How to Win the AdWords Game

August 29, 2005

This is an Ebook from Konstantin Goudkov. Relax and enjoy the reading.

Contents
1. Introduction
2. Terminology
3. The problem
4. Broad matching or exact keyword? - BOTH!
5. CTR and CPC
6. Low cost or high position?
7. Who is above and who is below
8. Ad copy
9. Finding your initial keywords
10. People who bid too high are stupid … or not!
11. If you are getting too much traffic, you are doing something wrong
12. Do not let your ads get down!
13. Negative keywords
14. Don’t be afraid to miss a part of the market
15. Tracking macro-actions
16. Tracking micro-actions
17. Tracking continued - testing the page
18. P. S.
You have permission to distribute this report under any conditions
as long as you do not add, modify, or delete any content or bylines.

1. Introduction
Hi, my name is Konstantin Goudkov.
Currently, I’m an affiliate manager for GenericGifts.com affiliate program. First of all, let me thank for you downloading this book and answer one question you might be asking yourself right now.

What does being an affiliate manager and using Google AdWords have in common?

Well, it is really simple. I needed a way to test new creatives and sales pitches before I gave them to my best affiliates.

After considering multiple options, I came to a conclusion that the fastest and the most practical way to do this would be to use pay-per-click engines.

Furthermore, of all PPC engines out there, I picked Google AdWords for my tests. Here are 4 reasons why:

● You can add/modify your listings almost in real-time (just a few minute delay). With other engines, you have to wait 3 to 5 business days to have your listings approved.

● You can run multiple ads for a single keyword at the same time.

● The ads’ positions are determined by a maximum CPC and CTR, not just by the maximum CPC alone.

● Google with its partners can provide an enormous amount of traffic quickly for any category imaginable.

For the reason #1,
I would like to point out that you do appear on the Google SERPs
sponsored links right away, but you still need to wait a few days to appear on its partners’ sites. I’ll go over the advantages of being shown on the partners’ sites later on.

Here is one more reason (probably the most important one) why I use PPC:

You always need to test your new ads, sales letters, and products with the highest quality leads. An offering that produces poor results with good leads will not produce better results with worse leads.

In other words, it is better to fail fast than to drag a poor offering forever. In my situation, there are only two practical ways to apply this principle to work:

● I could run e-mail campaigns for the past customers (customers who bought more than once and who bought a lot would be even better).
● I could utilize a source of fresh, highly targeted traffic.
One of the most valuable assets of any business is its customer base. Your customer base is not something you want to mess with.
Imagine how high the unsubscribe rate of the list would be if you started running 2-3 test mailings per day for your past customers!
Even if you forget about people who unsubscribe, wasting the time of your customers

– people who trusted you and your offering — by turning them into test subjects is simply not a nice thing to do. It is a sure way to ruin all your future attempts at the back-end marketing with those who did not unsubscribe. That leaves the second-best option, using PPC traffic.

A carefully planned PPC campaign can bring highly targeted leads in a short amount of time. Since I use AdWords for testing, I am on a limited budget. This makes things a lot harder because my tests must be overall profitable. If I waste my budget, I won’t be
able to test any more.

At the same time, the fact that I’m constantly testing implies that 90% of my efforts result in losses. Since only about 10% of my tests turn out to be profitable, they must generate enough money to cover themselves and to cover all failed tests. If unlike me, you are planning on using AdWords to its fullest potential, this report might very well save you a lot of time and money.

This report should not be considered a definitive guide in any way.
It is simply a description of common patterns in successful tests and techniques that I use. To get more information, send an e-mail to bcc-books-adwords@aboutrealstuff.com . Also, visit the message board I set up for topics discussed in this report.

I would be glad to answer your questions and discuss any Suggestions. I thank you once again for downloading this report.

2. Terminology
I use certain terms and abbreviations throughout this report. Below, you can find a brief explanation of each of them. Please note that some basic understanding of the Internet marketing in general and
knowledge of Google AdWords in particular are required in order for the report to make sense.

I include just a short explanation for each one of them, but you should be able to find much more information on the net.

