Google Hot Trends+Clickbank=$$$

Just to add my 2 cents to this. Since June I have been working off and on (maybe one weekend a month tops) on building out a site to automate a lot of this. Some of the features include:

-Grab both Google and Yahoo trend data and store in MySQL (Via cron)
-Dynamic page that uses keywords to grab RSS feeds from three large sources of content
-Dynamic link structure for easy crawling by SE spiders
-Mod_Rewrites for clean URL's
-Contextual CB ads
-Dynamic RSS feed
-Dynamic Sitemap (rebuilt once a day via cron, Holds ~45,000 URL's at any point)
-Pages SEO'd (generally pretty good, but probably could use a little attention)
-Everything is custom build in PHP by me.

Been storing keywords since June and DB is roughly 83,000 unique keywords.

Before spending the day cleaning up code, chasing down bugs, and testing, the site had roughly 1,000 pages indexed in Google with a very limited sitemap (three main URLs only) and no link building. Just launched new sitemap with 45,003 urls to Google, Yahoo, and Bing. Also working on link building the rest of the night (is 6:24pm here now, will work on link building for next 6 hours or so).

Tracking stats with both Google Analytics and Statcounter. Before all the updates to the site, traffic was only 15-20 a day roughly from SE's with no revenue (used Adsense before today).

I'll keep you guys updated on it if there is interest.

Thanks guys for all your help and knowledge from your postings. I've learned much!

It sounds like you crafted something beautiful there.
Is it a regular website or blog ?
It sounds like regular.

Are the RSS feeds full text news feeds, or only the teasers ?
Full text news feeds are harder and harder to get.

I'm still making about a good couple of hundred bucks a week from quick hits I did a couple of months ago.

It's great

Hey g-Vector

It is a regular site.

The rss feeds are teasers but I may make them full feeds. I can grab, slice and dice all the new content in the world and generate full text feeds and it may be a good idea as I can provide that outwards to anyone and receive the links back. I've got a free Saturday today so I will play with that and post back in a couple of hours if I go that way.

If anyone has a reason why I should stay away from producing full text feeds out, please let me know. Any additional input I can get is awesome.

As far as stats go, it's been too early in my opinion to see an uptick in site traffic from this approach. I need to do a lot of link building and build some link bait today for some automatic link generation to really boost traffic.

Indexed URLs did go up ~4.5% in the last 18 hours(since I launched overhauled dynamic sitemap) going from 1,090 to 1,140 currently. Will continue to watch and report if interest remains.

Like I said, I have a full Saturday and nothing to do, so I am going to play with this some more and report back if interest remains.

Thanks all.

@Yukinari84: Would you mind sharing your approach to this strategy since you are making money in the long haul and not just the short haul?

Man don't make me feel bad.
I didn't mean to ruin your free Saturday with work.
I'm just asking because I find it interesting what you wrote, not to put you at work.

With (full) feeds, you have to mention the source, unless you manage to re-write the feed on the fly to call it your own. You would be almost up there with the big boys where you're pulling the feed from.

The only problem with hot trends is that when you find them and create a blog for them even if you are fast enough, the trend might have gone down and most of them don't usually stay for a day. So anyone else is using this and had results?
I know it's a noob question just wanted to know and keep this thread alive if the method still works. And because this is not a saturated market, it will never be... right???

@g-Vector:

This is fun stuff for me, not work. This is my hobby time.

Rewriting the feed on the fly is easy. What I will probably do is create a feed for each keyword. This way each page has a feed. A page on my site is defined by a keyword and all content is grabbed to match that keyword. So I can grab all that content, then generate a feed from that content and put it out. I can easily have it changed enough and then embed links in the feed back to the different sections of my pages. Pages are made up of a news, blog, and video section and monetized with contextual CB ads.

@Thunderelvi: This method does work, but the biggest problem I see is the manual labor aspect of it. My goal is to automate the entire process. If I can do that, then I will bypass the delay of manual labor and strike keyword trends the second they are hot.

