How Not To Track Paid Search

I’ve been spending a lot of time in the paid search arena recently. I’m planning on putting together an in-depth paid search class soon and so I’ve been thinking a little bit more about it than usual.

One of the things that I see time and again with people that are new to paid search is that I feel they put emphasis on the wrong numbers. They are looking at the wrong paid search metrics.

The metrics that should guide the vast majority of your actions is the cost per conversion and conversion rate.

In order to accurately gauge the success of your paid search campaigns you need to establish one or more (preferably more) conversion metrics on your site.

This could include things such as:

  • Filling out a form.
  • Buying something.
  • Downloading a white paper.
  • Spending a certain amount of time on the site.
  • Watching a video.
  • Making a phone call.

Anything that is measurable and has value to you can be a conversion.

This may seem fundamental and basic to some but I see this as a hurdle for many time and time again.

I wanted to spend the rest of this article explaining why the other numbers are far less meaningful and how they can be misread.

Here is a typical dashboard report of some paid search campaigns.

We all want clicks. The more clicks that we get the more traffic to come to our website. The overarching issue however is we don’t need just any clicks. We need the right clicks.

If you are in the real estate business and you want a ton of clicks, simply bid on the phrase “real estate.” I assure you, you will get clicks. But that has nothing to do with your business. You very likely would be spending a ton of money on traffic that is too general and not ready to buy.

You need to look at the exact traffic that you have coming into your site and determine if it is good traffic or not. You don’t do this by just looking at clicks. You do this by looking at your cost per conversion. The cost per conversion determines whether or not the clicks are good or not.

I once did a campaign for a large brand that the key metric was impressions. They didn’t so much care about clicks. They just wanted people to see their ad.

If you are a big brand name that is a possible metric. But for most of us nobody knows our name. So impressions are pretty much meaningless.

Incidentally, this is how banner ads have always been sold. You pay for the impressions. Banner ad salespeople always talk about “eyeballs.” The goal is to get people to see your ad. Again, if you are a brand name that everybody knows then this has value. But if nobody has heard of you, you need people to get to your site. And furthermore you need them to engage in your site in a measurable way.

Even getting people to come to your website is not a good enough metric. As we saw in the first column, clicks can do that for you very easily. You need to get people to engage in your site in one way or another. That’s conversions.

CTR or Click through Rate
I’ve seen many people base all decisions on this number. They look at this percentage and decide whether or not the campaign is working.

Here is an example of why this is not a great determining metric:

Let’s say you sell a welding machine. But it isn’t a metal welding machine. This welds plastics and vinyl. However, you would like to come up for the phrase “welding machine.” So you create an ad that says: World leaders in vinyl welding. If you weld vinyl check us out.

An ad like that very likely will have a low click through rate. That’s because most people are looking to weld metal and other materials. The vinyl welding market is just a percentage of the overall welding market. So if people read the ad they are likely to only click on it if they are looking to weld vinyl. In this example we would actually be looking for a relatively low click through rate.

You will also see a low click through rate when your ads are displayed outside of the search network of the search engines. This doesn’t mean that it is a bad campaign. Again, you determine whether or not you’re having success by conversions and cost per conversion.

Who knows if you are spending too much or too little? Obviously, if things are working within an agreeable return on investment you could spend tons of money. This number by itself is meaningless. Whether or not you’re spending enough or not enough depends on whether or not the traffic is working for you. And you guessed it, you figure that out by looking at your conversions and cost per conversion.

Average Position
This is a number that you should experiment with. You might try raising it and lowering it. You’ll get different results based on the position you’re in. It will effect all the numbers. But what position you should be in will be determined by conversions and cost per conversion.

I have seen people obsessed with total conversions. They want as many conversions as they can possibly get. This can be a feasible strategy if you are looking to get significant growth in market share and capture a larger percentage of your industry. Some people are in that point of their business.

I’ve also seen people frustrated because their competitors are always in the first position and coming up for every phrase imaginable. This is painful because you know that you are losing customers to them. But you have to play your own game. If you are looking for top line growth where you are just trying to take over the market then fine. Plow as many conversions into your site as you possibly can.

But if your goal is bottom-line growth, where you are looking to make a profit, then this number is not where you need to be looking. You need to look at the next column, your cost per conversion.

Focusing on total number of conversions is for people who are looking for growth and increasing topline. It is a valid strategy for a business at a certain point in their evolution. However, most people I work with are not looking for this kind of growth. They are looking to make money now. If that’s who you are then seen me in the next column.

Cost per Conversion
For people who are looking to get a good return on investment this is the column that matters. Is this an acceptable cost per conversion or not? The answer to this question will be individual to your specific business. I can’t tell you in this article what your cost per conversion should be. But my experience has been once businesses understand this cost per conversion focus they already have an idea of what a good cost per conversion should be.

You will typically find that you can get a very low cost per conversion by finding very targeted, specific phrases. The problem with this is that there is typically a small amount of traffic in those niches.

The more general the phrase typically the more expensive the cost per conversion. But you also will find that you get more total conversions with these phrases.

The best campaigns usually have a nice mix. They will have more general phrases that cost more per conversion. And they will have niche phrases that are a lower cost per conversion. When you average the two together you oftentimes come up with an acceptable cost per conversion price.

Conversion Rate
This number is also very useful and can be a guiding factor. I’m oftentimes asked what a benchmark of the conversion rate should be. The bar I usually set is 2%. If your conversion rate is under 2% I feel that the campaign is probably underperforming. And if it’s above 2% I’m usually quite happy. That’s not to say that I don’t try to get a higher conversion rate. But 2% is the number that determines for me whether or not a campaign is healthy.

When I am looking at a campaign to determine success I typically look at the cost per conversion and conversion rate. Those numbers together tell me if I am on the right track.

While the other numbers are certainly not invisible to me, they are almost always secondary. I only start to make decisions on those numbers after I have worked on my cost per conversion and conversion rate.

If you are focusing this way already then congratulations. You are on the right path. But if you don’t have a conversion metric within your paid search campaigns you are probably not spending your money wisely. I would strongly encourage you to set up some conversions as soon as possible. It will change the way you think of your paid search campaigns.

Most recent posts by Sage Lewis

Real Time Web Analytics