« Day 7 Diary - It’s Judgment Time - Part 1 | Home | Day 8 Diary - Let the Judgment Continue »
Day 8 - The Most Important Day Of The Challenge
By Mike Mindel | August 9, 2007
This content has been summarised by Mike Mindel from day 8 training of the Thirty Day Challenge. The voice on the videos is that of Rob Somerville.
The Big One
You may want to sit down.
This is it.
Today we reveal the most extraordinary method for verifying traffic. This is the holy grail of internet marketing and you guys are getting it for nothing in the Thirty Day Challenge.
As experienced internet marketers will tell you, it is stunning.
It’s an absolute license to print money.
A Few Words of Warning
We originally, as we do with all of Dan & Ed’s stuff, release it to Immediate Edge members. This is not a pitch for Immediate Edge. In fact I don’t want you to go anywhere near there.
I only mention this as Rob mentions edgers in the video as it was made for the edgers originally. Just pretend he said Thirty Day Challenge. I mention this by way of disclosure up front.
The Jargon
There are a few terms in here which are jargon you have not learn yet. Again don’t freak out. Just focus on what Rob is doing on the screen to determine traffic.
Let me give you the context for this.
For so long when people have tried to get free traffic they have used all sorts of tools.
I think Wordtracker is by far the best (thanks Ed). But all these tools are educated guestimates of the traffic.
The only company that has the real traffic data is Google, Yahoo & the major search engines. But mainly Google with 50% of the traffic and growing. This is where we get the bulk of the traffic early on.
What Would Happen Prior to This Method
You do all this work on a keyword to try and rank for it and you achieve it. Then you ask yourself wheres the free traffic? It’s gone. Where is it! I don’t understand. Wordtracker or wherever said I would get all this traffic if I was in the number one spot.
This was an incredible frustration for years. It has been such a bug bear. It was like rolling a dice. There was no way of verifying this trafic. Until today.
That’s what is so exciting!
There’s No Excuse For a Dud Keyword
You can put just as much work into a dud keyword as a good keyword. There’s no excuse for working with a dud keyword after today. That doesn’t mean that down the track some of these phrases won’t be good. But the difference between the traffic you will get is enormous and it may take some time to build up particular traffic for these keywords.
So you’ve got to be careful!
The Secret
This will be the most successful thirty day challenge yet as we will be targeting little, minor league keywords that will be delivering you traffic straight away.
That is the secret.
Right now we are going to test each of your niches to see if you have keywords will actually stand up. What you might find now is this process will whittle them right down.
It All Happens Right Here
Get this right and the rest of the month is a downhill slope.
Get it wrong and you will have to keep revisiting until you get it right.
Google Trends
Finally, the truth about Google Traffic.
This is a fantastic traffic confirmation tool using Google Trends. This is a way of getting to the truth, specifically regarding the traffic potential that Google might offer in relation to a keyword.
Verifying Traffic
The problem is that sometimes the traffic potential predicted for keywords in Wordtracker are wrong. When you go out there with what seems like a good keyword & you create a Squidoo lens you don’t get anywhere near the traffic you were anticipating based on the initial keyword research.
This has been a challenge for quite some time now. So I had an inkling there was an issue with the Wordtracker numbers. So we (the edge team) compared the numbers in the Wordtracker database and the Keyword Discovery database.
Unfortunately Keyword Discovery is very expensive and we needed a more approachable way of confirming those numbers.
I had a flash of insight using Google Trends which for some time we have been using for cyclical (seasonal) traffic trends.
Since Dan & I rank very well for certain keywords we were able to correlate the Google Trends graphs with known traffic we are receiving from Google with lenses we happen to be ranked well for.
So at this stage I am going to offer you a baseline using the keyword ‘male yeast infection’
We can see that there is slightly declining traffic for that keyword over the last 18 months to 2 years (it picks up mid 2007).
But at the current time if you have a website that is ranked in the top 1 or 2 for that keyword you would expect to get 500 - 550 searches a day coming through from Google.
We know that because Dan happens to hold the number 2 position for that keyword. So this position here (when the video was taken)
correlates to 500 - 550 searches per day.