I haven't come up with a very effective method for getting the links back to my site automatically yet, so if anyone has an idea, please let me know. I can get the links and content out via RSS feeds that each direct back towards a targeted page on my site. I know the answer is here in BHW, because really, what answer isn't?

As always I'll keep posting on this topic if others are interested.

@THUNDERELVI:
Some trends don't die. They keep reoccurring.
Just let the articles/posts and comments age.

If you go to the Hot Trends site and click through on the newest ones in the list, you'll find that most news sources (Google News, Yahoo News) have "news" from the previous months or even year. When the hot trend was also a hot trend.
You have to realise that most of the hot trends are initially spreading like a rumour.
There is no real fresh news yet.
When a gossip/rumour/event has just entered the Hot Trend 100, it is based purely on search volume.
Hence the hunt by Google for sites with FRESH news about it.

It takes at least 1 hour for the first real news to be indexed. Otherwise there was a psyhic or clearvoyant reporter who already had the news typed up online.
Initially old/existing news is presented when you click through.
It might be years old, but people read it.

For example the news about the American Football player who was found shot in the head or whatever, initially was old and stale news (for more than half a day) about his affairs and his community work. Not relevant at all.
Same with the woman who had her umpteenth child. Initially only articles about her previous pregnancies and new borns.

@g-Vector:
That's exactly right. That's the big reason behind my approach to automation. My goal is to blend together major sources automatically so that I can run with first-movers advantage on any topic. And even when fresh news comes across the wire, I can be the first to grab new news from across the spectrum to again take a first-movers advantage on the current news for any given trend.

This is where getting that out there is important. Google will only come and index me so often, so I have to help it along as much as possible. Still sketching out some ideas on this one.

@bfellow:
OK, then.

You are right, Automation is the way to go. But you can not just automate all trends. Many are not commercially viable. Like I said earlier, the trend at number one is not necessarily the most profitable, or even impossible to monetize.

Automation depending on keyword matching would be ideal.
Or matching based on what I quoted earlier from the first step of the course.
But some of that would lean towards AI :

The Hot Trends WP plugin which I bought is pulling feeds depending on 2 parameters.
The position in the Hot Trend list and and the news source (fixed to Google News, Yahoo News, and YouTube videos).
Pulling too often from Yahoo news bans your IP, for hours, or days, and sometimes it's broken for more than a week, with parse errors like "expected a ';'at position..... blah, blah,blah".

It (the WP plugin) is a far from ideal mechanism, because you're pulling news for the same spots all the time. For instance I had syndications for spots #1, #2, and #3.
But if spot #5 was better to monetise, than the other 3, I would have to go in and add a feed pull, or change one of the others.
So I would call the plugin semi-automatic. You still have to watch it and go in when necessary, to manipulate it.
Even if the spots would change every hour, the spots you're pulling for could still be for something not commercially viable.

It would also blindly pull in YouTube videos which turned out to be a lure to someone else's CPA Offer : Title contains the exact hot trend keword, the video or slides contained a URL where to watch the full video, because YouTube had taken it down due to violation of whatever. Yeah right, the URL always went to a CPA or affiliate offer.

It also has a 3rd mandatory parameter, your purchase receipt# as protection against piracy. That is fine with me, but it is buggy as hell, sometimes it complains that the receipt# does not match, and will refuse to pull the feed. at:

Sorry about the long post.


EDIT: Craps I see you posted already again. That's punishment for my long post.

@g-Vector:

I ran into these issues back in June when I was first playing with this concept. The best way around that I can see is to have a script that does the following:

-Grabs _GET['Keyword']
-Checks against the DB to see if keyword already exists
-If yes, then check to see if there is stored content for keyword
-If yes, then check age of content.
-If age of content is over 24 hours, update content
-If keyword isn't stored, store keyword
-Grab keyword content and store in db

That's obviously very rough. But the thought behind it is to only grab new content that is in demand. If I have older keywords that arn't being demanded, I am not going to run blanket content grabs against RSS feeds to prevent me from being banned. And if something is in high demand, I am parsing the information to again keep from say Google News or YouTube from being hit hard from a single IP. If I only do one content grab per keyword in demand per 24 hours, then I can keep from being banned.

