Keyword Research : Go Deep or Going Wide First?

October 15th, 2007 by Amit

You know, a lot of you out there think if you have good keyword list then it’s just a matter of :

  1. Setting up tightly targeted adgroups
  2. Having effective Google ads that have the keyword in the headline, benefit in the description, & a strong call to action
  3. A landing page that converts

then you’re set, and ready to make windfall profits in your niche.

Right?

Wrong!

There’s one very common mistake that stops a lot of affiliates dead in their tracks.

I’ve helped and coached a lot of affiliates and I’ve seen this mistake come up again and again, and chances are you’re probably doing it too. I know I made this very mistake all the time when I got started.

Before I explain exactly what this often fatal mistake is that so many newbie ppc affiliates make, let me lay the ground work…

I want you to think hard about this statement:

When you first launch your campaign, the breadth of your keyword list is MUCH more important than the depth.

What in the world do I mean by that?

A broad keyword list is one that has lots of related and lateral keywords, and not just variations of one root keyword.

Here’s a simple example of a list of keywords that has breadth :

delete adware
delete backdoor
delete browser hijacker
delete cookie
delete Dialer
delete keylogger
delete malware
delete spyware
delete toolbar
delete trojan
delete w32
delete worm
destroy adware
destroy backdoor
destroy browser hijacker
destroy cookie
destroy Dialer
destroy keylogger
destroy malware
destroy spyware
destroy toolbar
destroy trojan
destroy w32
destroy worm
find adware
find backdoor
find browser hijacker
find cookie
find Dialer
find keylogger
find malware
find spyware
find toolbar
find trojan
find w32
find worm
remove adware
remove backdoor
remove browser hijacker
remove cookie
remove Dialer
remove keylogger
remove malware
remove spyware
remove toolbar
remove trojan
remove w32
remove worm
Now you can see these (trojan, worm, w32, etc) are all different types of spyware.

Now here’s an example of another keyword list that has depth:

anti spyware
anti spyware free
anti spyware software
antivirus spyware
best free spyware remover
best spyware
download free spyware
download spyware blaster
free anti spyware
free kazaa spyware remover
free removal of spyware
free spyware download
free spyware downloads
free spyware protection
free spyware removal
free spyware removal programs
free spyware removal software
free spyware remover
free spyware scan
mcafee anti spyware
microsoft anti spyware
remove spyware
spyware blaster
spyware blocker
spyware cleaner
spyware detector
spyware doctor
spyware download
spyware programs
spyware protection
spyware removal
spyware remover
spyware scan
spyware software
spyware terminator
spyware virus protection
totally free spyware removal

You can see that the above terms are related to removing or detecting spyware. And all the above keywords have the root word ’spyware’ in them. This is a keyword list with depth.

Now if you want a lot of depth you can dig up long tails, but I’ll get in a later post!

So what’s the fatal error that a lot of ppc affiliates are making, even if they have a excellent keyword list, that is one with both breadth and depth??

Most affiliate focus on going deep (deep means : launching one root keyword with lots of variations, such as the second spyware list) FIRST, before going wide ( wide means: launching broad keywords, ie the first keyword list).

Here’s problem with going deep first:

TOO NARROW A FOCUS ON THE MOST COMPETITIVE KEYWORDS

Most affiliates find every freaking variation and combination with spyware (or whatever the main keyword for that market is) in it until their blue in the face, setup their Google campaign, and launch.

Then they’re surprised when they hardly get any traffic bidding 0.20 (spyware is a high cost per click keyword), and their conversions suck since every affiliate is also bidding on spyware.

Frustrated they pause their campaign after a few weeks and move onto something else, without even attempting to try a bidding on : trojan, w32, etc, etc, or any of the other broader keywords.

They figure if spyware keywords don’t convert for them, it’s hopeless! Besides they lost $500 on the campaign, and don’t want to mess with it anymore.

Sound Familiar?

MISSING OUT ON THE MOST THE TRAFFIC & PROFITABLE NICHES

Even if you managed to get the spyware keywords to convert, with a sharp professional looking review site that compares every possible benefit and feature of 3 different spyware removers, you’re still may be missing out on the most profitable niches in the market.

What’s worse, in many markets a broad keyword list will deliver WAY more traffic than a deep one that focus on the main most competitive keyword.

This is why it’s so important to go wide FIRST before you go deep, it will assure that you get the maximum amount of traffic and allow you to identify the most profitable sub niches in the market you’re in.

