Increase in pages crawled per day
-
What does it mean when GWT abruptly jump from 15k to 30k pages crawled per day?
I am used to see spikes, like 10k average and a couple of time per month 50k pages crawled.
But in this case 10 days ago moved from 15k to 30k per day and it's staying there. I know it's a good sign, the crawler is crawling more pages per day, so it's picking up changes more often, but I have no idea of why is doing it, what good signals usually drive google crawler to choose to increase the number of pages crawled per day?
Anyone knows?
-
Nice find Ryan.
-
Agreed. Especially since Google's own Gary Illyes respond to the following with:
How long is the delay between making it mobile friendly and it being reflected in the search results?
Illyes says “As soon as we discover it is mobile friendly, on a URL by URL basis, it will be updated.
Sounds like when you went responsive they double checked each URL to confirm. From: http://www.thesempost.com/googles-gary-illyes-qa-upcoming-mobile-ranking-signal-change/. Cheers!
-
I usually analyze backlinks with both gwt and ahrefs, and ahrefs also doesn't show any abnormally high DA backlink either.
Agree the responsive change is the most probable candidate, I have a couple of other websites I want to turn responsive before April 21st, that's an opportunity to test and see if that is the reason.
-
Ah, the responsive change could be a big part of it. You're probably getting crawls from the mobile crawler. GWT wouldn't be the best source for the recency on backlinks. I'd actually look for spikes via referrers in Analytics. GWT isn't always that responsive when reporting links. Still, it looks like the responsive redesign is a likely candidate for this, especially with Google's looming April 21st deadline.
-
Tw things I forgot to mention are:
- something like 2 weeks ago we turned the website responsive, could it be google mobile crawler is increasing the number of crawled pages, I have to analyze the logs to see if the requests are coming from google mobile crawler
- the total number of indexed pages didn't change, which make me wonder if a rise in the number of crawled pages per day is all that relevant
-
Hi Ryan,
- GWT (Search Traffic->Search Queries) shows a drop of 6% in impressions for brand based searches (google trends shows a similar pattern).
- GWT is not showing any recent backlink with an abnormally high DA.
- we actually had a couple of unusually high traffic from Facebook thanks to a couple of particularly successful post, but we are talking about a couple of spikes of just 5k visits and they both started after the rise of pages crawled per day.
If you have any other idea it's more than welcome, I wish I could understand the source of that change to be able to replicate it on other websites.
-
I am not sure I understand what you mean, that website has a total of 35k pages submitted through sitemap to GWT, of which only 8k are indexed. The total number of pages indexed have always been slowly increasing through time, it moved from 6k to 8k in the last couple of months, slowly with no spikes.
That's not the total number of pages served by the site, since dynamics search results page amount to around 150k total pages, we do not submit all of them in the sitemap on purpose, and GWT shows 70k pages as the total number of indexed pages.
I analyzed Google crawler activity through server logs in the past, it does pick a set of (apparently) random pages every night and does crawl them. I actually never analyzed what percentage of those pages are in the sitemap or not.
Internal link structure was built on purpose to try to favor ranking of pages we considered more important.
The point is we didn't change anything in the website structure recently. User generated content have been lowering duplicate pages count, slowly, through time, without any recent spike. We have a PR campaign which is increasing backlinks with an average rate of around 3 links per week, and we didn't have any high DA backlinks appearing in the last few weeks.
So I am wondering what made google crawler start crawling much more pages per day.
-
yes, I updated to parameters just before you posted
-
When you say URL variables do you mean query string variables like ?key=value
That is really good advice. You can check in your GWT. If you let google crawl and it runs in to a loop it will not index that section of your site. It would be costly for them.
-
I would also check you have not got a spike of URL parameters becoming available. I recently had a similar issue and although I had these set up in GWT the crawler was actively wasting its time on them. Once I added to robots the crawl level went back to 'normal'.
-
There could be several factors... maybe your brand based search is prompting Google to capture more of your site. Maybe you got a link from a very high authority site that prompts higher crawl volumes. Queries that prompt freshness related to your site could also spur on Google. It is a lot of guesswork, but can be whittled down some by a close look at Analytics and perhaps tomorrows OSE update (Fresh Web Explorer might provide some clue's in the meantime.) At least you're moving in the right direction. Cheers!
-
There are two variables in play and you are picking up on one.
If there are 1,000 pages on your website then Google may index all 1,000 if they are aware of all the pages. As you indicated, it is also Google's decision how many of your pages to index.
