English titulky

← Configuring URL Parameters in Webmaster Tools

If you're the webmaster of a site with URLs parameters and you'd like to help Google crawl your site more efficiently, Maile Ohye gives pointers (for the more common cases) on how to configure URL Parameters in Webmaster Tools.

Získať kód na vloženie
13 Languages

Ukazujem Revíziu 1 vytvorenú 08/23/2012 od Teamocrite Amaro.

  1. >>Maile Ohye: Hi. I'm Maile Ohye. I'm a member
    of Google's webmaster support team. And I'd
  2. like to help you better understand how to
    use the URL parameters feature in webmaster
  3. tools.
  4. URL parameters is a fairly advanced feature.
    So some of this information might be more
  5. complex than one would expect. Before you
    further watch this video, please check out
  6. the URL parameters page to see if you have
    a message from Google explaining that we already
  7. believe we have high crawl-coverage of your
    site. And therefore no adjustments to this
  8. feature are necessary. The message would say
    "Currently Googlebot isn't experiencing problems
  9. with coverage of your site. So you don't need
    to configure URL parameters. Incorrectly configuring
  10. parameters can result in pages from your site
    being dropped from our index. So we don't
  11. recommend you use this tool unless necessary."
  12. For those of you who have that message, you're
    good to go. And no further viewing is even
  13. necessary. But for those of you who don't
    have that message please keep watching. And
  14. one of the main takeaways is that improper
    actions on the URL parameters feature can
  15. result in pages no longer appearing in search.
    Again, it's an advanced feature.
  16. The settings in the URL parameters are used
    by Google as a helpful hint in our crawling
  17. process. For stronger directives, you want
    to use things like a robots.txt disallow.
  18. Or a meta noindex. But using the URL parameters
    hint is still very helpful.
  19. In 2010, the Google store only sold about
    160 products. And that seems fine and fairly
  20. easy to crawl. But the thing is, that of these
    160-ish products, it actually created 380,000
  21. URLs. These URLs were created by things like
    different types of navigation. So Googlebot,
  22. in terms of crawling your site, doesn't just
    look at, say, 200 unique URLs. But actually
  23. has to determine which URLs to crawl of the
    380,000 that were created. You can see how
  24. Googlebot might want to be more efficient
    in crawling by looking at these two URLs.
  25. Now, the first one says, "category equals
    YouTube." Let's say that URL to 20 unique
  26. items on the page. But on the second URL,
    it's "category equals YouTube and size equals
  27. medium." So it's the exact same items. But
    now just say, filtered down to five. Because
  28. of the "size equals medium" parameter. So
    in this way Google would rather just crawl
  29. the first URL and reach all 20 of the items.
    Rather than crawling both URLs and seeing
  30. a redundant five items.
  31. Essentially, with your input in URL parameters,
    it helps us to understand your site better.
  32. So we can crawl more efficiently. By crawling
    more efficiently, we don't crawl as many duplicates.
  33. And that will save you bandwidth. And helps
    us to focus on your unique content. Rather
  34. than crawling the duplicate of information
    repeatedly. But if you want URLs removed,
  35. you can go to URL Removals in Webmaster Tools.
    Again, the URL parameters feature is to crawl
  36. more efficiently. It's not about removals
    or explicit robots.txt disallowing.
  37. Another background piece of information that
    I'd like to mention is that page-level markup
  38. is still taken into consideration in tandem
    with URL parameters. So if you have page-level
  39. markup. Like rel, "canonical", or rel, "next"
    "prev", or rel, "hreflang", that's fine and
  40. can still be used by Google. Even if you are
    using URL parameters. Just make sure that
  41. we can still crawl your page. Meaning that
    it's not robots.txt disallowed. Or you haven't
  42. set it to not be crawled in URL parameters.
    As long as we can crawl your page, we can
  43. still use the page-level markup.
  44. Since we've covered the background information,
    let's now talk about the types of URLs that
  45. are eligible for this feature. Here's one
    from the Google store. It says, "category
  46. equals office." Other URLs would be things
    like "category equals wearables" or "category
  47. equals wearables and size equals medium."
    These URLs are eligible for the feature because
  48. they come in key value pairs or name value
    pairs. What it looks like is "key equals value"
  49. and then perhaps an ampersand. And then "key
    two equals value two." And Google, when we
  50. see these parameters will treat this URL as
    equivalent to this URL. Because the ordering
  51. in the parameters doesn't matter.
  52. URLs that are ineligible for this feature
    are those that don't use the key value configuration.
  53. So if a site uses a bunch of plus signs to
    separate their parameters. Or they just use