Now if we scroll down and look at the traffic in relation to Regions, Cities or Languages
you can see the spread of that traffic is in the four Western countries: Canada, United States, United Kingdom, Australia. (I just performed the search on Google trends just now and Australia didn’t show). But the other three countries did as you can see - in similar proportions.
We’ll be exploring these bars later as there is a particular time when you are exploring these bars. For instance when the keyword you are searching does not register enough traffic to show in the main graph above.
We Have a Baseline
The real value of this tool is that now that we have a baseline. We have a known traffic keyword with a known volume. So now we can cross correlate a keyword and compare directly using Google Trends.
Note: The traffic for both the broad match ‘male yeast infection’ and exact match ‘male yeast infection’ is virtually the same
See there is only about 10% in it. So don’t worry if you see male yeast infection with or without quotes in the next few examples. It’s the same traffic.
Example: Updos Prom Hairstyles
Here’s an example where Wordtracker shows traffic for a keyword that doesn’t register in Google Trends.
Updos prom hairstyles registers significantly in Wordtracker
and Keyword Discovery
We can immediately go to Google Trends and enter in our baseline keyword ‘male yeast infection’ and then our comparative keyword ‘updos prom hairstyles’ and we get this
Google shows us the traffic associated with the broad match keyword ‘updos prom hairstyle’ against the traffic for ‘male yeast infection’.
Now we know that the blue line for male yeast infection represents about 500 - 550 page views / searches a day for a website ranking one or two (the lab has those rankings).
But now we can see that if we had a website ranking in position 1 or 2 for the keyword updos prom hairstyles then at certain times of the year relating to the prom seasons in the US there would be plenty of traffic.
In this case about 1500 page views / searches. That’s 3 x the baseline for ‘male yeast infection’. (Actually in my new diagram the difference is even more extreme. It looks like 4 x the traffic).
But there are periods of time in the year when there is no traffic at all. The reason is that the traffic associated with this keyword is highly seasonal and related to the prom seasons.
This is assuming that we hold a good ranking for the broad match for updos prom hairstyles (i.e. no quotes).
Why Are Wordtracker Results so High?
Wordtracker is taking the total of the last 90 days worth of traffic. So any spikes or surges of traffic in any discrete period during those last 90 days will increase the search volume overall.
E.g. if a keyword gets 0 traffic for January and February but 600 searches in March then the total is still 600. It will rank alongside a keyword that gets 300 searches for January, Feburary and March.
Back to the Hairstyles
Now lets try the same graph with the exact match i.e. with quotes
Now the picture is completely different. In effect there is no traffic for the phrase match ‘updos prom hairstyles’.
What is probably happening here is that on a broad match basis, most of the traffic here is associated with the keyword ‘prom hairstyles’ which one would expect.
But Google is telling us that there is no traffic for the keyword ‘updos prom hairstyles’.
Using Google Trends for Both Broad and Phrase Match
The Google Trends is extremely powerful as it shows you how much traffic you would expect to receive for both broad match (without quotes) and exact match (with quotes).
When you first setup a Squidoo lens you are probably going to rank well for the phrase match for that keyword.
The traffic in the first week or ten days is usually associated with the phrase match. Overtime as Google seeds you as more of an authority for a keyword then you will rank for the broad match.
But initially the traffic you get from Google will come from the phrase match rankings.
How To Build a Deck
Lets try another keyword where traffic exists for both broad match and phrase match. The keyword ‘how to build a deck’ gets
783 searches per day in Wordtracker.
With both broad match in Google Trends
we see the keyword is highly cyclical and the high traffic periods are during the summer time when naturally people are interested in building a deck. Whereas in winter time the keyword really drops off.
In periods like this
the traffic either fell to a very low level or Google was not updating its database in relation to that keyword. (I suspect traffic drops below the 250 searches / day baseline).
In off-season
the traffic builds from nothing through 500 a day and beyond.
In the peak seasons
we would expect nearly twice as much traffic.
