I read an interesting article yesterday on the Practical eCommerce website called ‘When to Ignore a 50% Conversion Rate’

It discussed the issue of data validation - meaning that if you launch a new keyword in your pay-per-click campaign, then the first visitor from that keyword goes ahead and converts, then your conversion rate for that keyword is 100%. However, because you’re only looking at the behavior of a single visitor, you don’t have a large enough sample size for that data to be valid.

Initially what struck me about this article was how obvious that was. But as I’ve pondered it a little bit, it occurs to me that many people may be making the mistake of making decision on too few data (although I suspect most marketers don’t make decisions based on a single click).

So, how big does a sample size need to be? Now, if you ask a statistician, I’m sure that you’ll get a different answer, but for the typical marketer I’m assuming that we’re looking for a best guess. Here are a few thoughts that might help your guess be more accurate.

1. If you’re looking at impressions, look for at least 1,000. If you’re looking at clicks, look for at least 200.

2. When comparing conversion rates of two Google ads, I like to calculate what percentage would be expressed by one more conversion. Simply multiply the number of clicks by 0.01 (or multiply by 1%) - the answer is how many percentage points would be represented by a single conversion.

3. Our friends at Vertster also have a little tool that measures the difference between CTR’s and tells you if the sample size is big enough.

I’d be interested in your feedback, and any other techniques you might use for this sort of data validation…

Our sister site has just been named a top 10 pay per click management agency in the U.S. Conversion rate expert, Mat Greenfield is the Founding Partner of Conversion Results, and a thought leader in the pay per click world.

Here are some of his articles on pay per click strategies:

Biggest AdWords Mistakes
Google AdWords: Structure For Success
Google AdWords: Get ‘em Where You Want ‘em

Read more on the: Top 10 Pay Per Click Management Ranking

From my discussions with many Google AdWords advertisers, I think it’s true to say that the concept of Quality Score is largely not understood, and is therefore ignored.  In fact Quality Score is just one of the many variables that Google allows you to adjust - if you know how.

Quality Score is used as a part of the calculation to determine how much you will pay, and where you will rank for a particular keyword.  The better your Quality Score, the cheaper your clicks.

You can manipulate Quality Score simply by ensuring that your keyword is 1) in your ad, and 2) on your landing page.  Beyond that, you’ll have to work to increase your CTR to impact your QS.

Some people that I talk to about Google AdWords seem to be under the impression that once you have things set-up, that your job is done.  However, I think there are a number of reasons why managing AdWords is an on-going job:

1. Because AdWords is essentially an auction system, bids are always going up.  As new competitors begin advertising, and words become more competitive your position (and results) will change.

2.  Your results are never perfectly optimized.  Better ads can be written and tested, the optimal position for a particular keyword can be identified.  Landing pages can be tweaked to maximize conversion rates.

Bottom line - if you think you’re done, you’re on a downhill slope.

Google AdWords Rookie Mistake #5: Not tracking conversions.

This might be the cardinal sin of Google AdWords. Conversion tracking might take a little bit of effort to set up, but it is a critical step in identifying which keywords create conversions, and which ones just create clicks.

Google AdWords Rookie Mistake #4: Not testing different ads.

I’m a HUGE fan of split testing – both on websites and for ads within Google AdWords. New advertisers should create 2 or 3 ads for each AdGroup, and after about a month, take a look which ad has the best cost per conversion. Ad split testing should be a continual process. One tip – I like to ‘pause’ my old ads rather than delete them, that way I can see what ads I’ve already tested.

Google AdWords Rookie Mistake #3: Bidding for Position 1.

I often see new AdWords advertisers over-paying for clicks because they are pursuing position #1. I think they assume that being in position 1 is always ‘best’, but I think when you consider the additional cost for position 1, often times positions 2-5 are much better value.

Google AdWords Rookie Mistake #2: Not understanding the difference between ‘Search Match’ and ‘Content Match’ (and setting single Campaigns for both)

Google does a nice job of allowing you to determine where your ads will show. But many AdWords advertisers use these setting indiscriminately. I would suggest that campaigns be set to either Search Match or content Match (but not both). The simple reason is that the ads that will appeal to individuals actively searching on a phrase will be quite different from the types of ads that will appeal for browsers who see content Match ads.

I was recently asked to contribute to a Google 101 guide. I came up with 5 rookie mistakes that I often see, and thought that I would share them here.

AdWords Rookie Mistake #1: Having too many keywords in a single Adgroup

Since an ‘AdGroup’ is Google’s lowest level of segmentation, it should ideally be used to express a single idea or theme. Generally I find that the more specific, the better. Typically I find that any Adgroup with more than 50 keywords is probably too broad. The exception would be Adgroups with a large number of mis-spellings.

More specific Adgroups does mean more management – but it also means a much better ROI.

I called Google this week with a question about Quality Score.  I had a keyword with 1 impression, 1 click, and 1 conversion, but it had a LOW quality score.  My previous understanding was that CTR (in this case 100%) was the biggest influencing factor on adwords quality, but it seems not.

Google Support reviewed the keyword and told me that because the word was not present in the ad, it was being awared a low quality score.  The ad was relevant to the keyword, even though the keyword itself wasn’t in the ad text - causing a lower QS.

This reinforces the case for very small adgroups, with only a few highly relevant keywords that can be featured in the ad.