Welcome to WebmasterWorld Guest from 18.207.136.184

Forum Moderators: buckworks & eWhisper & skibum

Message Too Old, No Replies

Undoing the Carnage - Plan of Action

     
4:10 pm on Jul 14, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


Preface:

Like many of you, I've enjoyed a nice, swift kick to the nads this week. My big money-earner campaign has suffered the most.

CPC is up 22%. Clicks are down 20%. Approximately 40% of my keywords have been disabled, with a few dozen additional keywords disabled each time I log into Adwords. I wouldn't be too devastated if things remained where they are, but I'm not confident they *will* remain where they are. If this keyword attrition continues, I will be in big trouble.

-------------------------------------------

Here's my plan:

1. Create a new campaign which will contain all the keywords from the old one (obviously, I'm going to leave the original campaign alone so that its keyword history remains intact).

2. Let's say I currently have 10 adgroups. I'm going to turn those 10 adgroups into 50 adgroups, with each sub-adgroup containing 1/5th of the keywords of the old source adgroup. Big adgroups could conceivably be a red flag. Heck, maybe I'll make 10 times the adgroups.

3. I'm going to create a keyword-specific URL for each and every keyword. I will programmatically rearrange and slightly tweak the content that I'm currently using for each adgroup on a keyword-by-keyword basis. Layout will be slightly altered, paragraphs rearranged. Sentence templates will be used to dynamically insert keyword-containing "relevant" content in several random locations per page. I may even add random outbound links with the link text set to the keyword (obviously, these guys will have to be well below the fold).

4. I might consider using dynamic keyword insertion in my ads. Long ago, I used to make use of this facility, but I found that my ROI was substantially better when I switched to entirely static ads. I also felt I was being more honest with surfers by telling them exactly what they were getting in the headline of each ad.

I hate having to do crap like this. I worked hard over several years to create a very profitable, effective campaign.

----------------------------

Any thoughts on this plan? Has anyone tried anything similar, and if so, what were the results?

Finally, will the existence of the keyword in an inactive state in the original campaign preclude its "reactivation" in the new campaign? Do I need to nuke my inactive keywords first?

4:19 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


Another 4 keywords have gone inactive in the 7 minutes that have elapsed since I started this thread.

Good times.

4:24 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:July 12, 2006
posts:38
votes: 0


I believe the system will ignore the new KW and stick with the one with the long history.

Also if you delete the KW, the past history remains in the systems "memory" and the new KW will not activate.

Otherwise a person could restart forever. No, they give you one shot at a kw and thats it.

I may be wrong, (I hope I am, but I don't think so) and if so someone please correct me. But I don't think it is that simple to just create a new KW in a new adgroup and start a new.

4:30 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:May 24, 2006
posts:24
votes: 0


More likely you'll lose the history and the keyword will still be marked as inactive and not worthy.

-Mike

4:31 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


ohwell, I know that there is some global memory concerning keywords you've used, but I figured - especially if I nuked the original keyword - that it would be given another shot in a new campaign/adgroup.

My inactive keywords by and large have stellar historical CTR's - I think it is the quality score, alone, that has destroyed them. I figured a new landing page per keyword might boost the quality score enough that they would be reactivated.

Then again, as others have pointed out, perhaps my CTR was simply too high for these keywords...and there is nothing I can do to address this.

Clear as mud, eh?

4:31 pm on July 14, 2006 (gmt 0)

Junior Member

10+ Year Member

joined:June 27, 2006
posts:128
votes: 0


It safe to say that the more broad a search term is, the more broad the information should be on the landing page, with more options and links. While more specific keywords terms require more specific information. Also, less options or links are required for a good quality score on more specific keyword terms.

Yes, large keyword lists are bad, especially if they point to one landing page (dynamic or not) because it is impossible to be relevant to most users when the keyword list is so large (unless you offer A LOT of information on your site and easy access to that information on your landing page). Best practices: make many ad groups, and make sure your keyword list in that ad groups fits a specific theme. Make sure the landing page fits that theme, presented in a way that makes YOU unique from the other sites.

