Know-How Exchange : Research/Metrics

You should start by asking yourself, "At what spending level would this campaign be profitable for me, given a scenario with a tight range of metrics?"

So you develop a "high, medium, low" set of assumptions for each metric. Use those to calculate a cost per sale. Then factor in the assumed profit margin, and see if the project is returning its cost (or more).

You won't get it right the first time, but that's OK. You'll begin to refine the estimates, see where the leverage is, and test your way to a mix that works for you.

The one thing you want to avoid is going into this knowing that the best set of assumptions still won't deliver a positive result. That's why you go through the exercise. After that, it is just a matter of testing your way to the best mix of keywords, copy, offer and landing page ... to maximize conversions and profitability.

Another thought: We've found that if your budget is, say, $20,000 for 6 months, you're better off spending $6-8,000 the first month, so you can learn what works and what doesn't very quickly. Then the remaining $12-14,000 will work harder for you than it would if you spent the whole amount at $3,333 per month over the 6 months.

When you move slowly/cautiously, it takes longer to get the mix right and you end up spending more of your money inefficiently, and ultimately getting less return on your total investment.

It's often a hard pill for a new/small business to swallow, but there's no way to know exactly what's going to work with AdWords until you try it. And if you don't have experience with enough ads/clicks, you won't be able to test enough iterations until most/all of your budget is spent.

Does it help to have experienced professionals to guide you? Of course. But the really good ones will tell you that they rarely get it exactly right the first time. They just know how to test the variables quickly and efficiently, and improve effectiveness at every opportunity.