SERP - search engine results page
ROI - return on investment
USP - unique selling proposition
CTR - click-through ratio
CPM - cost per thousand (impressions)
PPC - pay per click

widget - a hypothetical product
conversion ratio - a percentage of leads that perform a desired action (sign-up, sale)
attribute - a specific visual or conceptual element of a page, an ad creative, or a sales
letter (used in fine-grained performance comparison testing)
Also, all targeted keywords are written in an italic style.

widgets- a search term
“widgets” - a phrase match
[widgets] - an exact match
All links in this report are underlined.
link title - external links to outside resources
link title - internal document links

3. The problem
Let’s say you have a direct competitor. That competitor is selling the same product at the same price to the same target audience; therefore, both of you are bidding on the same keywords.

The questions is: How do you make your advertising campaigns perform better than those of your competitor?

I’m going to skip a lengthy discussion on marketing and sales, but instead I will outline a few factors that make a winner in the AdWords game. As you go through the list, you will notice that some statements are just derivatives of previous statements combined with some common sense and backed up by testing.

● Your ads need to rank as high as possible on the list; above your competition, if possible.

● You need to pay as little as possible for each click, while ranking high.
● Your targeting must be as strict and narrow as possible.
● You need good ad copy to attract the “right” audience.
● Your ads must have high CTR.
● You need high conversion ratio.
● Your site must sell better than your competitor’s site.

Keep those statements in mind while reading the rest of the report. Most of the content in the following sections expands on what I’ve just listed.

4. Broad matching or exact keyword? - BOTH!

Many people are afraid to miss a portion of the market. They try to be as general as they can afford when picking keywords. Such approach won’t get you too far unless you have VC’s millions for branding. On a small scale, the best way to compete is with many different specific terms.

All it takes is a few bucks a day!
Here is a basic plan:
● Find a a promising keyword.

● Test and track to maximize your profit for this keyword (think conversion and marketing costs).

● Narrow it down and split into multiple keywords if you have to.

● Make this keyword / ad / sales copy combination produce $10/day of profit after all marketing expenses.

● Wait for the local marketplace for that keyword to settle after your entry.

● Learn and understand what you did and how you did it.

● Rinse and repeat.

That’s right. All you need is $10 per day of profit per keyword.
If you get 10 such keywords, that’s $100/day. With 100 keywords, that’s $1,000/day! So how do you find those keywords?

I’ll reference a few tools that might help you brainstorm in a section called Finding keywords. For now, let’s just talk about the basic idea.
Try to apply the concept of testing page attributes (from the tracking section) to the keyword selection.

You need to find many keywords and put them all in different ad groups of one test campaign. Also, you obviously need different ad copies and tracking / landing URLs for each ad and keyword.

Initially, you should set a low CPC and use broad matching (use negatives to limit some obvious terms). Start testing and tracking once you get some traffic. If you haven’t made any sales yet then you must check your tracking data for micro-actions.

For example, keywords that generate more page-views per unique visitor are more likely to become highproducers once you start optimizing for them. After you have identified a few keywords that show promising results, start working on them.

Allow me to go over one example. Let’s say you have not made a single sale yet, but you found that an ad for the broad matching keyword widgets produces highest number of page-views per unique visitor. Next thing you should do is look at the tracking data for just this keyword widgets. If you find that people who search for red widgets make it deeper into your site, spend more time reading your content, or go farther along your checkout process, you start
working on the narrower term red widgets.

This becomes your new primary keyword. You need to place it into a separate ad group with a separate set of ads and tracking URLs (a different URL for each ad copy). Once you have done that, you need to increase the maximum CPC to get to the position 2 or 3 and let the games begin. From this moment on, you have to keep an
eye on your ROI and start optimizing your ads and your site for that particular keyword.

One of your ads must target the exact match [red widgets] and nothing else. Using exact match will help your ads get higher CTR, and we know what that means, don’t we?

As for the broad and phrase matches, you need to create another ad group for the phrase match “red widgets” and the broad match red widgets, all while using separate tracking URLs.

