I posted earlier about my first pay-per-click advertising campaign, and Jim Self commented, "I bet someone out there has crunched the numbers on what level of bid gets the best results."

If you're scratching your head as to what he's talking about, when you book a pay-per-click advertising campaign, you bid a certain price that you will pay if the ad is actually clicked on. Bid too low, and your ad never actually gets shown to anyone--which is what happened my first day. Bid really high, and your ad will get shown to everyone, right away!

Sounds like a good idea to bid high, right? But it's not, for a couple of reasons.

For one thing, you need to look at your potential revenue per customer--you don't want to bid a dollar a click if you only have one 99-cent book out and can only possibly make 35 cents off each customer. The math isn't always that simple--my first pay-per-click campaign was for a free book, after all, but the hope is that they'll buy copies of Trust. Some already are, which is awesome, but I have no idea what the conversion rate is, so I can't sit down and calculate my exact return on investment. But it really doesn't matter--I'm more likely to cover the cost of the campaign (which wasn't much--$71.22) if I keep the bid price low.

The other reason to bid low is that you set a daily budget, and once your campaign hits that cost, it closes down for the day. So, if you have a daily budget of $100, and you bid a dollar a click, your ads will stop showing after 100 people click. If you bid 50 cents a click, 200 people can click before your campaign goes dark. Ten cents a click? One thousand people!

If you're running an ad campaign where the click takes someone to where they can buy your book, then clearly you want to maximize clicks. In that scenario, getting as many clicks as you can before your money runs out matters far more than having the ad shown to everyone quickly.

The tricky thing is that, as Lindsay Buroker notes, Facebook suggests a range of bids per click that is very high. Lindsay said that Facebook suggested she bid almost a dollar per click; she wound up doing fine at 20 cents. I'm assuming that costs have gone up because of all the post-holiday advertising, because I initially bid 30 cents, and the ad wasn't being shown. I raised it to 50 cents, and the ad got shown fairly often.

That was for a short-term campaign--I was only running those ads for two days, so I couldn't wait and see if the bid price was going to drop. Now I've started a new campaign advertising Trang at its normal price, and I've set the bid price at 40 cents, which gets it shown some, but not a lot. I'm fine with that because this is a long-term campaign, so I can check on it every now and again to see if the bid price should be raised or lowered.

So, just to demonstrate potential-revenue-per-customer thinking:

The campaign for free copies of Trang cost me 50 cents per customer and has the potential to make me $3.44 per customer (assuming 100% buy through for both books, which is absurd, but we're talking potential here). That means if one out of every 6 or 7 clickers goes on to buy Trust, I will break even.

The current campaign for Trang is costing me 40 cents per customer and has the potential to make me $5.44 per customer. That means if one out of every 13 or 14 clickers goes on to buy both books, I will break even.

If I only had Trang out, the first campaign would be strictly a money-loser, and the second campaign would have only the potential to make me $2 per customer (one out of every 5 clickers would have to buy Trang for me to break even). If Trang was 99 cents, the second campaign would have only the potential to make me 35 cents per customer, which would make it a money-loser at my current bid price of 40 cents.

If Trials was already out and priced at $4.99, the first campaign would have the potential to make me $6.88 per customer, and the second would have the potential to make me $8.88 per customer.

In other words--get back to work!