Once you go wide and see which sub niche of keywords (malware, keylogger, worm ,etc) are producing the most traffic and conversions then focus in on those keywords and start going deep.

So if you find (just an example), that trojan keywords are getting a lot of traffic at a bid price where your Cost/Conv is low, then focus in and start going deep. That is, start finding as many variations and combinations of trojan keywords that you can find.

By doing this you’re now employing the 20/80 principle and focusing your energy where you’re getting the highest return. :)

So here’s EXACTLY what you need to do when launching your affiliate offer:

  1. Go wide, as wide as you can to, spend some time identify every related and lateral keyword in your market.
  2. Once you launch this broad set of keywords, your campaign will be getting traffic from all the major sub niches in your market. Tip: To get a good Google QS (Quality score) you may want to setup a separate page for each sub niche (ie a page for trojan, a page for w32, a page for browser hijacker etc)
  3. Let your campaign run and identify the sub niches where you’re getting the majority of traffic and sales. Focus in on those top sub niches and go deep.

This is how you build a ppc affiliate campaign with a solid foundation that will bring in profits for months, even years to come!

Posted in Keyword Research, PPC Marketing, Super Affiliate Mindset |

49 Responses

  1. Response by:  Andy on October 15th, 2007 at 8:54 pm

    MyAvatars 0.2

    Amit,

    You posted this just in the nick of time for me. I was just going to start a new campaign with what you are referring to as a deep keyword list.

    My only question is, do you create a separate adgroup for each one of those keyword phrases, or do you use dynamic keywords for testing?

    Thanks,

    Andy

  2. Response by:  Adnan on October 15th, 2007 at 9:22 pm

    MyAvatars 0.2

    Amit,
    Wow! I am not an affiliate marketer yet and have not researched too many other blogs on the subject, but I cannot imagine anyone else giving away this kind of trade secret for free. You probably saved a lot of marketers-to-be from making mistakes that could be fatal to their AM careers.

    I can even use the same philosophy on my blog’s keyword cloud to attempt to drive traffic. Keep up the great work; I’m hooked on this blog.

  3. Response by:  David on October 15th, 2007 at 10:53 pm

    MyAvatars 0.2

    Rock solid information. Thanks for a great post!

  4. Response by:  darren on October 15th, 2007 at 11:13 pm

    MyAvatars 0.2

    Thanks for the information Amit - your posts are awesome. I was wondering if you could tell me which software tools you use to find lateral keywords.

  5. Response by:  REC on October 15th, 2007 at 11:15 pm

    MyAvatars 0.2

    What about mis-spellings and QWERTY typos. Is that stuff of any use? Sometimes I wonder since Google does correct your spelling.
    So we’re going Broad and then “peel and stick”?

  6. Response by:  Brent on October 15th, 2007 at 11:25 pm

    MyAvatars 0.2

    Excellent info Amit. Do you find it important to revist the wide approach every once in a while? Like say wide - deep - then back to wide again to catch any sub-niches that have become hot since your initial wide approach?

  7. Response by:  Joe on October 15th, 2007 at 11:45 pm

    MyAvatars 0.2

    I think you mean “breadth” instead of “breath”.

  8. Response by:  Stockshaker on October 16th, 2007 at 1:14 am

    MyAvatars 0.2

    Amit & Community,

    BUT how to do you go about finding the keywords that fit in group 1? Is there a tool? Can’t you do this thru adwords keyword suggestion tool? Or is wordtracker the best bet?

  9. Response by:  AnotherAmit on October 16th, 2007 at 2:24 am

    MyAvatars 0.2

    StockShaker:
    I think you can use both tools. I feel that breadth is something that you instigate using these tools.

    Everyone:
    1)Please correct me if I am wrong in my suggestions

    2)How you are you guys tracking cost/conv while marketing affiliate programs? are you asking the Seller to include your google tracking code on their thank you page? Or are there any other ways you are doing this?

    cheers,
    AnotherAmit

  10. Response by:  Hadi on October 16th, 2007 at 5:50 am

    MyAvatars 0.2

    Great post Amit.

    I made exactly the mistake you mention of ‘going deep’ with my first ever campaign last week, with limited success.

    I figured my market was so niche that broadening the keywords would bring irrelevant traffic.

    Presumably, you have to run a pretty high cost campaign to experiment with a ‘going wide’ approach i.e. you’re going to get lots of irrelevant impressions and a poorer overall CTR and hence lower quality score?

    I managed a CTR of about 65% at one point with a deep/narrow approach and didn’t have to spend the earth to get the traffic - my biggest problem was a poor conversion rate from my landing page.