Now lets look at the phrase match estimation in Google by putting inverted commas around the keyword and
we’re still getting traffic! It’s not quite as much as a broad match but there’s still a lot of traffic tied to that specific keyword.
If people enter ‘how to build a deck’ into the search engine then most of the time they are going to enter them in that exact order. Whereas there will be other keywords where this is not the case.
Now lets look at all three together
We’ve got the baseline in blue. Notice the upswing at the end there. But remember that at the beginning of 2007 this correlated to about 500 searches a day.
Now see that both the broad match and phrase match for ‘how to build a deck’ have the same overall pattern and plenty of traffic overall. The broad match gets about 20% more traffic than the phrase match.
This means if you get ranked for a Squidoo lens (say) and start getting ranked in Google then traffic will arrive for your exact match.
When You Don’t Get a Graph
There are circumstances when the traffic potential for a keyword is so low then it doesn’t get displayed in the graph. In my experience if the traffic falls below 200 - 250 searches a day then you won’t see a graph.
That’s when the bar graphs come into play.
Intimo Lingerie
Here’s a keyword for which I (Rob) happen to hold the number 2 and number 4 position - intimo lingerie.
This keyword has absolutely no commercial value. I’ve never had a sale and never expect to have a sale. Intimo lingerie is sold by party plan and has no online purchase function. This is why I’m quite happy to share this keyword with you.
There’s very little traffic associated with this keyword. It’s way under 100 searches a day. So no red line appears on the graph. The blue line is our baseline ‘male yeast infection’ keyword not the ‘intimo’ keyword.
But we do get information in the lower graph.
Nearly all the traffic comes from Australia. Hardly any from anywhere else. This is perfectly consistent with the fact that Intimo lingerie is based in Australia. This confirms what I already know.
But it’s interesting to see that this is indeed the case.
When the Traffic Volumes Are Too Low
When the traffic doesn’t show in the upper graph I look in the Languages section and compare the blue and red lines.
Most of the time the traffic is English for both these keywords ‘male yeast infection’ and ‘intimo lingerie’.
So in effect we can assume that the blue bar represents the traffic volume we know we’re getting 500 - 550 searches a day.
Note the little red bar associated with ‘intimo lingerie’ and we treat it as a proportion of the blue bar then we’re getting 40 searches a day in position 1 or 2 for the position ‘intimo lingerie’.
I happen to be in position 4 for the keyword and I’m getting 20 - 25 searches a day which is consistent with the drop off in traffic from positions 1 through 4 on the first page of Google results.
This is entirely consistent with what Google Trends is telling me.
So remember if there is no traffic in the top graph then look in the Languages column to see if there is any comparative traffic between the blue & red bars.
Use it to confirm or deny the potential traffic for keyword as defined by Wordtracker.
It’s only telling you what the traffic potential is from Google. When you start out most of your traffic will
probably come from Google. So it’s entirely consistent with the traffic you would expect to receive in the first couple of weeks of setting up a Squidoo lens.
Overtime you will get traffic from many more sources than just Google.
Identify a Candidate Umbrella Phrase
You may recall before we were looking at the keyword ’speed reading free’. It had acceptable competition, under 1000 competing web pages.
We’re now going to explore whether it has any traffic in Google for that specific keyword.
Remember the blue line is for ‘male yeast infection’ which delivers 500 - 550 searches a day to the website. (I suspect this was true at the beginning of 2007 but may be a bit more now because of the spike).
But look!
There is no traffic for the phrase match ’speed reading free’. It will only register if there are at least 250 searches a day. There isn’t even 250 searches a day for ’speed reading free’.
Lets look at English languages section
Comparatively speaking we would expect about 10 searches a day. But it’s not zero.
This is less than the traffic predicted by Wordtracker. Wordtracker predicts 54 searches for ’speed reading free’ with about 60% of those searches coming from Google. So a prediction of 32 searches a day.
The actual traffic is a third of that - at around 10 searches a day.
As an umbrella keyword phrase this really doesn’t cut the mustard. It doesn’t justify the effort to capture some of that traffic.
