shamster25
Regular Member
- Oct 12, 2008
- 230
- 166
Why doesn't the OP just give me his url, the keyword, and I'll get the site to #1 then?
[SB Case Study] Starting with nothing, taking it to the top
Why doesn't the OP just give me his url, the keyword, and I'll get the site to #1 then?
marketingwise
Junior Member
- Nov 25, 2009
- 109
- 12
Sharp. That is my goal to be sharp like htis
Because it's a case study / experiment.
If he does that, the conclusion to this would be worthless.
Besides, you should know nobody can ever promise #1 spot, as there's far too many variables to take into account.
Maruk
Power Member
- Jun 15, 2009
- 570
- 901
1) no I dont't use names. I use my keyword as an anchor.
2) no I don't get only no-follow links. A lot of them are not sporting a no-follow tag. Btw there is no such thing as a do-follow link.
My bubble is still intact buddy.
Maruk
Power Member
- Jun 15, 2009
- 570
- 901
I can't think of anyway to just pick out random keywords.
Perhaps you can search for a tool that will randomize the whole list and then you select the first 100 or something.
gorang
Elite Member
- Dec 6, 2008
- 1,894
- 1,678
There's some PHP code I found using Google but in the end I scrapped that idea. I've just dumped the 355k of keywords into scrapebox and i'm using 15 private proxies to scrape from yahoo only since I can't find any public proxies that are good enough for the job.
It is still running now and I already have 200k URLS so all is going well. I don't think i'll reach 8 million like you did though.
Important note, i'm using blogengine, because I don't want to go over your comments you made. You scraped wordpress right?
Maruk
Power Member
- Jun 15, 2009
- 570
- 901
Yes I did use WP, but if you scrape 4 million blogengine blogs you'll end up with about 4k unique domains buddy.
I'm using a very large list of keywords and I'm going through them alphabetically.
This way, you end up removing more duplicates first.
If you randomise the list, you're still going to get the same number of blogs, but the duplicates will be in different lists, meaning more duplicate posting.
I hope that makes sense
gorang
Elite Member
- Dec 6, 2008
- 1,894
- 1,678
I'm using a very large list of keywords and I'm going through them alphabetically.
This way, you end up removing more duplicates first.
If you randomise the list, you're still going to get the same number of blogs, but the duplicates will be in different lists, meaning more duplicate posting.
I hope that makes sense
Actually the reason I wanted to pull out like 100 keywords is because I don't think my proxies will last for the full 355k keywords even in yahoo.
But you've given me a good idea. I could split the keywords into lists which start with the first letter. So a list full of keywords beginning with A, then B then C.
Lol, or at the end of the run I could just export the unfinished keywords in the report SB comes up with(I'm being a dumbass again today).
What I did is simply use software to split them up. I split them up into blocks of 5,000 and harvest 5,000 keywords at a time.
So now I have a keyword folder with quite a lot of files to use when harvesting. Once each keyword mini file has been used for harvesting I move it to a different folder (in case I need to use those keywords with a different footprint in the future).
The tool I use is called 'log splitter'.
The reason why I chose 5,000 is because SB seemed to harvest too many results with 10,000 and SB would freeze. I might just run 2 instances of SB at a time and harvest. 30 connections each. 5,000 can still get you like 250,000+ unique URLs.
Maruk
Power Member
- Jun 15, 2009
- 570
- 901
About 67% and yes they will be split evenly. SB handles all that!
Via the 'link lock'?
So your first round of blasts is just for the root and then you go through the same URLs again with a 2nd blast linking to all your sub pages using the keywords you've optimised them all for?
T
temud
Guest
If you want to pick a few random items from a list you can use Excel.
- copy the list to column A
- put a random number in each according cell in column B
- sort on column B
- e.g. take the first 100 cel of column A
I think the name of the function for a random number in an English Excel is: rand
Maruk
Power Member
- Jun 15, 2009
- 570
- 901
I am not sure about using the linklock yet, I might do name+review for the corresponding url. And all new URL's.
The first blast was for the root domain indeed and with the second blast I hope to rank the actual review pages for product name + review.
black101
BANNED
- Dec 2, 2010
- 1,436
- 1,640
its not because the keyword has 2900 searches a month it means its easy!!!!!! it depends on the competition not the searches.
you can have a 1000 searches a month keyword and 1'000'000 exact comp. pages. it would take a long time to rank
Maruk
Power Member
- Jun 15, 2009
- 570
- 901
All unique domains, I don't want to piss off any webmaster
gorang
Elite Member
- Dec 6, 2008
- 1,894
- 1,678
All unique domains, I don't want to piss off any webmaster
Same. Just wanted to check.
Anyway the Yahoo scraping has been good, 550k harvested, saved the non completed keywords and i'm not even into the C's in the list yet.
Interesting re: Unique Domains.
So, do you use a tool such as Loopline's to remove the root, followed by leaving just 1 random URL on that domain?
Then post to it and if it sticks, then do a site:domain.com and post to the rest?
I had a list of something like 250,000+ shrunk to about 1034 doing that method so can only assume that's how many different domains were in the list.
Or do you use a different method?
So, do you use a tool such as Loopline's to remove the root, followed by leaving just 1 random URL on that domain?
Then post to it and if it sticks, then do a site:domain.com and post to the rest?
I had a list of something like 250,000+ shrunk to about 1034 doing that method so can only assume that's how many different domains were in the list.
Or do you use a different method?