Curated by: Luigi Canali De Rossi

Friday, April 24, 2009

How To Get Your Images Indexed By Google Image Search

Sponsored Links

One of my greatest traffic sources is Google Image Search, and, if you like me, utilize plenty of high quality, well-selected images for your web site, it is likely that the same thing has already happened to you as well.

Photo credit: Mario Lopes

In case you are not yet benefiting from this extra powerful source of web traffic here are a few tips and a video from Google itself, that should help you understand better how Google Image Search works, how it indexes images, and how to get your published images to become a valuable source of web traffic for you as well.


Tips on How to Get Your Images Indexed


by Robin Good

Here some basic tips you can use right away to improve your ability to get your images indexed by Google and other major search engines.

  1. Since search engines can track pages / content through text, add a text description to each image by learning to use alt tags.
  2. Select images that have a very strong visual impact and which are "essential" by testing them against adverse viewing conditions. Learn more about How To Select Best Images For Web Publication.
  3. Rename the name of image files so that they clearly reflects in clear and simple words, separated by dashes, the actual description of the image contents.
  4. Surround the image with text that is relevant and complementary (topic-wise) to the image subject.
  5. The larger the dimension of the image are, the greater the chances of getting more Google Image Search originated traffic.


Google Image Search

Duration: 14' 45''

by Peter Linsley - Google

Full English Text Transcription


Hi everybody, my name is Peter Linsley, I'm a product manager at Google, working on Image Search.

What we thought we'd do today was to run over some slides that I presented at SMX West in February 2009.

The slides were sort of high-level introduction to Image Search.

First of all I thought we'd run over to the presentation I gave at SMX West, and then afterwards I'll run through some of the questions that came up which seem like topics of interest for webmasters.

Let's start the presentation.




First of all, our mission with Google Image Search is to organize the world's images.
We put a lot of focus on satisfying the end user, so when they come with a query and they have an image that they are looking for, our goal is to provide relevant, and useful images for that query.

Of course, the theory is that if they find what they're looking for, and they enjoy their experience, they'll come back and use us again.

What I wanted to get out of this talk as well was to start to engage a little bit more with the webmaster community.

If we look at what has come out of conferences like this where web search representatives from different companies, have gone out and had a conversation with the webmasters and found out what their pain points were, and we found a sort of ad hoc consortium came together and came up with things like the sitemap standard, or they came up with rel="nofollow", and they came up with robots wildcards, and things along those lines.

One of our hopes in Image Search is that we can trying start this dialogue and find out what sort of pain points you guys might have as webmasters, where we, the likes of Google and also other search engine companies, can trying come together and try to enhance the end user experience by finding an easier way for you guys to get your images both indexed and ranked.




I'm just going to move on to the first slide that I had:

I wanted to paint a little bit of a picture of image searchers. What they do and why they might be slightly different to the kind of audience you might be used to with web search.

Image Search appear in lots of places beyond images at You've probably seen images appear in universal search, so whenever you do a query like "pictures of San Francisco" there might well be a portion of the results page that's dedicated to show images for that results.

And the theory here is very much in line with our goal at Image Search which is that we are going to show you these results when we believe these are useful, and informative, and relevant to the query.

Images also appear in other places like on Maps. You might have seen a little row of images in Maps which come from property, and it is a really cool product if you haven't seen it before.

Images appear everywhere all over across our properties, and we're really just trying to align it with when they match the user intent and enhance the user experience.


Search Behavior


Image searchers also have a very unique search behavior. They are a very different animals to web searchers.

If you think about the paradigm when they do a query, it's not so much about what's the first result. We don't really have this sort of "I'm feeling lucky" paradigm. It's more about saying: "Here's a query. Well, here's 20 images that you might like". And users can consume those images in a heartbeat, and if the image they happen to like is at the bottom left-and corner or the bottom right-hand corner, so be it. They'll see that, there's something about the image that attracts them, and they'll click through.

The other thing they do is they search a lot of images, so there's a lot of next paging going on, they'll go very deep looking for the images they like. One of the reasons why this happens is that a lot of queries we see are very subjective in nature, so if you see a query like "waterfalls", then the waterfall that you like and the waterfall that I like might be on two very different pages.

There's no way, as a search engine, we can figure out what you are looking for.

So, there's a lot of next paging, users can consume results very quickly, and it's just interesting I think what it might mean to you guys, as marketers, that is not all about being in the first position in the first page.