    I think it will be worth trying a ‘go wide’ approach, so I’ll be doing that today!

    Thanks again

    Hadi

  11. Response by:  Vijay Teach Me $$ on October 16th, 2007 at 8:04 am

    MyAvatars 0.2

    First blog post to read in the morning and wham…. it hit like a 4×4 in the head…

    Thank you Amit, no wonder I have you as my favourite teacher.

    Regards
    Vijay

  12. Response by:  Doug Woodall on October 16th, 2007 at 8:40 am

    MyAvatars 0.2

    Hi,
    I came across your site thru a Google alert for the keyword Spyware.
    Ive found that the keyword that gets the most hits depends on what particular Spyware or Virus is ravaging the Internet at that time.
    But your plan is a good one. You really have to focus, failure to act in the business ecosystem of the Internet can mean lost sales.
    You have a great Blog here.

  13. Response by:  fields on October 16th, 2007 at 8:57 am

    MyAvatars 0.2

    Thanks Amit,

    I think I need to read and reread this article until it sinks in .. :-)

  14. Response by:  Jay R on October 16th, 2007 at 9:00 am

    MyAvatars 0.2

    By “breath” do you mean “breadth”? http://www.m-w.com/dictionary/breadth

  15. Response by:  Amit on October 16th, 2007 at 9:35 am

    MyAvatars 0.2

    Hi Jay,

    Thanks for pointing this out! :)

  16. Response by:  Amit on October 16th, 2007 at 9:49 am

    MyAvatars 0.2

    Hi Darren,

    To find lateral and related keywords & not just variations of the root I recommend the following tool: Wordtracker, Google Keywords Tool, Keycompete

  17. Response by:  Amit on October 16th, 2007 at 9:52 am

    MyAvatars 0.2

    Hi REC,

    You can actually build a whole campaign out of misspelling and typos. However, I’ll get into that in another post.

    I would focus on finding misspelling for the most high traffic terms, once you figure out what those are by going wide and collecting data on your campaign. Use the Wordtracker misspelling tool to find them.

  18. Response by:  chris risenhoover on October 16th, 2007 at 10:11 am

    MyAvatars 0.2

    Amit may have meant breadth, but breath may be better! If your campaign is not wide enough, your affiliate baby will not have enough air in its lungs to survive! Good post. The key to doing this effectively IMHO is making sure to create dynamic pages that pull the keyword in so that you can go wide without spending all of your time creating landing pages in the beginning. thank dynamic content as well (rss feeds, latest news, etc) that you can put in your landing page that will help the spider identify what you are about.

  19. Response by:  Tim on October 16th, 2007 at 10:32 am

    MyAvatars 0.2

    Very interesting.

    Also, many thanks Amit (and Andrew) for the fantastic “bonus”. That was worth the price of the course alone.

  20. Response by:  Walter on October 16th, 2007 at 7:45 pm

    MyAvatars 0.2

    Ooh. Reality adjustment. Excellent post, Amit! But how wide is wide enough? How big a keyword list would you start with? Would you start small, then expand on the most promising ones, or would you cast a huge net and then toss out anything that didn’t convert well?

  21. Response by:  moves on October 16th, 2007 at 7:54 pm

    MyAvatars 0.2

    I’m very close to launching an Adwords campaign in a competetive niche so this info is a huge eye opener for me. Thanks a lot Amit! Your blog is number 1 on my list.

  22. Response by:  Clyde Fernandes on October 17th, 2007 at 4:38 am

    MyAvatars 0.2

    I have given up my job and plan to get into ppc part time, will be reading a lot more of your articles. I know a lot on affiliate marketing but not much on PPC.

    You will be seeing a lot more of me here :) Thanks for this article, I will be able to make a much more confident start!

    Cheers Amit!

  23. Response by:  susan on October 17th, 2007 at 3:44 pm

    MyAvatars 0.2

    amit,

    where did you meet jerry palmer?

  24. Response by:  David on October 17th, 2007 at 11:18 pm

    MyAvatars 0.2

    Amit, how many keywords would you normally start out a campaign with? I know, I know..’it depends’. But generally speaking 10, 100, 1000, 10,000, etc? :)

  25. Response by:  Vanessa on October 18th, 2007 at 2:32 pm

    MyAvatars 0.2

    Hi, Amit:

    I’m curious about how broad you recommend going with your keywords before you find yourself in another niche…

    For example, I’m promoting a cellulite-reduction product. Not surprisingly, many of the *deep* keywords have a high cost per click and lots of competition. So in my keyword research, do I explore diet-related keywords, fat loss, etc.? Since they’re not necessarily “sub-niches” of cellulite reduction, are they too off-topic?