A Keyword That Does Meet the Criteria
Lets type in ‘build a deck’ into Wordtracker
Wordtracker is suggesting the highest traffic keyword is ‘how to build a deck’, a five word keyword. You wouldn’t expect so much traffic for a five word keyword than a three word keyword like ‘build a deck’.
So this is a good starting keyword for us to look at.
Let’s head on over to Google and enter ‘how to build a deck’ in inverted commas
the results show
23,000 competing web pages.
This is within our 25,000 - 30,000 or less suggested guideline so this makes a good candidate umbrella phrase.
Now over to Google trends
and the traffic for the exact keyword ‘how to build a deck’ is good. It’s as least as good as ‘male yeast infection’ for a good part of the year (during the summertime) so you are guaranteed at least 500 - 550 searches and probably about 50% more - could be as high as 750 - 800 searches a day.
But there is clearly under 250 searches a day in the winter. Rob thinks that Google doesn’t track this data during these periods but my suspicion is that the count just drops below the threshold of 250 searches a day and Google reports that as a straight line.
Wordtracker reports 800 searches total. Google is confirming significant volumes of traffic at particular times of the year.
A candidate umbrella phrase with good number of searches (80-100+), low competition (25,000 - 30,000 or less), good Google Trend graph compared to our baseline.
Summary
- Get candidate keyword from Wordtracker
and note number of daily searches.
- Go to Google Trends and enter ‘male yeast infection’, ‘keyword’. Keyword is your candidate umbrella phrase. Click Search Trends.
- Compare the graphs. Note that just after the beginning of 2007 correlates to about 500 - 550 searches a day (a known amount of traffic). See how the traffic of your phrase match keyword compares.
- Should be similar to 60% of Wordtracker estimate (Google share).
- Sometimes Wordtracker says there’s plenty of traffic but Google trends says otherwise. This can be because Wordtracker is counting totals only over last 90 days. Spikes in traffic in any discrete period can distort the number of searches overall.
- Select an umbrella phrase based on high number of daily searches (80-100+ ideal), low competition (25,000 - 30,000 or less), a fair amount of Google Trend traffic.
- Note that no graph at the top means there is less than 250 searches per day. But not necessarily no traffic. Compare red line to blue line and make a comparative guess as to how much traffic your keyword will get.
- Note: It doesn’t matter if you enter ‘male yeast infection’ with or without quotes as the traffic difference is only about 10% between them. I.e. you get similar traffic for broad & exact match. So don’t fret on that point.
Once you get your candidate keyword we can move on to stage 2 of The Thirty Day Challenge.
Comment from Mike Mindel
What a great, great explanation. I have never seen this kind of market research before online. This is truly an internet first. Well done guys.
I like it.
I like it a lot.
Caveats
Note: Often the word phrase, keyphrase or keyword phrase is used throughout one video and keyword in another. For consistency I use the word keyword to mean one or more words separated by a space. So ‘music’ is a keyword as is ‘world music’. There is no right or wrong answer but I aim to be consistent in all my articles. I will however keep the term ‘umbrella phrase’ as its specific to the Thirty Day Challenge.
Note: This video was created during the early part of 2007. I am running the same searches in August 2007. So there will be some differences in the graphs. But in all cases the basic concepts hold true.
-Mike
Technorati Tags: day 8 training, thirty day challenge, wordtracker, free keyword tool, mike mindel, ed dale, rob somerville, the big one, squidoo lens, squidoo, dud phrase, the secret, google trends, male yeast infection, baseline, updos prom hairstyles, high results, hairstyles, broad match, phrase match, how to build a deck, intimo lingerie, when traffic volumes are low, no graph, candidate umbrella phrase, umbrella keyword, candidate keyword, build a deck, build a deck above ground pool
Powered by ScribeFire.
Topics: Daily Summaries |
August 9th, 2007 at 12:47 am
Nobody is going to believe me, but I was thinking about two weeks ago that if I had a ranking keyword I could figure out the traffic volumes in this way (with Google Trends).
When I saw the video my jaw hit the floor.
August 9th, 2007 at 1:23 am
[…] Day 8 - The Most Important Day Of The Challenge […]
August 9th, 2007 at 1:26 am
I think Google is going to start wondering why there is a spike in their Trends logs of so many people trending on “male yeast infections”. Classic.
