GSA SER Scraper - Please explain
2014-04-27, 07:28 PM,
#1
Hi guys,
I bought GSA's SER a couple of days ago and so far I'm quite satisfied with it. However, I'm not quite sure about how to properly scrape for URLs and I was hoping you guys would be able to help me. As of right now I'm using the SER scraper tool, because I don't have Scrapebox or any other scraping tool and my projects aren't that big so I don't need huge link lists.

So here's how I approach the scraping process:
1. I take my KWs and punch them into the scraping tool and hit go
2. ???
3. Profit

No, seriously. That's it and the results haven't been that great for me. I get like 2-4k links for the KWs I put in. My niches are quite narrow so I only have about 5-10 KWs (+variations).
This is how I scrape for relevant and contextual links for my tiered structure.

Now I have no freaking idea how to scrape for secondary spam links. They obviously don't have to be contextual and quantity > quality, but how do I approach this? When just putting in the foot prints and no KWs SER doesn't come up with any results.

I'm a little lost to be honest. Is there anything like a tutorial or a video that explains all of that?

Thanks in advance guys.
Reply
2014-04-27, 08:45 PM,
#2
Hi,

5-10 kw's really isn't much at all to work with. When most of us scrape, we use 100's or 1000's of keywords to bring in more link targets. You should click the "add predefined footprints" button then click the "+keywords" button and import your .txt file with keywords. You might already be doing it this way, but just wanted to make sure.

My suggestion would be to grab a bigger list of keywords to work with. There are plenty of scrapers out there to help generate keywords :)
Reply
2014-04-27, 09:07 PM,
#3
Right...Agree with s4ntos.

You need lots of keywords to scrape url. All of GSA user are using same footprint. They might have same keywords [incase of less keywords]
1) limited number of links
2) same links to user.

We have to remember that we do not want target url related to niche [because we are blasting] so do not stick to your keywords only. More keywords more Links.

And after scraping links ...Remove duplicate urls, domain etc to Get awesome and very juicy list
Reply
2014-04-27, 09:14 PM,
#4
(2014-04-27, 08:45 PM)s4nt0s Wrote: Hi,

5-10 kw's really isn't much at all to work with. When most of us scrape, we use 100's or 1000's of keywords to bring in more link targets. You should click the "add predefined footprints" button then click the "+keywords" button and import your .txt file with keywords. You might already be doing it this way, but just wanted to make sure.

My suggestion would be to grab a bigger list of keywords to work with. There are plenty of scrapers out there to help generate keywords :)

Where is located the "add predefined footprints" button ?
Reply
2014-04-28, 12:31 AM,
#5
Big options button > advanced tab > Tools > Search online for URLS > add predefined footprints.

[Image: predefines.jpg]
Reply
2014-04-28, 01:11 AM,
#6
(2014-04-28, 12:31 AM)s4nt0s Wrote: Big options button > advanced tab > Tools > Search online for URLS > add predefined footprints.

[img]/gg9mx3g7x/predefines.jpg[/img]

its really that simple add keywords and search engine--> scrape-->remove duplicate domain and urls--> decent list for backlinks?????
Reply
2014-04-28, 02:42 AM,
#7
(2014-04-27, 08:45 PM)s4nt0s Wrote: Hi,

5-10 kw's really isn't much at all to work with. When most of us scrape, we use 100's or 1000's of keywords to bring in more link targets. You should click the "add predefined footprints" button then click the "+keywords" button and import your .txt file with keywords. You might already be doing it this way, but just wanted to make sure.

My suggestion would be to grab a bigger list of keywords to work with. There are plenty of scrapers out there to help generate keywords :)

Great reply s4nt0s, thanks. I've posted on the GSA forum, too and I think my KW list was just way too small. I'm actually scraping as we speak, using a KW list of 1-2k. Let's see how that turns out.
The way I did it:
1. Select 3-5 main KWs. Think of a couple of related niches to my main niche (i.e. my niche is 'Ford Mustang', then I'd pick a couple of KWs for 'cars', 'vehicles', 'car accessories', etc. and then put all of them into GKT for suggestions. After that I took the newly suggested KWs + seed KWs and put them into Long Tail Pro Platinum to suggest even more KWs.
Is that a way to approach this, or should I be looking into Scrapebox & Co.?