  26. Response by:  Amit on October 21st, 2007 at 11:19 pm

    MyAvatars 0.2

    Hi Venessa,

    I would definitely try bidding on keywords like fat loss, and the like. For a high CPC market like that it’s bidding on those lateral keywords that’s the key to profitability.

  27. Response by:  Amit on October 21st, 2007 at 11:20 pm

    MyAvatars 0.2

    Hi David,

    I general start a campaign with several thousand keywords. Depending on the size of the niche it can be ten of thousands of keywords.

  28. Response by:  Amit on October 21st, 2007 at 11:22 pm

    MyAvatars 0.2

    Hi Clyde,

    You’re in for an exciting adventure! PPC is the coolest game in town, in my opinion anyway. ;)

  29. Response by:  Amit on October 21st, 2007 at 11:23 pm

    MyAvatars 0.2

    Hi Walter,

    Cast a huge net of broad keywords, and launch then ALL at ONCE. Then throw out what doesn’t convert.

  30. Response by:  Kang on October 30th, 2007 at 9:25 am

    MyAvatars 0.2

    Hi Amit,

    Excellent tips!

    Very valuable information you have here
    on your blog. You’ve got yourself a
    regular :)

  31. Response by:  o2smedia.com on November 21st, 2007 at 5:55 pm

    MyAvatars 0.2

    Great article!….Im tempted to copy it on to my site for users..if i let everyone know you wrote it are you cool with me posting it?

  32. Response by:  Rex Bush on December 25th, 2007 at 9:29 pm

    MyAvatars 0.2

    I just started my first PPC campaign and made this very mistake of going deep before wide. Thanks for the data! It opened my eyes.

  33. Response by:  Richy Roo on January 28th, 2008 at 12:20 pm

    MyAvatars 0.2

    I think you have to be careful with what is being said here you need to establish both when creating a camapign, it’s all about consumer experience. With having “breadth campaign” you leaving you self open for a lot of unrealetted traffic which will cost you alot of money. If you run somethng like a search query report in Google Adwords you will soon establish which keywords are working well for you. Remember who you customer base is and what they are likely to type in inoder to find your site and product. There are some valid points but I don’t think this is “THE” way to do a PPC camapign.

  34. Response by:  Tyler Dewitt on February 4th, 2008 at 6:55 pm

    MyAvatars 0.2

    Amit,

    Want to tell you this was a great post now in the beginning stages of the launce I would imagine your going to lose money in a way well your not going to start profiting right away i should say.

    From what I visiualize in my head what your saying is to go deep with more competitive keyterms therefore you can find out what is converting and plus build your QS score up for that particular vertical keyword and then build lateral terms around it such as variations?

    The reason I said your not really profiting and most likely losing money in the beginning stages is because your building your QS up with your CTR and other things?

  35. Response by:  Steve on February 12th, 2008 at 4:41 am

    MyAvatars 0.2

    Amit,

    First off, thanks for all your tremenddous help. In reading your posts and watching your presentation at Affiliate Summit I have a question.

    You speak regularly of changing and managing campaigns with ten of thousands of keywords. Are you just setting up multiple Adwords accounts? Because it seems that you would be getting close to the account limitations in reagards to keywords/adgroups. I only ask because I was told by Google Adwords support that having multiple Adwords accounts is “not recommended” which I took as being frowned upon. Thanks in advance for your time.

  36. Response by:  Tyler DeWitt on February 12th, 2008 at 11:42 am

    MyAvatars 0.2

    Steve,

    I was told the same thing and I’m kind of disappointed in the support today its like they was giving me trouble about increasing my keyword limit to the max of 1 million per account and then they talked to me like I didn’t know what I was talking about which they said they was going to take care of it and send me an e-mail when the keyword limit i updated…

    but yes Amit I had the same question, Do you run a million keyword account and what do you think of the adwords support maybe its just me taking them the wrong way and there just trying to do there “job” by recommending other techniques, but I don’t really think there reps know what there talking about when it comes to building large keyword campaign

  37. Response by:  Tyler DeWitt on February 12th, 2008 at 11:48 am

    MyAvatars 0.2

    I’m not going to say what keywords I’m using, but there very relevant lateral keywords and I’m up to 4.4 million

  38. Response by:  Steve on February 12th, 2008 at 2:08 pm

    MyAvatars 0.2

    Thanks for the comments Tyler. Adwords support gave me a number of 50,000 keywords per account as a maximum. Which is making it very hard for me to grasp. Especially if you are exploring the recommendations of trying tons and tons of longtail keywords. I just don’t get it.