What a fabulous method.
PS Thank you Mike for your summaries and diary entries. Much appreciated.
August 9th, 2007 at 1:38 am
Mike Mindel, you are absolute gold!!! Your magnificent summaries supplement my meager notes on vids and podcasts. For the ‘Newbiest of newbies’ (me) your whole blog has been an invaluable resource. Thanks again, Brad
August 9th, 2007 at 2:23 am
LOL @ Male Yeast Infections will SPIKE to an all time high and hold there…Google will wonder why…the medical community will be alerted…and Dan can post all the updated on his MYI site. LOL
This is such a nice hard copy of today to have!
Bless you, sir!
August 9th, 2007 at 4:27 am
Hmmmm…first there’s Dan and his infamous pink thong.
Now we learn that Rob has the number 2 and 4 woman’s lingerie sites on the web.
I’m really beginning to wonder about these guys *wink*
August 9th, 2007 at 9:41 am
24k gold.
August 9th, 2007 at 9:54 am
That’s why I’m taking male yeast infection from the beginning of 2007 (same as two or three months in). That way any spiking from the Thirty Day Challenge won’t move my baseline of 500 (- 550).
August 9th, 2007 at 11:16 am
This is a big Diamond
August 10th, 2007 at 1:25 am
Mike,
You really helped me a lot to catch up in this challenge…
You’re blog is my “precious” toolbar as to Dan’s toolbar to TDC. Don’t worry I still do the core training.
When I succeed in this venture, I owe it all to you, to Ed et.al. and Dan’s “male yeast infection” lol.
August 10th, 2007 at 3:15 am
Hi Mike,
First of all thanks so much for your blog, it helps tremendously!!
I keep trying to install wordtracker, but all I get is the html for it. Got any suggestions?
Any help would be greatly appreciated.
Thanks
August 10th, 2007 at 9:53 am
Dearly Beloved Mike:
Greetings!
That was almost word for word as it was shown in the video, any help on what you use to do this e.g. software or something?
“Absolutely brilliant” is the word that goes for this and…
Thank you very very much for this recap.
Remain blessed and a blessing.
Regards,
Olakunle Solomon Fatoye.
August 10th, 2007 at 3:55 pm
Actually. It’s just me, my fingers, ScribeFire, SnagIt & Flickr.
August 10th, 2007 at 6:03 pm
Dearly Beloved Mike:
I can now see why people call you Mike “Magic” Mindel.
Thank you for the clarifications.
Remain blessed and a blessing.
Regards,
Olakunle Solomon Fatoye.
August 12th, 2007 at 12:41 am
Hi My eyes r out on the end of their stalks,truly amazing.thanks ever so much guys its nice to see there is light at the end of the tunnel.Stanley.
August 18th, 2007 at 7:07 pm
Well I hate to put a dampener on things, but I have run into a problem which doesn’t follow the GTrends/Wordtracker logic.
I signed up for the FULL Wordtracker system, used the UK database and found the search term “golf breaks in cyprus” which supposedly has 75 UK searches a day and only 2 competing sites. However when I drop that keyword phrase into the free Wordtracker keyword/Gtrends tool I get a zero return on searches!
Manually dropping that term into Gtrends and tracking against our familiar male yeast control phrase - produced a pinprick result in the language bar too.
So why would these 2 Wordtracker systems produce such radically different results one from the other - and if the full system is more accurate, how is it that GTrends reports virtually no searches in Google? 60% of Wordtracker’s reported 75 searches should still be a visible number after all.
Any thoughts anybody?
Regards
Adrian
September 2nd, 2007 at 1:17 pm
Interesting. I saw the video but somehow did not catch the whole drift. Read this article and it brought the video home. After seeing the video I started doubting my research. Then I checked my keywords against male yeast infection baseline and they are rating two, four and eights times higher in Google trends. It’s great to have a baseline rather than just a google trend line. Not only is there more bounce in my step but at least there’s a roughly quantifiable rationale behind it.
Thanks for this great post.