HOME    articles    tutorials    tool directory    about

Free Email
META Tag Generator

Managing Pay-Per-Click

With PPC, you're paying for each visitor to your site. What you'd like to do, as much as possible, is pay for visitors who buy, and not pay to get visitors who don't buy.

I know that some of my clients have dealt with agencies who will help them spend any amount they want on pay-per-click. When this is done, typically a large number of keywords is found and listed on Google. Ads are written quickly, and generous cost limits are put on each keyword. With this method, it's possible to spend a lot of money on pay-per-click, but this is not money well spent.

With PPC, you're paying for each visitor to your site. What you'd like to do, as much as possible, is pay for visitors who buy, and not pay to get visitors who don't buy.

How can you tell which visitors buy? If your visitors can make purchases on the site, then you can use what the ppc providers call "conversion tracking". It's easy to set up. They give you a snippet of code that goes onto your site, at the page that's displayed after a purchase--the page that says "Thanks for your business." Then at the ppc site, you can measure the cost in ppc advertising per purchase.

Once you can measure cost per conversion, you can work to lower those costs.

The first thing to check is how well your are qualifying the visitors you are paying for. You need for the combination of the ad copy and the keyword to inform visitors about what you're selling. For example, suppose you have purchased clicks on the term Gatlinburg, although you are actually renting Gatlinburg vacation homes. In this case you need to be sure that your ad mentions vacation homes in Gatlinburg, so that a visitor who comes to your site, who costs you money, knows what you are offering.

You also want to make sure your ads are working well. Google lets you run several alternate ads for each group of keywords. For each group of keywords, you should have several ads running all the time. Once a week or so, check which ads are generating the most clickthroughs, and delete the ad that's performing the worst. Then put in a new ad that's a small change from your ad that's working best, or put in an ad based on a whole new idea. If you do this over time, you'll improve the effectiveness of your ads, and the better they work, the higher you'll be listed in Google's list of ads on the right side of the screen. A high click-through ratio gives you better position, that otherwise you have to pay more for.

Finally, it's simply the case that some keywords bring you a higher proportion of customers than other keywords! So if you can measure purchases, you can tell which keywords are best for your site, and put more money into the terms that give you less expensive conversions. Note that you may be putting more money into terms that have higher cost per clilckthrough, if those terms are the ones that give you lower cost per conversion.

But what can you do if your site doesn't offer on-line purchases? First, consider adding on-line purchasing. It's well known that people who shop on the Web want to buy on the Web. These days there are a lot of inexpensive ways you can sell on the Web, and it's worth trying on-line sales. When you think about it, why would anyone with a Web site who can possibly sell on the site not allow their visitors to buy on-line? If they purchase on-line there is no cost of someone to take the call, or even a cost for a phone line. The money just comes in, and the service or merchandise can go out with minimal sales cost.

In spite of my advice, though, you still might not be selling on-line, or you might have a business that doesn't allow it--one of my clients is a custom tailor, and the essence of his business is a personal visit where he measures his customer, shows him thousands of fabric samples, and advises on matters of style and fabric. In this case, what do you do? How do you tell the good keywords from the also-rans?

You can put the ppc measurement code on a page that is as close to buying as you can get. For example, for the tailor, you can put the code onto a page where the visitor checks the proprietor's schedule to determine when he will be in their vicinity. You're not measuring purchases, but certainly before someone orders custom tailored garments they will visit this page, so the people who visit this page are very likely to be more interested in buying than those who don't.

There is also a method you can use to measure which keywords are good ones. My own experimentation has shown that measuring the average time that visitors spend on site is a good proxy for measuring which people buy. Those who spend longest on the site are highly correlated with those who buy. So if you use a log analyzer such as ClickTracks, that allows you to measure how long the visitors brought by each keyword spend on your site, you can separate the good keywords from the less good.

It's important to keep using these techniques and keep improving your ppc effectiveness. By continuous work on a campaign, I have changed the average cost per conversion from over $100 to under $5, a 20 to 1 improvement in buying customers brought in for a given ppc spend.

The Bottom Line: Don't just put a lot of keywords on your pay-per-click page and then let it run. Measure, measure, measure and keep adjusting what you do to improve performance.


Dave Roberts provides expert Web marketing services to help you meet your business goals. Visit his Web site at: DaveDoesItAll.com

HOME    articles    tutorials    tool directory    about

(c) copyright 2000-2010 Anventure.  All Rights Reserved.
privacy policy