Google Crawls Pages, Show Them in Serps then Drops Them

[zehrila]

I started this new site couple of months ago, its doing alright for its age. However, whenever i publish new content, i push that link in couple of social bookmarking sites to get my page indexed before my competitors. Google bot comes, crawls page and show me in Serps, but doesn't cache the page. After about 24 hours, it drops my page and then fully index it after 3-4 days. Meanwhile when it drops my main page, it caches subpages of that page.

So lets say this is my page, which i publish and push in Bookmarking sites.

Domain.com/product-name-widget (main)

and the subpages of the above mentioned pages are

Domain.com/product-name-widgets/specifications (sub)
Domain.com/product-name-widgets/details (sub)

I am wondering, why does it drop my main page after 24 hours and meanwhile cache my subpages?

[tedster]

After about 24 hours, it drops my page and then fully index it after 3-4 days.

Sounds like a fresh index ranking, followed by a main index ranking. I assume your question is about the 3-4 days where the page isn't in the index at all. At a guess, I'd say that the best cure would be if the URL attracts some natural backlinks from other sites.

...and meanwhile cache my subpages?

Once Google has discovered that any URL exists it usually will spider it. Do you mean to say that these sub-pages end up immediately in the main index and search results and do not go through the same cycle as the main page?

In order to diagnose what you're noticing, it would help to be very precise about technical words:

1. Crawl (or spider) - the most dependable way you learn about this event is from your own server logs

2. Index - this means process the crawl data so that the URL is then able to be ranked and show up in a search result

3. Cache - the only thing you can observe is Google's public cache. That's available through the cache: operator or the cached link within in the search results.

This public type of cache information is NOT the same thing as Google having a cached copy of the page in their back end to use in calculating rankings. The publicly viewable cache date is an undependable bit of data, although a back end "cache" must occur first. However, back end cache that is more recent than the public cache date is always possible

[zehrila]

True, its fresh index ranking and then couple of days later page show up again. I don't understand why it drops as 1: i submit it to couple of social bookmarking sites as well as FB and Twitter. I also do a short guest post of about 200-400 words on one of blogs where i have posting rights.

The process goes like this.

1: I publish new page.
2: Add it to social bookmarking sites and do a short blog post.
3: Page gets spidered and shows up in google when i paste exact URL. Most of the time, targeted keywords show up 1-3 in G. (not cached)
4: After sometime i can see other subpages showing up in serps for their keywords as well. (not cached)
5: After about 24 hours, main page disappears, putting exact URL in Google yeilds no result, but it shows all sub pages still there, but with Cache written next to them.
6: After couple of days, main page comes back in serps ranking 1-6 depending on how well it ranked when google first discovered it.

Apparently yes, subpages do not go through the same cycle. Even when i make a blog post, page gets discovered instantly (same domain), but never drops out and after about 10-20 hours, i can see the "cache" written next to the result. Even though, my main page has better on page seo, yet my blog posts on same topic out performs my main pages, power of wordpress? cms favourism?

I understand your points, i have read them before in some of your posts :)

[tedster]

My thought is that things might be different IF your new URL "catches fire" a bit, attracting a backlink or two that you can't create on your own steam alone. Your use of social media does give that process a chance to wake up.

However, the very predictability of your process (it's like a punch list) may be creating the same type of result in Google, over and over again. Maybe try a new form of publicizing the next article and see what you can learn.

Today's ranking algo is broken up into different query types - how a new URL is handled depends very much on the type of content, how important freshness is for a given query, if a given query is more for an informational search or a transactional search - things like that but broken down further in a granular way that still gets a bit inscrutable for me.