In short, ask yourself, will someone searching for 'widgets' find enough information on 'widgets' to make them happy.

Obviously, the broad term 'widgets' require more work to make them happy than 'sticky blue widgets.'

Google has a broad array of metrics to determine if a visitor FOUND WHAT they are looking for, and were happy with their visit.

Remember that the average user is very very novice compared to WebmasterWorld visitors. You have have to adjust your point of view when thinking about the question: "Will my visitors find what they are looking for given the search term?"

WES

[edited by: RockSolidWes at 4:39 pm (utc) on July 14, 2006]

4:31 pm on July 14, 2006 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:Apr 1, 2006
posts:1536
votes: 0


Once we are tagged as low quality, nothing seems to work...
4:37 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


If this continues, and my efforts to undo the damage fail, I may consider switching to a new domain, and setting up a new corporate entity so that I can legitimately sign up for another adwords account.

New domain + new adwords account + "quality" enchancements = keywords remain active?

Who knows. But it's worth a shot. This is my livelihood, afterall.

4:42 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


RockSolidWes, this e-commerce campaign of mine has maintained roughly a 7% rate of conversion for the last two years. I think it's safe to say that people are finding exactly what they are looking for.
4:42 pm on July 14, 2006 (gmt 0)

Junior Member

10+ Year Member

joined:June 27, 2006
posts:128
votes: 0


"Once we are tagged as low quality, nothing seems to work... "

It will not be an overnight fix. Earlier in this post I gave some suggestions. If you follow these suggestions, Google is constantly re-evaluating landing pages and adjusting bids. Also make sure you read carefully the landing page guidelines on Google's site. I know it is the standard reply, but really read and ponder and try to get Google's point of view.

4:44 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:July 12, 2006
posts:38
votes: 0


>>I figured - especially if I nuked the original keyword - that it would be given another shot in a new campaign/adgroup.<<

I don't think so. You can certainly change the landing page for the orig KW and the orig adgroup of course and mess around there and it may very well work. But I dont think you can delete a long standing KW and start a new kw and expect the system to start the new KW with a fresh history or even to fool the system since it is now in a new adgroup.

I think you have to work on the long running KW/adgroup and try things there.

The guys who wrote the system are way too bright for simple stuff like that to get by them. : )

But I do hope I'm wrong and there is a way to start a new. I would love to know myself. Anyone willing to point us in the right direction it would be highly appreciated.

4:52 pm on July 14, 2006 (gmt 0)

Junior Member

10+ Year Member

joined:June 27, 2006
posts:128
votes: 0


spc90210,
I understand your convern: But conversion rates can sometimes be misleading for determining a users experience. We have been hit hard, and we have excellent conversion rates. But have been told our user experience metrics were low on certain keywords.

I look at it this way: What is a usefull attitude to have. The best attitude for me is to learn the new system, and work with it. I went through a phase of demanding a rereview, and after they rejected everything, we have decided to work very closely with our rep and ask questions on what Google wants.

Google has trust in the computer algo that determines landing page quality. We either learn it or go someplace else.

5:01 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


Wes, did your rep give you any concrete details on these user experience metrics? If so, do share! ;)

I think it might finally be time to give my rep a call; for whatever reason, I've always strongly disliked the notion of having any google employee - even my rep - manually peruse my stuff.

5:44 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:July 12, 2006
posts:13
votes: 0


"The best attitude for me is to learn the new system, and work with it. I went through a phase of demanding a rereview, and after they rejected everything, we have decided to work very closely with our rep and ask questions on what Google wants."

Exactly my attitude as well. However having my rep tell me that she see's no reason why my keywords went up 2000% and now awaiting a reply from the quality review team is nerve wracking. My friend was left with having to review the guidelines. Thats too vague, and I'm sure nobody wants to spend all the time on the phone with their rep :)

I already spend all day going over spreadsheets and cutting loss leaders, increasing ROI, spotting trends and so forth. If someone would tell me "please add a link to your privacy policy".. now *that* would be extremely helpful.

