select an object, one can do one of two things: decrease the distance between the cursor and the object, or increase the size of the object. On SeatGeek’s maps we are constrained by the layout of venues, so our only option is to increase the marker size.

The natural way to increase the target area of a marker is to expand it to the shape of its section. However, it turns out this isn’t straightforward.

How Our Map Pages Work

First, a little background on how our map pages work. We use an excellent library, Leaflet, as the foundation for our interactive maps. The maps themselves start out as vector files. These are then rasterized into a bunch of tiles, such as this:

Leaflet handles the logic for mouse interactions and displaying the correct tiles depending on the current view. The markers are a custom Leaflet layer (for performance reasons, but that is a whole other blog post). Then, we overlay a vector path as an SVG layer when a marker is either hovered or clicked.

First Attempt at Section Interaction

A while back, when we first implemented our current style of maps, we considered adding polygon interaction instead of just using the markers. Given that we had the SVG paths of all of the shapes for the highlighting purposes, we decided to add all of these SVG elements to the map so that we could use the built-in event handling that browsers provide.

Unfortunately, that resulted in terrible performance on the map page. At the row level, we can have as many as a few thousand SVG elements being drawn at the same time. Combine that with all the markers we have to draw, and the map grinds to a halt. We decided to shelve the section interaction and move on to other features.

A Renewed Attempt

With the launch of our new map design, called the Omnibox, click and hover interactions became much more central to the interface.

The breakthrough was realizing that we could implement our own logic for hit-testing, or checking if a given mouse position is inside of a polygon. This means we didn’t have to add any additional elements to the DOM (like we did before with the SVG elements).

The naive approach would be to iterate through every polygon and check if the mouse is inside it using the ray casting algorithm.

However, we can do even better. By using a spatial data structure, such as an R-Tree, we can reduce the lookup complexity from linear to logarithmic. Below is an example of a 2D R-Tree. Imagine that the lowest level of the tree contains references to the actual polygons. Each node in the tree represents a bounding box that is the union of all the children below it in the tree.

Luckily enough, we were able to find a leaflet-compatible implementation of an R-Tree by Calvin Metcalf. Basically our use of it looks like this:

1. On page load, convert all of the SVG shapes to geoJSON and insert into the R-Tree.
2. On mouse events, transform the mouse position to the map coordinate system and make a throttled call to our intersection testing function.
3. Call a search on the R-Tree with the transformed position of the mouse.
4. The R-Tree will return a list of all the leaves (shapes) whose bounding boxes had been intersected.
5. Iterate through the shapes and perform the ray casting algorithm to check for intersection.
6. Return the intersected shape.

Results

The user can now hover over any point in the polygon, instead of just the marker! It works on sections:

And it works on rows:

Most importantly, all of this works without any impact on performance.

If you think about SeatGeek like we do, then in your head you probably picture an event page. You know, the page that has the venue map, the colorful dots, and the big list of tickets from all over the web, ranked by Deal Score. All the best reasons to know & love SeatGeek are encapsulated in this one single page. And it is not only the functional core of SeatGeek, it’s also our most highly-trafficked page type.

With so much riding on the event page, we’re constantly working on incremental and under-the-hood improvements. We normally avoid committing obvious, disruptive changes, but a few times in SeatGeek’s history we’ve launched major redesigns of our event page—the most recent of which happened earlier today.

Here I’ll give an overview of the latest changes and, for posterity, a quick tour through earlier SeatGeek event page history.

Today’s release

Inspiration

In the year and a half since we launched the last major version of the event page we started making mobile apps. Designing for mobile devices forced us to reconsider the SeatGeek experience from scratch, and once we launched our apps—in particular our iPad app—they became new sources of inspiration for the website. For example, we began to think much harder about conservation of screen real estate.

Internally, today’s milestone inherited the name “Omnibox” from an eponymous Google feature. Not Chrome’s address bar, but rather a more obscure reference to a CSS class found in the new Google Maps’ control panel. Although many people have griped about Google Maps’ recent update, we admired the idea of having a single page element whose content could change based on interactions with the underlying map.

What changed

In the main sidebar, we swapped our large, color-filled section headers and deal score labels for more elegant circles and lines that more closely resemble our latest iOS designs. We also moved the filter controls and box office link from the top of the sidebar to the bottom. The result is that ticket listings get higher precendence on the page.