Increase your CPC a bit. Your maximum CPC should be higher for the phrase match, but not as high as what you have for the exact match. Don’t forget to keep track of the micro-actions. This becomes the mini-version of your original keyword test. But this time, the base for your new test is the term red widgets. Once you identify the winners here, rinse and repeat. You can look at is as branching with one given path of:

“all initial keywords” > “widgets” > “red widgets” > “red widgets sale” (or whatever red widgets leads you to).

You should keep branching until you either get to a point of being profitable and on top of your competition or until it becomes impractical to spend more time on smaller branches.

Keep in mind, these directions will ruin you if you don’t apply some common sense. Many keywords should be promoted without waiting for them to surface in your tracking, while many others should be eliminated without even trying.

There are circumstances when pushing broad matching alone proves to be more profitable, but don’t forget to test and track your advertising efforts so you can base your decisions on a solid data and not on your gut feeling.

5. CTR and CPC
Once you employ testing and tracking principles that I describe in the Tracking macro-actions section, you should be able compete with your conversion ratio. In other words, you will be able to pay more per each click and rank higher on competitive terms, while still making profit.

For now, let’s talk about another factor that affects the position of your ads: clickthrough ratio. Guys at Google have figured out a smart way to sell their inventory. They only care about cost per impression, be it CPM or any other measurement. In other words, they don’t care about your clicks, yet they want to get the most buck for the served search queries and display relevant offerings.

It might sound a bit counter-intuitive, but bear with me for a minute. I leave their AdSense service out of this report. You can start by disabling content sites in your AdWords account and later on figure out if you want to advertise on those sites or not. Anyway, let’s get back to the deal with CPC and CTR. Let’s take two ads, one with 5% CTR (ad A) and another one with 1% CTR (ad B).

If the ad A gets 5 clicks per 100 impressions and the ad B gets only one click, Google would charge the owner of the ad B five times more for the same ad position. Let’s look at the numbers. If the owner of the ad A sets maximum CPC to 20 cents then Google would make $1
for each 100 impressions (more or less).

100 impressions * 5% CTR = 5 clicks * 20 cents = $1

At the same time, Google would have to change the owner of the ad B one dollar per click just to make the same amount of money.
100 impressions * 1% CTR = 1 click * $1 = $1

It comes down to the same buck for Google per each 100 queries (impressions). So if you are the owner of the ad A, you only need to bid 21 cents to get ahead of the ad B.

Does it start to make sense? Sounds good in theory?

In reality, Google does seem to apply a damping factor to its calculations. I have no proof of that, just a gut feeling. So if your CTR is 5 times higher than that of some other ad, your CPC for the same ad position would not be 5 times lower. Still, even if it’s just 2 times lower you win big!

Combine lower cost for each click with a higher conversion ratio and you become unbeatable! Keep on reading to find out how to get and maintain high CTR for your ads.

6. Low cost or high position?
Two words: impulse buying. When calculating your ROI, don’t assume your conversion ratio will stay the same once the position of your ad has changed. Don’t even assume that your average order
size will stay the same. Why would you get an order if your ad is nowhere near the top? Here are a few reasons that come to mind:

1. Your sales pitch and the site layout made the visitor buy.
2. All sites above yours did not offer what the visitor wanted.
3. Other ads did not appeal to the visitor and did not grab his attention.
4. The visitor checked other sites, but your prices were lower.
Well, #1 sounds warm, fuzzy and feeds your ego, but don’t kid yourself.

There is no way a visitor could have read your perfect sales pitch before he actually clicked on that ad down there, in the no man’s land. All your optimization efforts have minimal effect if you ad is not among the top few.

What about #2?

Well, in that case, either the world is full of incompetent marketers or your visitor did not use the anticipated keywords. You should start testing as described in the Broad matching or exact keyword? - BOTH! section. There is a chance that the keyword is a
high-producer waiting to be discovered.

If it turns out that the order was an isolated occurrence then don’t waste your time with this keyword any longer. You can either take it off or leave it there and hope for more luck again.