Love you guys you're so fast at your responses.
So this works, but we have to automate it and leave the automated programs do their work all the day on a dedicated server 24x7. That would be our GOLDMINE. For e.x:

1. A program extracts the hot trends from GOOGLE/YAHOO and all the keywords related to the trends, as the second the hot trends appear(if they're volcanic/fire the better) without letting them reach their peaks. The competition for the keywords related to the trends are checked(typed in google with quotes. If it's less than 100k it's very good. The Program notifies you.

2. A simple program/plugin extracts the articles/feeds from the web or similar sites based on keywords, then a spinner/re-writer makes this articles unique and fully SEO with the keywords related to the trend which we got in step 1.

3. The articles are posted in a blogspot/wordpress blog, which has a nice theme/template, in a well formatted manner and also contain pictures and videos from youtube related to the article type/niche.

4. I don't know if this can be automated, but we put little effort to placing adsense blocks on the blog in the best positions and also put some cloacked links to CB affiliate/own products or CPA offers.

5. After the creation of the blogs is done, index them to google and start sending traffic.
You guys already know better than me what are the best automated or not, methods to drive traffic to the blogs. If the blog has success and the topic still remains popular and searched, different posts(articles + pic + videos) are made daily to it on different time intervals by using an automated plugin.
If not forget about the blogs.

6. This entire creation of the blogs fully SEO and monetized, submission and pinging to the social bookmarking/pinging/rss-feeds sites should not take more than 1 hour, max. 1 hour and 30 min. The idea is that you dominate the market for that day, and the best thing is that your site is live and indexed(in front page #1 of G.) when the peak reaches.

So, when thousands of people search for that hot trend, your site is on the #1 spot and they get to your site. You can do this to all the trends, but we want to make sure they can be monetized. Ok guys these are just suggestions, I don't know how many things of what I said can be done on autopilot but manually they can be done but will take time.
And if you say time=money, then in this case it's totally 2x more true!!!

Plz any suggestion from you guys to take this method to the next level and make it fully/almost automated is appreciated. Less Work To Be Done=The Better.
Thank you!

And the suggestions of g-Vector were good too...
Sorry didn't see that posted!

@THUNDERELVI:

Essentially I have the site part all done. Everything is automated. I take a little bit more of a shotgun approach as I grab all Google Trends and Yahoo Buzz keywords and generate pages for them. Ads on the pages are CB ads and dynamic. If the ads don't pull relevant cb oroducts, then they grab the most popular cd products, so I am basically hoping a popular product will catch someones eye on a keyword that doesn't reflect a product in CB. Make sense?

I don't do any check against Google SERPs as I go after every keyword that is published on my site. I have thought about building a backend with some serious reporting features in it and this would be a good tool to have. I have some firefox plugins that I use for checking up on the SERPs. I need to make this more automated thought as I still have to generate new keyword text files from the DB and manually upload them to my rank checker.

My afternoon right now is all about automating the link building process. I've got some notes down and am making headway. Hoping to test a system by early this evening and will post.

THUNDERELVI and g-Vector, since you two are active and interested in this area, PM me and I'll hook you guys up with my URL so you can see what I'm working on. I just don't want to openly post it here just yet.

bfellow said:

@g-Vector:

I ran into these issues back in June when I was first playing with this concept. The best way around that I can see is to have a script that does the following:

-Grabs _GET['Keyword']
-Checks against the DB to see if keyword already exists
-If yes, then check to see if there is stored content for keyword
-If yes, then check age of content.
-If age of content is over 24 hours, update content
-If keyword isn't stored, store keyword
-Grab keyword content and store in db

That's obviously very rough. But the thought behind it is to only grab new content that is in demand. If I have older keywords that arn't being demanded, I am not going to run blanket content grabs against RSS feeds to prevent me from being banned. And if something is in high demand, I am parsing the information to again keep from say Google News or YouTube from being hit hard from a single IP. If I only do one content grab per keyword in demand per 24 hours, then I can keep from being banned.

