How to start tuning your campaigns with keywords and bids

Always double-check your maximum daily budget first and be certain that it is still set at an amount you are willing and able to risk. Also, be aware that for some high-volume keywords on some search engines, your daily budget limitation might be reached so quickly that it is exceeded before the ad can be shut off. I have had campaigns spend several times my daily budget within an hour or two on more than just one occasion, so if you are bidding on a keyword with obvious volume potential, you need to be especially cautious.

Tuning Your Keywords

You have already covered the keyword selection process, how and when to group them into ad groups, and when to use keyword phrases, exact match, broad match, negative match, and the like. The key now is to simply revisit this same process. Can you think of any keywords you missed? Negative matches? Maybe Paramount Pictures has decided to release a movie remake of The Munsters, and your Munster Jobs campaign should now include some negative keywords to weed out those interested only in the upcoming movie and not in the job site you are advertising.

Maybe you change some of your broad matches to phrases or exact match. Make any changes you feel might be necessary, but do not make changes without any reason just to see what happens. If you went through the original keyword selection process carefully and thoughtfully, this area will likely need the fewest changes when tuning.

Tuning Your Bids

Obviously, the higher you bid, the more often your ad will run and the higher your position will be, which will result in more clicks (budget allowing). You will also see higher costs and lower returns per click. The lower you bid, the less often your ads will run and the lower your ad position will be, but your costs will go down and your RPC will go up. So the question you should ask is, do you need higher volume or higher margin (RPC)?

Your problem is clearly your plummeting click volume. In this case, you would raise your maximum CPC bid, probably by at least the 5 cents per click you seem to be saving already in your actual CPC. Then you would continue to monitor your total return and continue to raise your maximum bid in increments until you find the bid that gives you the highest total return.

If your volume is fine, but your margin (RPC) is very low, then you head in the opposite direction. You would lower your maximum bid incrementally while still monitoring the total return to determine the most profitable bid.

How much you should raise or lower your bids each time is another question. I cannot provide you with an exact answer or a simple formula, but there are a couple of things to keep in mind.

Certainly, you should never raise your maximum bid much higher than your earnings per click (EPC), as this could result in your campaign not just losing money per click, but also in generating more clicks (potentially a lot more clicks) and losing more money. And, if you raise or lower your bids only in tiny increments, this process could be agonizingly long. Outside of these two pieces of advice, you are left to your instincts. But be aware, the more aggressively you proceed through this process of tuning your bids, the more closely you should monitor the results.

Tuning Your Ad Text

I said it before and I say it again: Ad copy is king. In performancebased paid search algorithms such as Google, MSN, and (soon) Overture (i.e., Yahoo!), a good piece of ad copy can lower your costs and increase your volume, so be sure to invest some time in testing your successful ads against new ones.

I have already discussed how to write good ad copy and things you should consider when putting an ad together, so your focus now should be on implementing what you have learned: writing multiple ads, comparing them against each other, and sorting out the great from the merely good.

First, you should never edit an existing successful ad that is, an ad that has generated a substantial number of clicks with a positive EPC unless you have a more successful ad already in place. I have been told by Google that ad history is a factor in its paid search algorithm, and I am sure Google is not the only one doing it. Always create a new ad, even if it will have only minor changes from the original, and run these side by side. You must be sure to run these ads long enough to get a statistically significant sample.

This is not a text on business statistics, so if you are not sure what that means, just wait for two days or until you have 200 clicks, whichever comes first. Do not run more than two or three ads against each other at once. If you have half a dozen different ad ideas you want to try, run the first two against the existing ad, and whichever of the two challengers appears the weakest can be replaced with one of the ads you haven’t tried. Do not replace your original ad, even if it is underperforming the new ad(s), until the more successful ad has built up a strong history.

This history does not necessarily need to include more total clicks than the old ad in some cases you may have spent years building up the ad history for the old ad but the new ad should generate a significant number of clicks, operate for at least one full performance period, and the all-time CTR for the new ad should be at least as high as the all-time CTR of the old one.

Your primary metric for comparing these ads will be the click-through rate (CTR) of each ad during the time they run against each other, but you will also keep an eye on your total return. As discussed in article 6, if your ads overpromise or fail to filter out the bad leads, then the increased traffic may not lead to increased commissions, and CPC savings might not offset the drop in EPC. But once again, the bottom line is always . . . the bottom line.

When you believe you may have created a better ad based on a higher CTR, monitor your total return for at least a few days to ensure that it does indeed go up. If the total return goes down significantly, your new ad is a lemon, and you should stick with the old one. If the total return is significantly higher, your new ad is a keeper, and you should run it beside the old ad until it has built up a similar history, at which time you will delete the old ad. If your total return is not significantly higher or lower but is instead rather flat, then you should probably still transition to the new ad, though somewhat more cautiously. Even though this new ad may not be making more money for you, your higher CTR is probably translating to higher costs for your competition, and as long as you are not losing money to accomplish this, in the long run it can only help you.