| 1:16 pm on Jun 17, 2007 (gmt 0)|
You could always install tracking code and see what campaigns / products are selling and reduce the spend on other campaigns.
Then after you know what is working you could spend more on the higher converting campaigns that would give a better ROI.
| 4:44 pm on Jun 18, 2007 (gmt 0)|
One of the things that worked best for me was to target the #5-8 positions, rather than trying to go for #1-3. We actually got BETTER conversions in some cases, with much reduced cost by doing this, because I took a look at the ads that were in the top spots, and worked very hard on really focused text that explained exactly what we were doing; a lot of our competitors were showing vague or generic type ads. Made a big difference in the bottom line.
| 4:48 pm on Jun 18, 2007 (gmt 0)|
We used DayParting to reduce our spend on weekends and Midnight to early morning. This greatly reduced our cost with little effect on sales.
| 4:49 pm on Jun 18, 2007 (gmt 0)|
Using negative keywords can save you a lot. Watch your log files to get more ideas. Run the ads only during your business hours or figure out the most profitable time and only run ads then. Work to get your quality score up to great. Make sure that when somebody comes to your site from an ad that the keyword they typed in is on that page. Set up broad match ad groups and put those same terms as exact negative keywords. Make another group and have those terms as exact matches. It is very important to look at your log files and to set up something that tracks conversions.
| 9:53 pm on Jun 18, 2007 (gmt 0)|
|What would be your first things to look at to reduce that? |
Some really good tips here, already.
I am going to be very literal, CestMoiPeridot, and say that the very first thing to do is to accurately understand whether your client is making a profit from their advertising expenditure or not.
And, if this is not known, then as rj87uk suggests, it's very important to get some sort of tracking going - from a source that you are comfortable with, whether it's Google conversion tracking, Google Analytics, or a third party.
A few thoughts, to expand:
* If you know that your client is making a profit from their advertising, then this puts you in a good position in which you can take your time and experiment with 'optimizations' such as those suggested in this thread.
* In a very real sense, if your clients are currently making more than they spend, then if they spend more they are also likely to make make more - even if you change nothing. It makes a great deal of sense to try to maximize that, of course.
* If you know that your client is not making a profit from their advertising, then things become more time-sensitive. At this point it might be best to pause campaigns which your client deems to be less important, and focus on the most important ones - then apply lessons learned to the paused campaigns later on, and re-start them.
* If you don't know whether or not your client is making a profit from their advertising, then everything you might do to improve things almost amounts to guessing - and tracking of some sort becomes a must-do.
My $0.02 ;)
| 6:42 pm on Jun 19, 2007 (gmt 0)|
Is it necessary to add the phrases as negative exact matches? If the phrase is listed in group 1 as a broad match and group 2 as an exact match, there could be a conflict when exactly that phrase is searched for?
| 8:57 pm on Jun 19, 2007 (gmt 0)|
Whoever is doing this work should get Perry Marshalls book "The Ultimate Guide to Google Adwords" from Amazon. I got my copy from my local Borders bookstore and they had it in stock so you might check there.
Unless the company is one of my competitors - then keep spending the $30K lol. But seriously, it's $20 that will most likely save them tens of thousands over the course of the next 12 months.
| 12:57 am on Jun 20, 2007 (gmt 0)|
What kind of stuff does that book cover? Has anyone else had a look at it?
| 2:22 am on Jun 20, 2007 (gmt 0)|
Tonearm the reason you do this is so that you can keep all the random kw's you did not think of in one ad group. The keywords that you did think of should be in an exact match group so you can write a better ad and have a better landing page and have a better QS.
| 5:58 am on Jun 20, 2007 (gmt 0)|
Take out the best performing keywords and ads and bid on them more. Either reduce or pause all other keywords and then the PPc budget can be reduced.
| 2:25 pm on Jun 20, 2007 (gmt 0)|
ogletree, I really like your strategy. I'm wondering what happens when someone searches for "blue widgets" and I have broad match blue widgets in one ad group, and exact match [blue widgets] in another. Could the broad match ad end up showing, or will the exact match always show in this instance? I'm just wondering if it's really necessary to specify -[blue widgets] in the broad match ad group in this case.
| 5:04 pm on Jun 20, 2007 (gmt 0)|
If you have blue widgets as a broad match and have blue widgets as an exact negative match in the same group you will not show up for the term blue widgets for that ad group. It will go to another group that has it as an exact match. I also create another group the same way and set them to phrase match. The more ad groups you have the better. Every keyword entry in adwords has its own quality score. You want to keep all broad and phrase matches away from exact matches. Use as many exact matches as you can. It is all about keeping your QS up and keep lower quality stuff away from higher quality stuff.
| 8:30 pm on Jun 20, 2007 (gmt 0)|
Very interesting. Is the benefit of high QS mainly being able to bid low for a keyword?
| 8:56 pm on Jun 20, 2007 (gmt 0)|
PPC is getting to be more like SEO these days. It factors in even if you bid high. If you have a higher quality score than somebody else then you will rank higher and pay less.
Quality Score = (keyword's CTR, ad text relevance, keyword relevance, landing page relevance)*
| 10:01 pm on Jun 20, 2007 (gmt 0)|
Sounds a whole lot like SEO. Thanks ogletree.