Thousands of bottom-feeding keywords that seem to produce overall profit are nothing more than a bunch of ants carrying a piece of food together. While it might seem like they are working together on a common goal, it has been scientifically established that ants just pull the food each in their own direction. What seems to be a result of a great collaboration is actually nothing more than a few ants on one end of the piece being a bit stronger than a few ants on the other end.

What would happen if you removed the ants off the wrong side of the piece of food?

- The piece itself would move in the right direction must faster.
What would happen if you removed half of your deadbeat keywords and spent that
money on higher maximum CPC for the other half of your keywords?
- You would make more profit in a shorter period of time.

Reason #3?
Well, congratulations. If that’s the case, what is your ad doing down there?

If you can’t raise maximum CPC because your ROI would become negative then refine your keywords. Assuming that your site is optimized, there is no other logical solution.

And the reason #4.

If you beat your competition with price and you got a visitor with an ad at a low position — congratulations!
You just got yourself one of those price-conscious, value-oriented, hard-core comparison shoppers. Good luck with your future back-end marketing efforts as well as attempts to increase prices to make some real profit next time. I don’t see anything wrong with using a bunch of bottom-feeding keywords, just don’t let them stay that way. Try to grow and convert them to high-producing keywords that are worth the time you spend.

Don’t forget on more important thing. Having your ads displayed among the top few makes you highly visible on Google partners’ sites. So not only does your reach increase dramatically, but the quality of traffic changes as well.

For example, AOL has much more consumer-oriented audience while the Google site is still utilized more by people who are looking for information.

7. Who is above and who is below
You should always be aware of your ad’s neighbors. Study them and take the information you gather into consideration while writing your ad text or your sales copy. Also, keep them in mind while brainstorming on the unique selling point you want to present for this particular keyword.

For example, if the ads around you point to competitors’ sites with similar offers and higher prices then you should press on the price difference. You need to let your visitors know that you offer lower price for the same exact product.

If your neighbors offer slightly different products then you should push the advantages of your particular offering. Your main selling point is to persuade visitors that your product is right for them.
In case you are surrounded by similar products offered at the same or lower prices, try to sell yourself and not so much the product. You should put some testimonials around and try to win the trust of your visitors. Your best bet is to stress the comfort and safety of dealing with you. Instead of talking about the product’s features, you should point out your fair return policy, your strict privacy statement, and your fast shipping methods. You need to make those your primary benefits.

In any case, make sure your ads and your sales copy match each other so that the perception of your offer and your company is consistent throughout the whole buying experience.

While doing so, make sure to stand out of the crowd that surrounds you. One more thing to note about your neighbors:

When do you absolutely have to beat some other ad in ranking?

- When you don’t have a unique selling point, your prices are higher, and the ad above you offers the same exact products.

In all other cases, don’t rush and test before you make a move on someone’s position.

8. Ad copy
Many smart marketers say that title is the most important element of your campaign The choice of titles can either keep you in business or drive you out of it, be is a sales pitch header, an e-mail subject line, or an AdWords ad.

There are a lot of great books about producing effective titles. I am not going to cover it in great detail. Also, I will not stress the importance of testing here. It should be obvious. The AdWords ad is just a title with a description which acts more like a subheader.
What differentiates PPC ads from almost any other medium is that you pay per click.

With an e-mail campaign, you need to get as much attention as you can. The same goes for a banner ad paid for on the CPM basis. It holds true for an ad placed in a leased advertising spot for a predefined time period as well. But with PPC, you don’t just want any attention. You want the attention of the right audience.

What further complicates things with AdWords is the fact that you need to balance the amount of attention you generate (which affects the CTR of your ad) and the degree of targeting you apply. If too few people become interested in clicking on your ad, your CTR will go down and you will start paying more for each of those clicks.
On the other hand, if you generate too many clicks, but the visitors are not really looking to buy, you are simply wasting your money.
With AdWords you have two somewhat independent means of controlling the performance of your marketing campaign: the copy of your ad and the keywords you pick.