The new version of the section info view (below, on the right) looks very similar to the old, with the notable exception that it doesn’t appear in a popover overlaid on the map, but rather in the sidebar. Popovers had a lot of annoying characteristics, not least of which was that they were starting to feel decidedly web-1.0-y. As an added bonus, under the new sidebar scheme, it’s now possible to apply search filters to tickets within individual sections.

If you can believe it, the old version of the ticket info view (below, on the left) was actually a second popover that appeared beside the first popover containing section info. Now that all this information is in the sidebar, the map won’t get cluttered (which was especially problematic on smaller viewports), and the ticket details are much more legible.

Last but not least, we moved the old event info bar (seen in the top half of the image below) into the site header. This frees up more space for the venue map. In order to make room for event info in the new site header, we consolidated the category page links (i.e. “MLB”, “NFL”, etc.) into a dropdown off the main SeatGeek logo.

To really get a feel for the wonders of the new event page, you should really check it out. Go try a search, or check out one of these random event pages.

Earlier event pages

Here we take a walk down memory lane, through the annals of SeatGeek event page history. We’ll begin with the very first 2009-era event page before venue maps even existed and end on today’s latest event page release, ten notable iterations later.

Full disclosure: there’s a gratuitous, self-indulgent number of screenshots ahead. Only the most obsessive SeatGeek fans need read any further.

#1 The original SeatGeek event page was launched—along with SeatGeek itself—in September 2009. It contained no venue maps. SeatGeek was all about price forecasting, and making recommendations about whether to buy a ticket now, or wait to buy. If you wanted to buy, there was a table of tickets from various sellers, sorted by price.

#2 In early 2010, SeatGeek licensed the rights to use venue maps, provided by a third party named SeatQuest (now defunct). According to engineer #1, working with SeatQuest maps was reportedly a nightmare.

#3 Before long, ticket listings and venue maps started stealing screen real estate away from the price forecast part of the page.

#4 The event page’s first major redesign happened in Summer 2010.

#5 Soon after the Summer 2010 redesign, we scrapped SeatQuest in favor of our own venue maps, which should look a lot more familiar to current SeatGeek users. Also worth pointing out that by now the price forecast feature is relegated to a small-ish button area above the map, and restricted to signed-in users.

#6 Sometime in early 2011, we made the long-standing switch from a lefty sidebar to a righty—a change that would persist all the way until yesterday.

#7 In mid/late 2011, we redesigned the site again. Note the dark blue primary colors, and the new styling for the sidebar.

#8 In the first half of 2012, the dark blue from the previous version softened into the lighter SeatGeek blue of today.

#9 This update featured some new sidebar styling and abolished the permanently overlaid search bar in favor of a more compact event info bar. This version reigned supreme from Fall 2012 all the way until March 12, 2014.

#10 Omnibox: The cleanest SG event page yet. (Note the lefty sidebar, a clear throwback to the year 2010.)

Back in December, we released an epic update to our iOS app that featured tons of big additions and improvements for iPhones. But this time, it’s all about the iPad.*

This is much more than just a quick catch-up release. We took stock of all the additions from v2.0 for iPhone, and thought holistically about what the ideal experience for each one would be like on a tablet. So, iPad users, sorry about the wait. But we think you’ll find that it was well worth it.

* Well, technically it’s not all about the iPad. We also made a few UI tweaks to the iPhone app, and added a “Recent Search” list to both apps so you don’t have to keep typing the same things over and over again. Anyway, the rest of this blog post is focused on big, new iPad features.

Demo time

Like gifs? Great! Here’s one:

If that was too fast for you, we’ve also got a few higher-res screenshots that you can inspect more closely down below. But to really get the best feel for the new app, we’d strongly suggest installing it now →

Logging In

Now that you can log in (you couldn’t before), we can remember things for you – like artists, teams, and events that you’re tracking. Later, when you’re logged in to SeatGeek somewhere else like on your laptop or your iPhone, you’ll have access to all the same preferences.

You can log in with a SeatGeek account or a Facebook account. (One nice thing about Facebook accounts is that we can help you automatically track performers for you if you’ve already liked them on FB.)

Tracking

As noted above, you can now track artists, teams, and events right in the app. Tracked items are easily accessible in your new My Events and My Performers screens. We’ll even send you reminders when new shows are announced or when a tracked event is approaching. Plus, now that we know a bit more about what you like, our event recommendations for you will get better and better.

Here’s what the My Performers screen looks like:

To track an event, just tap the heart in the corner of any event screen:

What! JT’s coming to town?!