  54. a directory structure. Or they use their own
    type of encoding. None of these types of URLs
  55. can actually be used. Because this feature
    requires the name value pairs.
  56. Alright. I know that was a long intro. But
    now let's get started with the feature. Step
  57. one is to specify parameters that don't change
    the page's content. So you can ask yourself,
  58. "Do I have parameters that don't affect page
    content?" Things like a session ID. An affiliate
  59. ID. Or a tracking ID. These types of parameters
    don't change page content. And so in the feature,
  60. you can actually mark them as "does not change
    content." And once you've said that, Webmaster
  61. Tools will put one representative URL as the
    setting. And then Googlebot will act accordingly.
  62. Once step one is completed for all the parameters
    that don't actually change page-content, then
  63. let's move on to step two. Which comes in
    two parts. The first part is to specify the
  64. parameters that change page content. So you'll
    select. "Yes, this changes, reorders, or "narrows"
  65. page content." And then you can have a type
    of page content effect. Whether that sorts,
  66. "narrows", specifies, etcetera. And we'll
    cover more of this in depth. Then the next
  67. part is step two, 2B. which is to specify
    Googlebot's preferred behavior. So given that
  68. parameter, how would you like Googlebot to
    actually crawl those URLs. And we'll talk
  69. more about this as well.
  70. [music in background]
  71. The first parameter I'd like to cover is the
    sort parameter. And we're covering this first.
  72. But it's a fairly complicated parameter and
    setting. The sort parameter is something like
  73. "sort equals price ascending" or "rank by
    equals bestselling." Any of these types of
  74. parameters that just change the order that
    the content is presented. These are sort parameters.
  75. Once you identify a sort parameter or sort
    parameters on your site, the next part is
  76. to specify Googlebot's preferred behavior
    for when they see URLs with this parameter.
  77. Now this can get pretty tricky. So I have
    two scenarios here. Let's go through the first
  78. scenario. You could ask "Is the sort parameter
    optional throughout my entire site?" Meaning
  79. that the sort parameter is never displayed
    by default. But only with manual selection.
  80. If you can answer "Yes" to that question,
    that "Yes, it's optional throughout my entire
  81. site." Then go on to question two. "Can Googlebot
    discover everything useful when the sort parameter
  82. isn't displayed?" Meaning that we can actually
    crawl all of your items even if no sort parameter
  83. is present on the URLs. If that answer is
    "Yes" and the first answer is "Yes" then it's
  84. likely that with your sort parameter you could
    specify "crawl No URLs." Once you've applied
  85. this setting, please verify that the sample
    URLs displayed in Webmaster Tools are in fact
  86. not canonicals. So they're just duplicates.
    And that the canonical URLs, the URLs that
  87. you really want crawled and indexed, can be
    reached by regular user navigation.
  88. If the first sort parameter recommendation
    didn't apply to your site, then hopefully
  89. the second recommendation will. The second
    recommendation is, if the same sort values
  90. are used consistently site-wide. The questions
    to ask yourself are, "Are the same sort values
  91. used consistently across my entire site?"
    So a negative example of this, where the user
  92. would say "No" is if you're a Webmaster selling
    things like coins and coin albums. So for
  93. your coins, you might have "sort equals" with
    the value "year issued." But "sort equals
  94. year issued" doesn't apply to the selling
    of your coin albums. So it's not used consistently.
  95. If the answer to the first question is "Yes,"
    Then you can ask yourself the second question.
  96. Which is, "When a user changes the sort value,
    is the total number of items unchanged?" If
  97. that answer is also "Yes" then it's likely
    that with your sort parameter you can specify
  98. "only crawl URLs with value x" where X is
    one of the sorting values that's used site-wide.
  99. If neither of those recommendations apply
    to your sort parameter, then perhaps select
  100. "Let Googlebot decide."
  101. The second parameter that I'd like to cover
    is the "narrows" parameter. "narrows" filters
  102. the content on the page by showing a subset
    of the total items. So you probably see this
  103. on an e-commerce site. Where in the navigation,
    a user is able to select, if they only want
  104. to see items that are less than 25 dollars.
    Or 25 dollars to 49.99. aAll of this is narrowing
  105. the content of the total items.
  106. Examples of the "narrows" parameter are "size
    equals medium." "Less than equals 25." Or
  107. "Color equals blue." If the "narrows" parameter
    shows less useful content. Shows content that's
  108. just a subset of the content from the more
    useful URL, which doesn't include the "narrows"
  109. parameter. Then you might be able to specify