Even though it seems intuitive enough to consider both of those factors in the same context, most people still seem to concentrate on one more than the other. Some people consider both of them, but not at the same time. What you really need to do is match a title that would appeal only to your most targeted audience (while filtering out related niches) with keywords that would most
likely be used by that same targeted audience.
The better you match, the better your advertising campaigns perform.

The easiest, although somewhat backward-looking, way to balance these two factors is to produce a strict ad copy that only appeals to your most targeted audience. After that, you need to narrow your keyword down to raise the CTR for the ad as high as you can. You might also have to split the ad in two, three, or more separate ads as a result of your efforts to increase CTR.

If you target the keyword widgets, and set your title to “Widgets”, you would get high CTR (assuming you are among the top few ads), but your conversion ratio would be somewhat low. You could increase your conversion by changing your title to “Buy Widgets”, but that would lower your CTR and you would start paying more for each click. Two things would happen:

● Half of the people who are only looking for information about widgets would go past your ad, which would effectively lower your CTR.

● The other half of the information seekers would still click on your ad out of curiosity, which would cause you to simply lose money.

What if you changed your keyword to buy widgets?

That would take care of the CTR and the conversion ratio, but would drop the amount of traffic you receive. So what?
Find 10 keywords (and 10 ad copies) for: buy widgets, cheap widgets, etc…

The higher CTR and conversion will let you occupy higher ad positions and that would compensate for the market you’ve lost by not advertising for simply widgets. If 10 keywords are not enough, find 100! In the end, having 100 highly targeted ads for targeted keywords is much better than having one general ad for a general
keyword.

Use the method I described in the Broad matching or exact keyword? - BOTH! section
to find those keywords. And by all means, keep running a test campaign for widgets to discover new keywords, but don’t make it your primary campaign! Well, that was easy.

Now a slightly harder problem. What if you have reached the strictest reasonable degree of targeting you can apply to your keywords, but those keywords still seem to be too general?
Let’s say you sell sliver widgets for $200+, and you are aware of a huge market for cheap $20 sliver widgets. It would be reasonable to assume that a large portion of people who search for silver widgets are looking for those $20 widgets and would not be interested in buying from you.

Your target keyword is silver widgets. You cannot target anything more specific, like silver widgets for $200, because that particular keyword would only bring you literally 2 clicks per year. That’s just ridiculous. Well, go back and do some testing. You’ll find a ton of keywords like quality silver widgets, luxury silver widgets, expensive silver widgets, etc. Make them your primary

keywords and dominate that marketplace. What if that’s simply not enough and you absolutely must target silver widgets?
Well, the first thing you need to do is change the copy of your landing page for that keyword. It does not have to be the same page you use for any other campaign. In your copy, acknowledge the existence of the cheaper market and explain the benefits of paying more. I’ll just leave it at that. The art of writing persuasive copy is
way out of the scope of this report. Just do your best on your side before proceeding any further. Also, don’t forget about your ad’s neighbors while brainstorming about the copy.

Once you are done adjusting your website copy for that keyword, here comes the tricky part. Many people will tell you to put something like “…starting at $200″ or some other negative qualifier in your ad copy. They will tell you to do that and hope for the best.
Well, that’s nice, but that will lower your CTR and you will pay more for each click. Don’t just limit your traffic and hope for the best, but instead test both ads. Check if your website copy can convince enough of the general widget buyers to go for quality.
Nobody, including yourself, knows in advance which ad copy will affect your bottom line better:

● being more specific in your ad text, having lower CTR, and as a result paying more for each click of presumably more targeted traffic or

● knowingly attracting a portion of leads somewhat outside of your core target audience, having higher CTR, and paying less for each of those less targeted clicks So we are back to the importance of having good tracking, that’s all.

You would be surprised to see how often negative qualifiers in the ad copy actually decrease ROI (I’m talking only about AdWords advertising here). Don’t leave anything up for chance. You need to test and track no matter how far, deep, or targeted you’ve gotten.
Once you have established which is the better way to run that campaign for that particular keyword, move on to the next one. All you have to do is be more productive (even by a tiny bit) than your competition. The fact that you are testing and not guessing will help you be more productive, especially if your competition is just guessing.