1. Know the dollar amount or percentage of the advertising spend that actually generates results.
What Logic 361 is saying here is that since the ROI of keywords that don't convert is zero, then you should only bother calculating the ROI of keywords that do convert. The point of this is that the ROI of keywords that convert should be used to determine your conversion goals for future advertising.

2. Look at the branded keywords in your pay-per-click campaigns.
The idea here is that if your competitors aren’t bidding on their own brandname, you shouldn’t either. Now, this is, arguably, a point of contention because prevailing wisdom would dictate that if your competitor isn’t bidding on their own brandname, then there’s a double opportunity to be had by bidding on your competitor’s name. However, as the legality of this practice is something that is always in a state of flux, then you might be best off just playing it safe. After all, if your competitor isn’t bidding on their own name, then by not bidding on their name as well, you (1) aren’t missing out on some facet of competition, (2) are avoiding the possibility of having legal action taken against you, and (3) not giving your competition an incentive to get that much more competitive. You should, however, be bidding on your own brandname.

3. Know the amount of advertising expense rolling over each month that is not generating results and rebalance spending.
Now, this point seems to go almost without saying, but Logic361 puts them in the contexts of both the holiday campaign and the longtail. First, they encourage you to assess whether a non-productive keyword can be made to convert. After all, there's a reason that you wanted to bid on it in the first place. Perhaps you can geo-target it? Secondly, even if a keyword isn't costing you much, if it's not converting on the longtail (4 months +), then perhaps that negligible cost could be better invested in keywords that are converting either in the short- or long-term. Finally, bear in mind that that there's a huge difference between holiday ad-campaigns and longtail campaigns. So always keep in mind just precisely what your goals are in bidding on any keyword, and analyze their conversion rate (ROI) against the backdrop of the intentions you had when bidding on them.

4. Evaluate your ad group organization based on the performance of keyword concentrations.
Here, Logic361 is suggesting that the fewer keywords in an ad group, the better the ad group should convert. Now, of course, the fewer keywords there are, the less of a margin for error there is in making sure that they're optimized for variables such as geography and landing pages. However, the lesson that can be drawn from this is to keep your ad groups small so that unoptimized keywords can be more easily identified and addressed -- whether that involves geo-targetting, landing pages, the longtail, or dropping them from your campaign altogether.

5. Review the number of ads in each ad group.
If you're running your campaign on Adwords, Logic361 suggest running only 2 ads per ad group. This is because AdWords will keep ads with a high click-through-rate (CTR) at the forefront, and that will keep newer ads that have not yet had a chance to demonstrate a high CTR from ever doing so.

6. Build on ads that work by implementing and monitoring reporting that distinguishes Champion Ads from Challenger Ads.
This is simply a case of how ad campaign are greater than the sum of its parts. In a word, by improving on campaigns that have been successful thus far will yield a disproportionately high increase in CTR.

7. Evaluate your ad group organization based on the performance of click rate concentrations.
Now, here, Logic361 is suggesting that you evaluate your ad performance based solely on CTR. However, more often than not, a conversion is more than just a mere click. All CTR tells you is what keywords and ad copy attract the most attention from the users. If those clicks aren't converting into sales, however, then you should be considering whether the content of your landing pages is (1) appropriate, and (2) relevant to the ads that are generating the clicks.

8. Examine keyword inventory word count concentrations based on performance.
This is simply a case of keeping the longtail in mind. If a use searches a single work, they are likely looking for information rather than products. You should, then, be bidding on series of words. As an advertiser, try to put yourself in the shoes of a consumer that might be searching for the products that you sell, and anticipate what you might search for if you were that shopper. You should also consider using negative keywords to make sure that your ads aren't being rendered for similar but irrelevant search queries.

9. Evaluate the use and implementation of match types.
In other words, consider whether you your PPC campaign coordinated with your SEO and traditional marketing campaigns. The idea here is that, if so, (1) users are probably being exposed to your marketing elsewhere, so (2) your brandname or services should seem ubiquitous. This will make you seem like that much more of a trustworthy seller.

10. Evaluate keyword inventory and strategy based on the performance of click rate concentrations.
Basically, if your keywords are generating a high CTR but aren't converting, then you're probably making some serious PPC mistakes. For example, you might want to pre-qualify clickers with your ad copy, or reconsider whether the #1 spot really converts (it often does not). After all, one of the 8 basic rules of PPC is to let your placement vary.