I've kept my keywords very relevant, and I've made sure my ads are very honest as far as what people will find after they click.
I'm currently spending about $25/day for about 500 clicks. Can anyone point me toward any techniques to maximize return and/or minimize cost?
As for ROI tracking, what do you guys do with that info? Maybe review your ROI every week and cancel ads based on that?
- Grant
Robsp -
I actually don't use the AdWords conversion tracking system. I append a special URL parameter so I know if an order was placed that came from AdWords as long as they don't end their session before converting *and* deny my cookie. I'm not sure of the best way to handle the info though.
- Grant
I can recommend the Adwords system as this not only tells you which group or keyword generated sales but also what ad did, which helps a lot when finetuning your proposition. It works for us in over 60 campaigns.
The 30 day cookie is also key as in *many* cases we do not get same day sales in our campaigns (I would like the cookie to be 90 days)
- Grant
1. Do I think the keyword should work?
If it's a keyword I simply tried as an experiment, unsure if it will convert, then I have to examine the time it'll take to do the following vs just deleting the keyword now.
If it's a keyword I think should work (and has decent volume), then I'll go through the following process.
2. Examine the time it'll take to create ads and landing pages and invest time vs the profitability of the keyword if it start to convert. If the keyword has limited volume on a low ticket item, the invested time might not be worthwhile. If it's a high volume keyword or a high priced item - the testing process is usually worthwhile. Even if it only makes 1 in 3 keywords viable, the profits from those keywords often take care of the vested work.
3. Examine the ad.
Is the ad accurately describing what the user will see when they click on the ad?
4. Examine the landing page.
Is there vital information about the keyword or ad missing from the page?
5. Split test the ads.
Create several ads that give different offers. Often keywords don't go to user intent - so determine the intent of people who are going to buy from you by using several ads with different offers to see if any do convert.
6. Split test the landing pages.
If it's a good keyword, then it's worth it to me to create 2-5 unique landing pages to test conversions.
If after all the following have failed, (and I've often consulted through the process with others in case I missed something important), then the keyword gets deleted.
Make 3 ads per adgroups with a different approach to the proposition (which is of course specific to your business) and measure over time which one converts best. Keep fine tuning the ads until you make a decent profit.
