Determining visitor value

The first thing that you need to do when you begin considering PPC strategies is to determine how much each web-site visitor is worth to you. It’s important to know this number, because otherwise you could find yourself paying far too much for keyword advertising that doesn’t bring the traffic or conversions that you’d expect. For example, if it costs you $25 to gain a conversion (or sale) but the value of that conversion is only $15, then you’re losing a lot of money. You can’t afford that kind of expenditure for very long.

How To?
To determine the value of each web-site visitor, you’ll need to have some historical data about the number of visitors to your site in a given amount of time (say a month) and the actual sales numbers (or profit) for that same time period. This is where it’s good to have some kind of web metrics program to keep track of your site statistics. Divide the profit by the number of visitors for the same time frame, and the result should tell you (approximately) what each visitor is worth.

For Example

Say that during December, your site cleared $2,500. (In this admittedly simplified example, we’re ignoring various things you might have to figure into an actual profit and loss statement.) Let’s also say that during the same month, 15,000 visitors came to your site. Note that this number is for all the visitors to your site, not just the ones who made a purchase. You divide your $2,500 profit by all visitors, purchasers or not, because this gives you an accurate average value of every visitor to your site. Not every visitor is going to make a purchase, but you have to go through a number of non-purchasing visitors to get to those who will purchase.

Findings
Back to the formula for the value of a visitor. Divide the site profit for December ($2,500) by the number of visitors (15,000) and the value of your visitors is approximately $.17 per visitor. Note that I’ve said approximately, because during any given month (or whatever time frame you choose) the number of visitors and the amount of profit will vary. The way you slice the time can change your average visitor value by a few cents to a few dollars, depending on your site traffic. (Also note that the example is based on the value of all visitors, not of conversions, which might be a more valid real-life way of calculating the value of individual visitors. But this example is simply to demonstrate
the principle.)

The number you got for visitor value is a sort of breakeven point. It means you can spend up to $.17 per visitor on keywords or other promotions without losing money. But if you’re spending more than that without increasing sales and profits, you’re going in the hole. It’s not good business to spend everything you make (or more) to draw visitors to the site. But note the preceding italicized words. If a $.25 keyword can raise your sales and profits dramatically, then it may be worth buying that word. In this oversimplified example, you need to decide how much you can realistically spend on keywords or other promotions.

visitor.jpg

Maybe you feel a particular keyword is powerful enough that you can spend $.12 per click for it, and raise your sales and visitor value substantially. You have to decide what profit margin you want and what promotions are likely to provide it. As you can see, there are a number of variables. Real life is dynamic, and eludes static examples. But whatever you decide, you shouldn’t spend everything you make on PPC programs. There are far too many other things that you need to invest in.

Conclude
Popular keyword phrases can often run much more than $.12 per click. In fact, some of the most popular keywords can run as much as $50 (yes, that’s Fifty Dollars) per click. To stretch your PPC budget, you can choose less popular terms that are much less expensive, but that provide good results for the investment that you do make.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.