Regarding the footprints - no, I don't use footprints right now. I understand these are just filters, aren't they? I mean if I only put in my KW list then SER will scrape for pretty much anything related to these KWs. If I use footprints + KWs then it'll return me a list for only the selected platforms which include my KWs, right?
I will definitely do this in the long run, but I wanted to do some test blasts first, without filtering out specific platforms and then see the results and cross out the ones with really low verification rates.

Also, I'm kinda wondering how I'd create a 'broad/global list'. The way I'm doing it now is just useful for contextual links and are handy if quality > quantity. But I'm building tiered links to my website with secondary links fueling my tiers. And these secondary links can be pretty much anything - the more the better.
How would I create such a list? Should I simply go broader and add more niches and KWs to my main KW to get a couple of thousand KWs and therefore more scraping results?
Let's take the example from above - main KW: 'Ford Mustang'. And to create a bigger/global list I'd go broader and use seed KWs like: cars, tyres, driving, etc. or even broader and use completely unrelated KWs? If this is the right approach then I guess I'll need a list of hundreds of thousands common KWs, right?
Reply
2014-04-28, 08:35 AM,
(This post was last modified: 2014-04-28, 09:32 PM by s4nt0s.)
#8
(2014-04-28, 01:11 AM)Conley18 Wrote:
(2014-04-28, 12:31 AM)s4nt0s Wrote: Big options button > advanced tab > Tools > Search online for URLS > add predefined footprints.

[img]/gg9mx3g7x/predefines.jpg[/img]

its really that simple add keywords and search engine--> scrape-->remove duplicate domain and urls--> decent list for backlinks?????

Yes. Some people will filter further by keeping only PR domains, etc. It's really up to you how far you want to filter down the list but overall that's basically how its done. Keep in mind SB or Gscraper are great scrapers for this sorta thing since they have a lot of list filtering options built in.

(2014-04-28, 02:42 AM)tixpf Wrote:
(2014-04-27, 08:45 PM)s4nt0s Wrote: Hi,

5-10 kw's really isn't much at all to work with. When most of us scrape, we use 100's or 1000's of keywords to bring in more link targets. You should click the "add predefined footprints" button then click the "+keywords" button and import your .txt file with keywords. You might already be doing it this way, but just wanted to make sure.

My suggestion would be to grab a bigger list of keywords to work with. There are plenty of scrapers out there to help generate keywords :)

Great reply s4nt0s, thanks. I've posted on the GSA forum, too and I think my KW list was just way too small. I'm actually scraping as we speak, using a KW list of 1-2k. Let's see how that turns out.
The way I did it:
1. Select 3-5 main KWs. Think of a couple of related niches to my main niche (i.e. my niche is 'Ford Mustang', then I'd pick a couple of KWs for 'cars', 'vehicles', 'car accessories', etc. and then put all of them into GKT for suggestions. After that I took the newly suggested KWs + seed KWs and put them into Long Tail Pro Platinum to suggest even more KWs.
Is that a way to approach this, or should I be looking into Scrapebox & Co.?

Regarding the footprints - no, I don't use footprints right now. I understand these are just filters, aren't they? I mean if I only put in my KW list then SER will scrape for pretty much anything related to these KWs. If I use footprints + KWs then it'll return me a list for only the selected platforms which include my KWs, right?
I will definitely do this in the long run, but I wanted to do some test blasts first, without filtering out specific platforms and then see the results and cross out the ones with really low verification rates.

Also, I'm kinda wondering how I'd create a 'broad/global list'. The way I'm doing it now is just useful for contextual links and are handy if quality > quantity. But I'm building tiered links to my website with secondary links fueling my tiers. And these secondary links can be pretty much anything - the more the better.
How would I create such a list? Should I simply go broader and add more niches and KWs to my main KW to get a couple of thousand KWs and therefore more scraping results?
Let's take the example from above - main KW: 'Ford Mustang'. And to create a bigger/global list I'd go broader and use seed KWs like: cars, tyres, driving, etc. or even broader and use completely unrelated KWs? If this is the right approach then I guess I'll need a list of hundreds of thousands common KWs, right?