There's also a lot of novel use cases on Image Search which we might not be apparent. Users use image search for inspiration. they want to get a haircut, or a tattoo, and they are looking for ideas. So, "tattoo ideas" and then they go through the pages looking for some inspiration. They'll refine that query... there's a lot sort of exploring and browsing with intent.

Users also use Image Search for shopping. They use it for research, health queries, or sometimes they just use it to kill time, just for the fun of it.

Another really interesting use case that we've seen is using Image Search as a visual dictionary.

There's a googler in Germany who's learning German, and if he hears a noun or a word that he's not too sure of what it is, he'll type it in and he knows exactly what the word means, even though he's not looking it up in a text dictionary.


How Image Search Works


This is a slide on how does Image Search work:

Simply put, as a webmaster, you'll see Googlebot come along and download the HTML as normal. Then what happens is, we pass through your page and we look for references to images.

Typically, references to images can come in one of the two forms:

  • It's either an "a href", when you're linking to an image directly,
  • or it's in-lined image with an "img src" tag.

Then what happens is, we come along and crawl the images, and then we go through this process of classifying it. What we are trying to do here is to figure out how to bucketize these images correctly.

One classification we do, is to work out if "It's a photograph or not?" Another one might be: "Does it contain a face?" Other buckets might be things like: "Is it line-art?", "Is it black and white?", "Is it an unsafe image?" that we can only show when SafeSearch is disabled.

This sort of classification goes on, and the reason we do that, is we found that image searchers really like to slice and dice their results. They like to do a query and then look at it and say: "Well, these images are sort of nice, but I really wanted to see just images with faces in them".

If you've seen across the top of the results page, there's a blue bar which contains some drop downs where you can filter the results down to just photographs, or just faces, or just line art, and so on and so forth. These filters tend to get used quite heavily.

We like to try and bucketize things off so they show in a more relevant context.

Finally, of course, the images are indexed and that's where we scroll them away, we have an index of the image, with all text associated with that particular image.


Identifying the Duplicates


Another part of this process is about identifying the duplicates.

If you think about the way images are typically deployed online, you might put an image up and a particular page will refer to it, and another page might refer to it, and you might have other pages on your site that refer to it. Every now and again the image will get copies, and maybe it gets copied as is, or maybe gets transformed slightly. But as far the user is concerned, it is still very much the same image.

The next process we go through is one of trying to cluster all of the very similar or identical images and trying treat them as one. And this is very much the same as the way things are done in the Web, where web pages are analyzed for duplicates, and then one sort of canonical winner is picked out of that entire group. The same thing happens with Image Search.

We try hard to identify all the duplicates, and again the main reason for us doing this is that when somebody comes in and types "blue widgets" we really don't want be showing them exactly the same blue widget 20 times, we want try and cluster them all together and say: "Here is one interpretation, and here is another one".

There are multiple images. Our goal is to try and cluster these and figure out which is the best one, and at the same time we have multiple pages that are including that image.

Another task is the runtime and query time to try to figure out which one of these referrers makes the most sense for these particular images that we've chosen. And the answer for this is we're trying to choose the best one.

We're trying to choose the best image that meets the user's intent more accurately, and maybe it's about size, or maybe it's about quality, or something like that. The referring pages that are included in that images are selected based on "how good it is" essentially, and that could be one of many things such as its relevance to the actual query itself.

And finally, ranking is performed on a whole lot of signals and typically we don't get into details of the signals, but it's very much like Web, there's more than one signal that we use to try and figure out what the most relevant image would be.


Best Practices


The next slide is on best practices, so you say and you think: "This sounds great. I've got good images that I think will be useful for users as well... What can I do about it?"

Probably the best bit of advice we can give is to really focus on the user. You might be thinking: "What exactly does it mean? What can I go and do tomorrow to focus on the user?"

The answer is pretty simple.

If you think of a user who comes to Google Image Search, and what they might be looking for, it might take one use case, such as "coloring pages". Maybe they're looking for a site which has a lot of coloring pages and they trust to use Image Search to get there.

The first thing am I going to do is come along and type in "coloring pages" and they're going to look at the results and maybe they see something that they like, maybe they don't, they might hit next page a few times, and all of sudden one element will catch their eye. They like it for some reason, and maybe it's just the quality of image itself. Or maybe it's the snippet, maybe there's something around the size, or the host name, and this draws their attention. Maybe, it's, "I know that site, I'm going to click through. I trust it".