Redesigned Home Screen

We thought to ourselves why bother making an update if you can’t tell the difference right away? So we also added some new hotness to the home screen:

Tapping the Explore area near the bottom lets you browse all upcoming events near you:

Go get it

Now that you know what’s in store for you, all that’s left is to hit the App Store to make sure you’re running the latest version.

As always, we hope you like it and will let us know if you find bugs or have any cool ideas.

seatgeek open sourced seatgeek/businesstime

A simple python utility for calculating business time aware timedeltas between two datetimes

Earlier this week we released businesstime, a simple python library for doing math with datetimes that’s aware of weekends, holidays and business hours. We’re using it to more accurately calculate shipping costs for tickets listed on SeatGeek.

Here’s a quick demo of how it works:

>>> datetime(2013, 12, 26, 5) - datetime(2013, 12, 23, 12)
datetime.timedelta(2, 61200)
>>> bt = businesstime.BusinessTime(holidays=businesstime.USFederalHolidays())
>>> bt.businesstimedelta(datetime(2013, 12, 23, 12), datetime(2013, 12, 26, 5))
datetime.timedelta(1, 18000)

You can grab it on pypi or checkout the code on github. We hope you find it useful!

Earlier this week we shipped version 1.2 of our now two-month-old Android app. There were a few main focuses for this release:

Tablet support. Whereas the previous version was built for smartphones only, the new release extends support to all the popular 7- and 10-inch Android tablets. (This one’s for you, Nexus 7 and Galaxy Tab owners.)

Landscape support. When the app was smartphone-only we could get away with only building for portrait, but tablet users sure do love to hold their devices sideways. So you’ll find that 1.2 supports landscape mode – for both smartphones and tablets.

Performance & stability. Since version 1.0 launched, we’ve kept a close eye on app reviews and error logs, tweaked the internals, squashed some bugs, and cured as many crashes as we could diagnose. The result is that the latest version should run fast and smooth.

We’ve also added support for deep-linking. If you’re a developer or an API partner looking to refer app users to SeatGeek, you’ll find the relevant docs at platform.seatgeek.com.

And, last but not least, we’ve stepped up our game with a new, more Android-esque app icon:

If you haven’t already, hit the Play Store now to get yourself a fresh copy of the app. And be sure to let us know what you think.

Today we’re excited to announce the public launch of Cross-Event Search, a new feature that helps users find the best tickets across multiple events at once.

Suppose you live in NYC and want to see what all the Book of Mormon hype is about. You want to go to whatever upcoming show is cheapest. With Cross-Event Search you can see the listings for all future performances (over 400 events!) in a single UI. Here’s the live page for that search.

The ability to search for & compare tickets for more than one event at once is not only a SeatGeek first, but also – to our knowledge – a ticketing first.

When to use it

Cross-Event Search comes in handy when you have more than one chance to see an event and – rather than pick solely by date – you prefer to choose based on which event has the lowest price, the closest seats, or the best deal.

The basic formula is: one performer + one venue + multiple events.

In addition to the Book of Mormon example above, here are a few more common use cases:

• Home games: The Giants’ season is underway and you want to take advantage of some nice San Francisco weather and a 9-game home stand with a group of friends. You don’t care who they’re playing — you just want to get into the park and have plenty of money left over for beer & hotdogs. This page compares all Giants’ home games in the first two weeks in June.
• Back-to-back concerts: Jay-Z has graced your local arena with multiple nights of concerts. Tickets originally sold out in minutes, but you’re sure there must be at least one show on SeatGeek with a great deal.

How to use it

To start Cross-Event Searching, visit any performer page on SeatGeek and click the Compare Events button. (This will work for any major sports team, music artist, theater show, etc.)

Check the boxes next to the events that you want to include in your ticket search, and click the green Compare N Events button. (You’ll also be able to change your selections on the next page.)

Next you’ll be taken to a regular-looking SeatGeek event page, where the venue map & sidebar will be populated with ticket listings from all your selected events.

Deal Scores for each event are indexed to the same scale, so you can browse for tickets just as you would normally. In other words, an 89 on one night is still a better deal than a 79 on another night. (See also: Absolute Deal Score.)

To adjust which events are included in your search, click the Comparing tickets for… button at the top of your event page and check or uncheck boxes next to a few more events.

Finally live

As many patient SG feature-requesters can attest, Cross-Event Search has been a long time coming. Historically, our underlying software was not structured to support queries like this. But that changed in June 2013 with a first place-winning SeatGeek hackathon project named “Sandcrab,” which paved the way for Cross-Event Search.

We’re pumped to release this feature into the wild. As with everything we build, please, please hit us up with feedback.