The second factor which is most likely the case in your situation is that Google only has two ways to index your pages. One is to submit a sitemap in GWT to all of your known pages. So Google would then have a choice to index all 1,000 as it would then be aware of their existence. However, it sounds like your website is relying on links. If you have 1,000 pages and a home page with one link leading to an about us page then Google is only aware of two pages on your entire website. Your website has to have a internal link structure that Google can crawl.
Imagine your website like a tree root structure. For Google to get to every page and index it then it has to have clear, defined, and easy access. Websites with a home page that links to a page A that then links to page B that then links to page C that then links to page D that then links to 500 pages can easily lose 500 pages if there is an obstruction between any of the pages that lead to page D. Because google can't crawl to page D to see all the pages on it.
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Odd 404 pages
Evening all, I've performed a Screaming Frog technical crawl of a site, and it's returning links like this as 404s: http://clientsite.co.uk/accidents-caused-by-colleagues/js/modernizr-2.0.6.min.js Now, I recognise that Modernizr is used for detecting features in the user's browser - but why would it have created an indexed page that no longer exists? Would you leave them as is? 410 them? Or do something else entirely? Thanks for reading, I look forward to hearing your thoughts! Kind regards, John.
Technical SEO | | Muhammad-Isap0 -
Why are my Duplicated Pages not being updated?
I've recently changed a bunch of duplicated pages from our site. I did get a slightly minimized amount of duplicated pages, however, some of the pages that I've already fixed are still unfixed according to MOZ. Whenever I check the back-end of each of these pages, I see that they've already been changed and non of them are the same in terms of Meta Tag Title is concern. Can anyone provide any suggestions on what I should do to get a more accurate result? Is there a process that I'm missing?
Technical SEO | | ckroaster0 -
Pageing page and seo meta tag questions
Hi if i am using paging in my website there is lots of product in my website now in paging total paging is 1000 pages now what title tag i need to add for every paging page or is there any good way we can tell search engine all page or same ?
Technical SEO | | constructionhelpline0 -
Mysterious drop in the Number of Pages Crawled
The # of crawled pages on my campaign dashboard has been 90 for months. Approximate a week ago it dropped down to 25 crawled pages, and many links went with it. I have checked with my web master, and he said no changes have been made which would cause this to happen. I am looking for suggestions on how I can go about trouble shooting this issue, and possible solutions. Thanks in advance!
Technical SEO | | GladdySEO0 -
Problem wth Crawling
Hello, I have a website http://digitaldiscovery.eu here in SEOmoz. Its strange since the last week SEOmoz is crawling only one page! And before it was crwaling all the pages. Whats happening? Help SEOmoz! :))
Technical SEO | | PedroM0 -
Mass 404 pages
Hi Guys, If I were to have to take down the majority of my site, taking all content and links pointing to that content down. How would the search engines react? Would I get a penalty for the majority of the site all of the sudden missing? My only concern is the loss of traffic on the remanding pages. Thanks!
Technical SEO | | DPASeo0 -
Product category paging
Hi, My product categories have 2-3 pages each. I have paging implemented with rel=next and rel=prev. from some reason Google GWT now reports the pages as having duplicate titles and description. Should I be worried? Should I set a different title like "blue category - page x" ? Thanx, Asaf
Technical SEO | | AsafY0 -
Consolidate page strength
Hi, Our site has a fair amount of related/similiar content that has been historically placed on seperate pages. Unfortuantely this spreads out our page strength across multiple pages. We are looking to combine this content onto one page so that our page strength will be focused in one location (optimized for search). The content is extensive so placing it all on one page isn't ideal from a user experience (better to separate it out). We are looking into different approaches one main "tabbed" page with query string params to seperate the seperate pages. We'll use an AJAX driven design, but for non js browsers, we'll gracefully degrade to separate pages with querystring params. www.xxx.com/content/?pg=1 www.xxx.com/content/?pg=2 www.xxx.com/content/?pg=3 We'd then rel canonical all three pages to just be www.xxx.com/content/ Same concept but useAJAX crawlable hash tag design (!#). Load everything onto one page, but the page could get quite large so latency will increase. I don't think from an SEO perspective there is much difference between options 1 & 2. We'll mostly be relying on Google using the rel canonical tag. Have others dealt with this issue were you have lots of similiar content. From a UX perspective you want to separate/classifiy it, but from an SEO perspective want to consolidate? It really is very similiar content so using a rel canonical makes sense. What have others done? Thoughts?
Technical SEO | | NicB10