  39. Response by:  Tyler DeWitt on February 12th, 2008 at 2:16 pm

    MyAvatars 0.2

    Steve,

    Yes I agree I think my campaign will do good all my quality scores have came up just by spending, but same issue here adwords is limiting on my number of keywords, but I think I got them to uncap 2 accounts at 2 million keywords.

    I would also keep in mind there just Google Ad Reps and nothing more I had to kind of explain to them what I was doing and then they was like that was a good idea the guy I was talking to was treating me like I was a moron or something, but then again I guess he don’t know who I am…

    But yes I’m trying to find out how some of these sites are organizing there keywords rather its through multiple accounts or what because I know of what that got to b eusing at least 10 million keywords and I can’t even get my adwords editor to upload 1 million it freezes up or returns post can not be sent or some error message I don’t know…

    But yea I’m trying to get an idea on how some of these other affiliate sites are managing there keywords rather it being through a ton of accounts like 200 accounts or if they just get there keyword limit unlocked

  40. Response by:  Tyler DeWitt on February 12th, 2008 at 2:18 pm

    MyAvatars 0.2

    sorry about the typo when I said

    I know of what that got to b eusing at least 10 million keywords

    I mean I know of one site that is using at least 10 million keywords

  41. Response by:  Steve on February 12th, 2008 at 2:28 pm

    MyAvatars 0.2

    Ok, so there is a way to get your account limits lifted. Well that makes sense now. I guess I need to contact them again. It sort of bothers me though that they have told me both times that there is no way to increase your keyword limitation above 50,000.

  42. Response by:  Tyler DeWitt on February 12th, 2008 at 4:00 pm

    MyAvatars 0.2

    Steve,

    Yes I know you just have to give them a good reason to increase your account try MCC and then create more accounts to be honest with you I think it probably better to have 50k keywords and numerous accounts or something like that because the adwords interface slows down big time when your running a lot of keywords and for some odd reason I can never get the adwords editor to upload the keywords properly its like slow, freezes up, or return a post error message when trying to upload that many keywords…

    I guess we can wait and see what Amit suggestions on this would be because I’m just as curious as you are on how they manage these keywords.

  43. Response by:  Steve on February 13th, 2008 at 2:47 pm

    MyAvatars 0.2

    Thanks Tyler, I will give the MCC a try. And wait anxiously to hear Amit’s response. Cheers!

  44. Response by:  Steve on February 16th, 2008 at 4:04 am

    MyAvatars 0.2

    Does Amit usually make his way through these old threads???

  45. Response by:  Tyler DeWitt on February 16th, 2008 at 10:13 am

    MyAvatars 0.2

    Steve,

    No hes not very good at replying to his comments hes probably pretty busy, but from my research I would imagine to use less keywords that convert the reason I say that for is because from my research it seems that less keywords are better to use that actually bring traffic and convert maybe a 200-2000 keywords.

    I do know I will start promoting through organic search at the same time I’m promoting through PPC the reason being for is because I’m pretty good with organic search were in the top 10 for website design I have a few sites I’m going to rank for web hosting :), but this PPC stuff driving me up the wall :).

  46. Response by:  Joshua on June 9th, 2008 at 4:09 pm

    MyAvatars 0.2

    What is a good tool for longtail key generation? Anybody have a good one they prefer?

  47. Response by:  c.m.Raja sekaran on August 14th, 2008 at 11:01 am

    MyAvatars 0.2

    dear amit,
    wonderful information

  48. Response by:  money making job in home on September 19th, 2008 at 6:51 pm

    MyAvatars 0.2

    sweet, beautiful and very useful. thanks.

  49. Response by:  olatunde on November 1st, 2008 at 3:58 am

    MyAvatars 0.2

    What are lateral keywords? And point by Hadi, that;

    “Presumably, you have to run a pretty high cost campaign to experiment with a ‘going wide’ approach i.e. you’re going to get lots of irrelevant impressions and a poorer overall CTR and hence lower quality score?”

    What do you think is the solution to this, Amit? It’s going to be a ‘pretty high cost campaign’, is there anything one can do to bring the cost down while achieving a high quality score?

    Thanks,

    Olatunde
    http://www.ecoversalive.com


Leave a Comment




Please note: Comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.