9:39 pm on July 14, 2006 (gmt 0)

Junior Member

10+ Year Member

joined:June 27, 2006
posts:128
votes: 0


I donít think any of the REPS know the formula Google uses to determine if a landing page has a decent quality. However, I gained some understanding on the subject.

Google Solves Problems by Computers.

Computer Solve Problems by Math.

Google had a problem with poor user experience. Therefore, they found a need to evaluate landing pages. There are many variables they can assign to a landing page based on the data Google collects. For all we know, they can use Analytics, Google Toolbar, what percentage of the user clicks the back arrow button after first glance on a website; in addition to on page factors such as number of outgoing links, keyword saturation, and amount of text and graphics.

From my best guess, there is probably a non-linear mathematical formula which computes this data and assigns a score.

I really think the percentage of visitors that click the back button upon first glance on a website is a strong component, yet I do not have confirmation. (On a side note, does Google have the technical abilities to determine HOW LONG a user stays on a website before they click the back arrow?). That is why providing many links and quality content is important. If your keyword is broad, you want a user to spend time on your site before returning to the SERP.

I am going to test a quintessential MFA (Made for Advertising) page alongside a content rich page with awesome links designed to engage the user. My hypothesis is that while many people will stay and click around a MFA page, I will see a high percentage that leave immediately as compared to a content rich page.

[edited by: eWhisper at 10:24 pm (utc) on July 14, 2006]
[edit reason]
[1][edit reason] Please no requests for stickies - See TOS. [/edit]
[/edit][/1]

10:10 pm on July 14, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


Thanks for the info, Wes. The wheels are finally starting to turn in my head...

Your studies also sound interesting; I may very well take you up on your offer to share your findings. :)

Since I'm all about the path of least resistance, the first thing I'm going to test is using distinct landing pages for each and every keyword; currently, I only use a handful of landing pages for all of my adgroups. I'm very curious to see if adding on-page factors will improve my quality score whatsoever. The verbiage will be tweaked in a distinct way on each page based on the keyword searched on. I'll also shift page elements around a bit. The keyword will appear several times on each landing page. I'm doing this first as this will be trivial for me to automate. I have my doubts about it doing any good, however.

My next endeavor will be to add dramatically more internal links to other pages on my site. I currently try to funnel everyone immediately to a "buy now!" page. So people are either clicking the "buy" button or they are immediately backing out, I'm guessing. I used to have more linkage, but I found that the more concise I was with my sales message....the more I made all roads lead to "buy now!"...the more money I made. My conversion will drop if I give customers more options. Then again, I suppose I could make these internal links very discrete and out of the way.

I'm beginning to think that you may be correct about the whole back-button business. It could be as simple as measuring the time differential between when your ad is clicked and when the user clicks another adwords ad. The shorter that time interval relative to your competitors, the less users are getting out of your site. If this truly is about improving user experience, something of this sort could be a major factor.

If this all about on-page factors...this is easy to deal with. I can tweak / optimize pages, add discrete links to other internal pages, etc, in a relatively automated fashion. If this, however, is just as much (or more) about off-page factors...then this particular campaign could be in trouble. This should be interesting.

I'm definitely willing to share the results of my tests.

10:28 pm on July 14, 2006 (gmt 0)

Senior Member

joined:Jan 3, 2003
posts:1023
votes: 0


"On-page" and "back-button" my arse.

When user searches for info, or comparison-shopping, or this is just a freeloader, then giving him a page full of info is great. No back-button result. However, if that particular user sees a widget selling page with big red BUY button, he will immediately click "back". So here's my other guess about their "quality score":

Eventhough you think you convert at 5% (or whatever percent), G$$gle's algo thinks it is a 95% back-button page, and hence is a "bad user experience".

G$$GLE, LOOK. WHEN I PAY HARD CASH, I DON'T GIVE A RATS ARSE WHAT YOU THINK ABOUT MY PAGE, AND I PROBABLY KNOW WAAAY MORE ABOUT ITS CONVERSION THAN YOUR ANALytics. BEAT IT, GET US OUR PRICES BACK!

What an arrogance.