Then they land upon your page. And the question is: "What sort of experience are you immersing them into?" "What sort of experience are they getting now they've come to your page" given that they were looking for coloring pages?

"Do they see the current page they just clicked on above the fold?" "Is it large enough?" It's one thing to send people to "coloring pages" page where they see very small thumbnails, and another thing is to show what they just clicked on and say:

"Here it is. Here is some descriptive text, here are some related pictures. Here is some commons from users, the readings, and all sort of things."

It's really about immersing the user into a very image-centric experience. These are the kinds of landing pages, these are the kind of images we've observed that our users tend to like.

Again, our intent is to try and match up the intent with the end user.


Image Search Tips


  • Focus on the user

  • High-quality images are always good, if you're taking photographs to put in your site, go and buy a digital SLR, learn how to use it, get a good lens... take really nice high-quality pictures.

    You don't necessarily have to take up the whole screen with the photograph or the image, but just large enough is usually what the users like.

  • Above the fold,

  • and plenty of descriptive text, and fundamentally the impetus to all of image search is a search query and the extent to which you have a lot of descriptive text that's on topic and talks about what's in the image.

    Maybe you want to expose EXIF data, maybe where the image was taken, maybe has a nice title across the top.

All those sort of things are really good clues for us to figure out when an image is relevant and not, but more importantly it's useful for the end user. They can read the description, read the caption, and learn a lot more about the image.




The last slide... I talk about resources:

  • We have the Webmaster Help Centers, where you can go and read a lot more about Image Search, and we also have forums, where you can post questions about Image Search.

    We really encourage webmasters to come to these forums, and post all of their questions or concerns. We monitor this very closely, and we pick up these concerns, and we will take a look at them.

  • There's also a web search help and forum for end users, so if you are an end user of Image Search and you have questions, you can leave them there.

  • The other thing is to monitor the Google Official Blog because that's where we typically put our announcements of new features, and changes, and news... specifically around Image Search.


Pages Load Times and Analytics Data


That was the end of my presentation which I gave at SMX. In a nutshell.

At the end of the presentation we had Q&A and I wanted to pick up with some of the questions that came up during that time:

Q: The first question was: "Hey, you guys mentioned large images are a good best practice, but I have a concern with that because I don't want load really the large version of the image that I have because it takes the page a whole lot time to load up. So how do I balance.... how do I manage the trade-off there?

A: I think the answer to this question is just to show an image that's large enough. Typically 2/3 of the screen, maybe one sort of ruler thumb, but the point here being that users tend to like to be able to see the image, as opposed of being a very small thumbnail.

A good way to get around this, to allow the users to see the larger version if they wanted to, is to turn your image into a link to either the larger image itself or another HTML page that includes the larger version of the image.

Ultimately no-one wants to see an image that's larger than the browser size.

Q: Another question that came up was about Analytics, and somebody was saying: "Hey, can I get Analytics information, traffic that's coming from Image Search?

A: The answer is: "Absolutely". In the referral string that we send across, that the browser sends across, both the query that ranks the image, plus the image itself, is sent in that stream.

One slight difference with Image Search of course is we are not necessarily sending people to your pages as much as we're sending people to your image on the page, and there could of course be more than one image.

Bypassing upon the referral string you should be able to get all the Analytics you need to know to what queries sent the user there, and what image sent the user there.

Google video originally recorded by Peter Linsley for the Google Webmaster Central YouTube Channel on March 2nd, 2009 as "Google Image Search".

Photo credits:
Mission - adistock
Goals - Tatiana53
Search Behavior - Lars Christensen
Identifying the Duplicates - Elena Elisseeva
Best Practices - Sergey Galushko
Image Search Tips - Christophe Testi
Resources - Ramon Grosso Dolarea
Pages Load Times and Analytics Data - Michael Osterrieder
Tips on How to Get Your Images Indexed - Vasyl Yakobchuk

Robin Good and Peter Linsley -
Reference: Google Webmaster Central YouTube Channel [ Read more ]
Readers' Comments    
blog comments powered by Disqus
posted by Daniele Bazzano on Friday, April 24 2009, updated on Tuesday, May 5 2015

Search this site for more with 








    Curated by

    New media explorer
    Communication designer


    POP Newsletter

    Robin Good's Newsletter for Professional Online Publishers  



    Real Time Web Analytics