Click to expand...

Looks good to me.
But when a trend does have regular updates (every few hours) you'll lose out on that. So you need some way to override it for one or more trend (look further down). That, and I do not see automatic matching for monetisation yet. But I realise that that one will be quite hard.

Also, you can safely pull a feed more often than once every 24 hours when you check it's announced update schedule. I know that FeedWordPress only does the actual pull when the feed was refreshed by the feed supplier.
In WordPress's backend, the FeedWordPress plugin displays the time in minutes and hours when the next announced update of the feed will be ready to pull. Break that schedule and Yahoo ban's your server's IP after a few attempts. Google is a little more forgiving.
Many people didn't understand why the FeedWordPress plugin would refuse to pull content when it was commanded by another script or plugin to do the pull, while in fact it covered their ass.

@g-Vector:

I believe I have the matching of CB ads to relevant keywords figured out, just need to increase the volume of traffic first to test, then make adjustments from there.

You are right on missing out on content, especially for trends that have a life of less than a day. Checking the announced update is a good idea. Another idea I have had is setting up scripts on a cron schedule at several ip's. Say I set up 12 ip's that run a script that just checks for updates. If an update has occured, then it grabs the content and deposits it in my central DB. And from there my DB can send out the content to the pages, avoiding a ban and updating the site on regular intervals every two hours (or more with more IP's).

THUNDERELVI said:

And the suggestions of g-Vector were good too...
Sorry didn't see that posted!

Click to expand...

And I just missed your's while I was typing. Hot Trend visitors do not clicking much on Adsense ads. Check out the WP blogs of the writer of the Hot Feed WP plugin which I posted eralier. These have Adsense ads on it. He told me that he made money through Adsense but not as much as expected. Check his (from US only) traffic figures at Compete and you will see what I mean if I tell you that he made less than $100/month in Adsense from that kind of traffic.

Also the bounce rate with Hot Trend visitors is extremely high, around 76% for those 2 sites when I last checked at Alexa.

bfellow said:

@Yukinari84: Would you mind sharing your approach to this strategy since you are making money in the long haul and not just the short haul?

Click to expand...

If you actually take time to read through the whole thread, you would have seen that I already did post what I do.

I don't do this method for the long haul.

It's a quick cash method for me. I just so happened to land on the right combination of KW and marketing for some of my promotions is all.

bfellow said:

@g-Vector:

I believe I have the matching of CB ads to relevant keywords figured out, just need to increase the volume of traffic first to test, then make adjustments from there.

You are right on missing out on content, especially for trends that have a life of less than a day. Checking the announced update is a good idea. Another idea I have had is setting up scripts on a cron schedule at several ip's. Say I set up 12 ip's that run a script that just checks for updates. If an update has occured, then it grabs the content and deposits it in my central DB. And from there my DB can send out the content to the pages, avoiding a ban and updating the site on regular intervals every two hours (or more with more IP's).

Click to expand...

You also have to take care that Google sees the feed/content as coming from the same domain. They shouldn't see that you are pulling from outside the domain where you are using the (rewritten) feed. The version of the Hot Feed plugin cloaks the URL of the feed pull so it looks local.
Two examples of a URL to syndicate a feed would be ull from Yahoo News for the #2 spot: Pull video from Youtube for the #1 spot:
&yn=1 : Pull from Yahoo news
or
&gn=1 : Pull from Google News
and with &yn=1 and &gn=1 both omitted : Pull video from YouTube.

&n=2 : For Hottrend# 1 - 100 (In this case for the #2 spot)
and with &n=spot# omitted : Pull for the #1.

YOUR_RECEIPT_NR = purchase order # (buggy piracy protection)

Don't mind Yukinari84. He has a vision we don't. I mean this. Just check up on his earlier posts.
He's maybe one of the clearvoyant reporters I mentioned.