I just halfed my budget, and doubled one on Yahoo! and MSN. Ironically, my conversions from G$$gle's traffic is way down in July, so it was an easy decision.

And I don't think creating new account will do, as it will be the same eCommerce landing page where 95% of freeloaders still click "back".

3:44 am on July 15, 2006 (gmt 0)

Preferred Member

10+ Year Member

joined:July 26, 2005
posts:486
votes: 0


G$$GLE, LOOK. WHEN I PAY HARD CASH, I DON'T GIVE A RATS ARSE WHAT YOU THINK ABOUT MY PAGE, AND I PROBABLY KNOW WAAAY MORE ABOUT ITS CONVERSION THAN YOUR ANALytics. BEAT IT, GET US OUR PRICES BACK!

This is exactly why Google has had to make these changes. Because you don't care about Google's visitors or their experience. You ONLY care about selling your product. Nothing wrong with that, but it does not work well as a long-term strategy for Google, so they need to make changes so both their search results and their ad results remain viable going forward.

By your logic, they could just sell out all their inventory to porno sites, because a percentage of people might even sign up, and the porno sites know they will sign up and convert. The rest of the people who had a bad experience who cares about...

Like I said, not your problem, but it is Google's problem, and they are trying to do something about it. It is one thing if someone clicks on your ad and says, wow good information I will note that for the future, but I am not buying today, and "Boy that was a waste of my time". The more time wasted, the less trust in ALL of the ads, and the less effective they are for everyone. If everyone stops clicking on ALL ads,it doesn't matter how smart you are or how well you think you know your business, nobody will end up there via Google.

5:58 am on July 15, 2006 (gmt 0)

Moderator This Forum

WebmasterWorld Administrator skibum is a WebmasterWorld Top Contributor of All Time 10+ Year Member

joined:Sept 20, 2000
posts:4472
votes: 1


You ONLY care about selling your product. Nothing wrong with that, but it does not work well as a long-term strategy for Google, so they need to make changes so both their search results and their ad results remain viable going forward.

If people are clicking on ads, finding what they want and making a purchase, ads don't get much more relevant than that. Most advertisers are not running ads on Google to provide information and most are only there to sell their products. If they weren't generating sales from the ads, they wouldn't be advertising.

Could this whole thing be a duplicate content penalty applied to the sites/pages in the paid listings?

6:17 am on July 15, 2006 (gmt 0)

Full Member

10+ Year Member

joined:Apr 12, 2004
posts:203
votes: 0


Because you don't care about Google's visitors or their experience. You ONLY care about selling your product.

I care about Google's visitors "experiences" as much as Google cares about its advertisers.

Is that a better comparison?

99% of Google's income is from advertising.

6:51 am on July 15, 2006 (gmt 0)

New User

10+ Year Member

joined:July 12, 2006
posts:38
votes: 0


The fact is, as any WM will tell you who has been in bus a while, the only ones who click on search engines anyhow are newbies who don't know the web. That is why the traffic converts. (Granted that is more true of adult) After a while they don't need SE's they have bookmarks. And they sure don't hold the engine reponsible because joe webmaster's page gave them a bad experience, was a little too hard to navigate, or whatever.

99 out of 100 are not going to associate the landing page with the search engine.

A search engine is just a tool like a mouse. Do I blame the mouse when it clicks the wrong link.

They prob got some complaints from .0000001% of users and 40% of WMs are now paying for it.

10:59 am on July 15, 2006 (gmt 0)

Senior Member

WebmasterWorld Senior Member 10+ Year Member

joined:July 11, 2006
posts:771
votes: 0


For what its worth, I, like the rest of you, have spent hours analysing and tweeking Adwords to try and reactivate keywords. I have found a way, but to publish how in a public forum, that is regularly scrutinized by Adwords employees would be suicidal. All I can say is bat on, there is a way around this for the moment.

From personal experience, trying to reactivate a keyword for a specific display url is futile, within 24hours you will be back to square one. It does not seem to matter whether you start a new campaign, adgroup.....nothing seems to work. So with that in mind, and the fact that there is a way - you should have a good idea where to spend your time and effort!

