Sorry this is not a beginners guide to Yahoo Search Marketing. It presumes understanding of basic search advertising concepts and familiarity with Yahoo’s ad network

Since Yahoo Search Marketing (YSM) moved to the Panama interface, a lot of people are unclear as to how to setup a successful campaign. While YSM is now more like Google Adword, it is still unique with its own quirks.

But here are some important facets of the new YSM engine:

• Like Google, Click Through Rate (CTR) is important. The higher your CTR, the better Quality Score (QS) you’ll get. The better the QS, the more impressions you’ll get with higher positions.
• Unlike Google, QS seems to be calculated one-time (when your campaign first starts). It doesn’t seem like the YSM-bot comes back to check the site ever again. Or if does, it doesn’t seem to impact the QS ever again.
• After campaign startup, QS seems to be almost 100% based on CTR. Good CTR, QS goes up. Bad CTR, QS goes down.

The secret of success on Yahoo seems to be entirely based on setting up your campaign up correctly so that you’ll get good CTR so that QS will go up naturally over time.

Some basic concepts behind my technique:

• The adgroup’s ad must be relevant to your keyword(s). If YSM deems that there is no relevance, that keyword won’t trigger any/many impressions for that ad .

Solution: Grouping (keywords with common words) is critical and mandatory for success. By grouping, you’ll make sure that every keyword in that adgroup has at least one common word. Then it’s as simple as making sure that those specific word(s) occurs in your adcopy. If you can get two (or more) common words, even better for your QS.
Caveat: It’s widely recognized that Keyword Replacement does NOT help ad relevancy scoring. Just because you throw in a {KEYWORD} into your ad, doesn’t mean you’ll get any quality benefit. If you’re grouping keywords, type the specific word(s) into your adcopy in rather than using {KEYWORD}.

• Irrelevant keywords (even with good CTR) will hurt your bottom line

We all know that irrelevant keywords simply cost you money (if they’re clicked on) or lower your CTR (if they’re not clicked on).
Solution: If you spend time to delete irrelevant keywords (during the campaign building phase), you’ll lose less money. AND (definitely) more importantly, you won’t take that initial hit to your CTR.

• Bids affect ad position, which affects CTR.
  • highly competitive keywords will damage your CTR (you’ll get impressions, but they’ll be really bad positions, so you’ll never get any clicks).
  • you can’t detect/delete these highly competitive keywords during the pre-campaign phase (see point above) because you don’t know which ones are competitive until your campaign starts
  • you are unlikely to want to delete them, because the keyword looks legitimate to you and most people will rather write it off as “bad luck” and keep it in the campaign (where it will continue to hurt your CTR)

There is a theory that says you should start bidding at $0.10 for all keywords (and raise them over time). While it “nearly” guarantees immediate profitability, it results in a campaign that cannot sustain itself over the long run. Its lifespan is limited simply because:

As long as those bad keywords exists in your campaign, lowering your CTR and QS. With a low QS, YSM will give you less and less impressions until you give up the niche as unprofitable.
Solution: You need to bid high enough to score a good position for as many keywords as possible.

So now that you understand the premise of the technique, this is the simple step-by-step for implementing it:

1. Group your keywords. Try to group by finding AT LEAST one common word. Then ensure that ALL the common word(s) are in each ad for that adgroup. With this technique, you should only need 1 ad, per adgroup. You can run more ads if you’re doing split testing, but 1 ad should be all you need.
2. Talk to your Affiliate Manager, and find out what the eCPC for your offer is. Set your bid for ALL your keywords to be ONE-THIRD of that eCPC. This should give you enough margin to make some mistakes (and still eek out a profit).
3. Make sure EVERYTHING in the campaign is approved before turning it on. Yes, EVERY adcopy, EVERY keyword.
4. Set your campaign’s daily spending limit to be 20 times your bid setting. So if you bid $0.20, then your spending limit will be $4.00.
5. Start the campaign and come back to it once your daily limit has been reached (probably 1-2 hours for most niches I can think of). Review EVERY adgroup in the campaign (sort by impressions) to see if you have any crazy-high impressions. If they don’t make sense, or are obviously not targeted, delete them straight away.
6. Reset your daily spending limit to something you’re comfortable with. Resume campaign.
7. On a daily basis (until the campaign has been “stabilized”), do the following:
   • If a keyword has low CTR, look at the Average Position. If the position is >10, then decide whether you are willing to bid more. If you won’t increase your bid, then delete that keyword.
   • Obviously if the position is <10, and it still has a CTR, then it’s an un-targeted keyword (delete it) or your adcopy is bad (fix it).

So does this technique really work?

I have a campaign that has:

• 80k keywords
• I got my first 5-out-of-5 bars QS after 1 week. All adgroups were 5-out-of-5 bars within 1 month of the campaign setup
• When it first started, this campaign’s daily charges were just over $100. I now pay over $1k per day, because that’s how much my impressions/clicks have increased.

You decide for yourself

I’ve been really focused on building some solid Google Adwords campaigns recently, and I’ve come to the realization that there is something seriously screwy with Adword’s default “broad match” option. Turns out Google’s definition of “broad” is pretty damn broad!

As an example, a broad match of against “cingular ringtones” will catch:

1. “ringtones” - Apparently Google figures that “cingular” isn’t necessary
2. “razr ringtones” - It seems that Google figures there are alot of Cingular RAZR phones, so “cingular” can be arbitrarily replaced with “razr”
3. “cingular caller tones” - Just because Google believes that “ringtones” is closely related to “caller tones”

This kind of (crazy-super) broad matching sucks because:

1. If you hardcode your ad’s title or if you use Dynamic Keyword Insertion (which will display the triggered keyword “cingular ringtones”) then your ad’s title will say something like “Cingular Ringtones” even though the searcher might have a Verizon phone. RESULT: your Click Through Ratio (CTR) will drop to hell faster than you can delete that keyword.
2. If the searcher was looking for “razr ringtones” (because they have a Verizon RAZR) and your ad comes up with “Cingular Ringtones”, guess where your CTR is headed to?
3. For those that don’t know, Caller Tones is what your CALLER hears when they call you. The Ringtones affiliates offer ringtones _not_ caller tones (which I believe only can be offered by the carriers). RESULT: You’ll get clicks that’ll suck your bank account dry, with no conversions.

And remember in Adwords, CTR is key to your Quality Score (QS). Lower QS will result in lower placements and more expensive bids. So impressions without clicks (lowering your CTR) will hurt your entire campaign.

So how do you overcome these broad match problems?

1. Find those other words that Google thinks are the same, and put them in that adgroup’s Negative Keyword List (NKL).
2. Once again put it in your adgroup’s NKL. In fact, I put a negative keyword against every other adgroup I have. So my cingular adgroup has verizon in the group’s NKL, and my verizon adgroup has cingular in the group’s NKL.
3. Yes, you guessed it. Put “caller tones” in your campaign-level NKL.

In summary, with Adwords, be very careful of your broad matches. Negative keywords are probably more important than your keywords.