1) Yes that's a great way to do it and of course you can always drop those keywords in SB if you want to generate even more keywords. Also there is a guy named furykyle that sells a pretty massive keyword list. That's what I use to scrape a lot of targets.

2) If you're only scraping with keywords you will be pulling in a bunch of random platforms that SER isn't capable of posting to so it will make SER do more filtering instead of running through a more targeted list where it would be doing more posting. You can do it either way but using footprints helps bring in platforms its capable of posting to so its more efficient.

3) If your second tier can be anything, then I think your second method is going to work better with a massive list of common kw's. As I mentioned before furykyle has a list. You can also look around on dictionary type sites or see if there's a database of kw's online somewhere. That should give you plenty of sites to scrape.
Reply
2014-04-29, 12:56 AM,
#9
s4nt0s

1) As of right now, I don't own SB, so I guess I have to stick with my method for now.

2) Yes.. I just found out the hard way. I've let SER scrape over night and I ended up with ~8k URLs and SER wasn't able to post to most of them (I'm talking 90% junk..)

3) That's what I've assumed. I might have to buy one of those huge KW lists and just scrape 24/7 to build a decent verified list.


Follow up questions:
a) I've downloaded a couple of free link lists and right now I'm just spamming the shit out of 'google.com'. This way I'm trying to find out which engines suck and which engines are worth posting to (verification rate wise). In the next step I was going to only use footprints, combined with my KW lists to only scrape for these platforms. My final results might not turn out to be 100k+ links, but I'll end up with very targeted URLs and therefore have a highly increased LPM, won't I?

b) Since you already brought up ScrapeBox - would you say it's an absolute must, even for a beginner like me? Right now I don't have that many proxies and no VPS (running everything on my computer).

c) I've come up with this method and I'd like to hear your opinion on it: I'm downloading a couple huge free URL lists and just let SER submit them. This way I'm able to build a solid global verified list and in addition I'll get some results so I can compare platforms against each other and throw out the ones not worth posting to them.
-> Bonus question: If I only want to verify links, will I still have to submit + verify them or can I make SER verify my link list without actual submission?

Thanks s4nt0s, you're incredibly helpful.
Reply
2014-04-30, 01:01 AM,
#10
Alright, unfortunately no answer from you yet, so let me just sum up how it went for me:

With my newly created KW list (1-2k) + foot prints of all the engines I want to post to, I'm already at 18.5k identified and 9.5k unknown. That's pretty good IMO, since these URLs will serve as contextual tiered links. I won't need 500k of those, since their quality is more important than their quantity.

While the scraping continues I've downloaded a couple of free verified lists and right now I'm just blasting the shit out of 'google.com' to build a global verified list that's as big as possible. This list will be used for secondary/junk links. Right now I'm firing away @ 35LPM, which is quite good IMO, since this is @100 Threads and using only scraped public proxies.

I guess I'll keep growing my verified global list until I deem it big enough and then start my first project with the very targeted list, I've scraped in the meantime.
Additionally to that, I took your advice and searched for a 'generic KW list' and I found one. It's not that huge, only 7.5k generic KWs, but I assume if I were to punch them into a KW suggestion tool of my choice, this number might easily double, or triple.

Even though I solved some of my problems it'd still be great if you could answer the questions above.
Thanks again mate.
Reply


Possibly Related Threads...
Thread Author Replies Views Last Post
  Kontent Machine VS Wicked Article Creator ,WHICH IS BEST TO RANK FOR GSA SER? trafficbeast 7 2,031 2017-09-14, 07:28 PM
Last Post: fionix
  Planing to buy GSA SER, GSA AWS, GSA SEO Indexer EagleEyes 7 377 2017-04-24, 07:37 AM
Last Post: EagleEyes
  GSA Search Engine Ranker - Is it REALLY worth buying? TylerDurden 12 1,753 2017-03-09, 07:33 AM
Last Post: arvinzx20
  GSA SER Adult site ranking advice RaZoR 1 442 2017-01-21, 03:16 AM
Last Post: shaunm
  safest and most recommended gsa list? Fierce 0 293 2016-12-08, 09:26 PM
Last Post: Fierce




About Us | Contact Us | CPA Elites | Advertise | Stats | Staff Team

© 2013-2017 CPA Elites Ltd
Enhanced by MyBB and WallBB
Return to top