    "Crawl No URLs." For example, a useful URL
  110. is "category equals YouTube." And the less
    useful URL is "category equals YouTube and
  111. size equals medium."
  112. Here, I might specify "size" as a "narrows"
    parameter. And then because it has less useful
  113. content, I can say "crawl No URLs." But before
    I specify "Crawl No URLs" it's good to verify
  114. a few things first. First, be sure that the
    "narrows" parameter won't also filter out
  115. useful pages that you'd like crawled and surfaced
    in search results. So if you have brand or
  116. category pages that you'd like to show to
    users, be sure that when you select "Crawl
  117. No URLs" that those brand and category pages
    won't be affected. Second, verify that example
  118. URLs that might be displayed in Webmaster
    Tools are really URLs that provide non-useful
  119. content. When compared to the parent URL.
    So again, you see content like "Category equals
  120. YouTube and size equals medium." And you know
    that the "size equals medium" that narrows
  121. parameter just isn't' useful.
  122. If the behavior of "Crawl No URLs" just isn't
    optimal for your site because it affects too
  123. many important brand or category pages then
    perhaps let Googlebot decide.
  124. The next parameter is "Specifies." "Specifies"
    determines the content displayed on a page.
  125. For example, "item id equals android t-shirt"
    or "sku equals 495." The "Specifies" parameter
  126. is responsible for the actual content. So
    you'll likely select "Crawl Every URL." After
  127. "specifies" is "translates."
  128. Unless you want to exclude certain languages
    from being crawled and available in search
  129. results, like auto-generated translations.
    Unless that's the case, then it's likely you'll
  130. select "Crawl Every URL."
  131. As an aside, one best practice that I'd like
    to mention, and this is by no means a requirement,
  132. is to put your translated original content
    not in a URL parameter, but actually in a
  133. sub-folder or a sub-directory. The reason
    why is that in a sub-directory or sub-folder,
  134. it helps Google to better understand your
    site's structure. And that this applies to
  135. translated or regional content.
  136. The last parameter is "Paginates." "Paginates"
    displays one component page of a multi-page
  137. sequence. Examples are "page equals three."
    "Viewitems equals ten through 30." Or "start-index
  138. is 20." With paginates, because you want us
    to crawl every page to reach all of your items,
  139. it's nearly always "Crawl Every URL."
  140. Congratulations. You've finally gotten through
    the discussion about the different parameters
  141. and the desired Googlebot behavior. And now
    you can go back to your site and repeat this
  142. process for the different parameters that
    you have. You don't have to do this for every
  143. single parameter. But just including a few,
    and specifying the right Googlebot behavior
  144. can really help with your crawl efficiency.
  145. We're almost done. But you might be asking
    one last question. Which is, "What about multiple
  146. parameters in one URL?" For example, "sku
    equals 234 page equals three and sort by equals
  147. price" etcetera. How does URL parameters work
    when there are multiple parameter settings?
  148. The answer is to remember that URL parameters
    is about crawling your site more efficiently.
  149. So you can imagine that all URLs that we know
    about for your site. So if you're the Google
  150. store, all 380,000 URLS begin as eligible
    for crawling. And then we work as a process
  151. of elimination. Not inclusion. So we take
    our knowledge of your site. Combined by each
  152. of your settings in URL parameters to slowly
    weed away the URLs that shouldn't be crawled.
  153. Until at the very end, we have a smaller subset
    of good URLs.
  154. To recap. If you're now more comfortable with
    the URL Parameters, please utilize this feature
  155. for more efficient crawling of your site.
    You can start with specifying the parameters
  156. that do not change page content. Those should
    be easier to do. And then after you have done
  157. that step, the next step is to specify parameters
    that change the page content. And remember,
  158. if you can't determine, don't guess. Please
    "let Googlebot decide."
  159. One more time, the sorts parameter. If the
    sorts parameter for your site never exists
  160. in a URL by default, then "Crawl No URLs."
    Or if site-wide, the same sort value is always
  161. used, then "Crawl URLs with value x." For
    "Narrows" if your "narrows" parameter causes
  162. non-useful filtering for searchers, like size
    or price, then you might select "Crawl No
  163. URLs." Just be sure to double-check that none
    of your important pages will be affected.
  164. For "specifies" it's usually "Crawl Every
    URL." And the same applies to "Translates"
  165. and "Paginates."
  166. Yay! I think we've covered a lot of information
    about URL parameters. Here are links to more
  167. resources if you want further help. Thank
    you and best of luck.