Sorry I can't be more specific, and I'm really sorry that Google has destroyed so much hard work that you guys ahve put in overnight.

2:01 pm on July 15, 2006 (gmt 0)

Junior Member

10+ Year Member

joined:Apr 23, 2003
posts:158
votes: 0


Any other PPC Search Engines worth doing?

We all know about:
Yahoo
MSN

What about Miva? (Formerly FindWhat)

Any others?

3:23 pm on July 15, 2006 (gmt 0)

New User

10+ Year Member

joined:July 14, 2006
posts:10
votes: 0


Hi...im new here and i have to say i enjoyed reading
this thread very much.
i have also been affected very hard as of the new changes,
and i think we should all join forces to share our expiriences so we could possibly get some sense out of the new changes and possibly resovle them.

i agree that their is no sense in griping about the situation, rather except it for what it is and try to deal with it if possible.

i would like to share with you some of my thoughts...
all my landing pages were affected by the change and are now disabled, and also a good portion of my direct link ads, however some of my
adgroups that i refer traffic directly to the vendor were not affected at all, and i think these adgroups are the key to everything.
apparently something in these sites meets the google robots fancy..(im sure my keywords also have to do with it,however i have much more specific keywords to other sites which were all disabled,it has to be the vendors page itself).
perhaps if we list lots of sites which the ads that refer to them were not affected we could come up with something in common to all these sites , a possible design/lay out which works.

another thing im thinking about is the death of landing pages, it seems G declared open war on these type of pages.
remeber that a few months ago G inroduced us to the rule that only one ad in every search page can refer to the same URL, that in turn caused a lot of us to create landing pages so we can overcome the problem.
So now G simply decided to eliminate the landing page possibility as well...that makes sense to me.

[edited by: eWhisper at 3:28 pm (utc) on July 15, 2006]
[edit reason]
[1][edit reason] Please don't drop links. See TOS. [/edit]
[/edit][/1]

3:52 pm on July 15, 2006 (gmt 0)

New User

10+ Year Member

joined:May 30, 2005
posts:6
votes: 0


My campaigns were all nuked in April so I've been at this awhile. At first I tried completely revamping my campaigns by eliminating the bad converters and creating many adgroups and renovating my landing pages. Also tried using new landing pages (optimized) and using only one keyword per landing page, etc. etc. etc. Tried just about everything and got no results. All of my remaining keywords remained nuked and the new bid prices did not even budge.

The only way that I found relief was to create new campaigns using landing pages on a NEW DOMAIN. I used the same keywords but did shorten the lists and make the keywords go to a specific landing page that was optimized for those keywords. Using a NEW DOMAIN seemed to do the trick. Hope this helps.

4:19 pm on July 15, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


kcollier63, that does indeed help. And it may be something I'll have to try.

I hate having to play these games, though.

4:39 pm on July 15, 2006 (gmt 0)

New User

10+ Year Member

joined:July 12, 2006
posts:38
votes: 0


kcollier63, Thanks a lot, good stuff. May I ask did you delete the old KWs and start new adgroups (you mentioned you started new campaigns) or just do this in the same adgroups as before.

I wouldn't think starting a new adgroup with the same KWs would matter but perhaps it does. (normally new adgroups with the same KWs will not activate as the system still "sees" the old, now deleted KWs.)

Thank you again for your help. Highly appreciated by a lot of folks I'm sure.

6:21 pm on July 15, 2006 (gmt 0)

Junior Member from GB 

10+ Year Member

joined:July 14, 2002
posts:115
votes: 0


I second the new domain theory, that is most certainly the case. One of my accounts is suffering the $10.00 limit on all keywords. The improving quality stance is simply a deceptive smokescreen for filtering out advertisers that don't fit. Its a complete lie.

Using the original domain all keywords, are now set to $5-10 which they know is a unworkable price. No improvement in page quality makes any difference at all. Only when I change the domain do the ads return to their correct bids. All the statements about quality are total deception, it has nothing to do with it, once your domain is marked then nothing you do to your landing pages will bring it back. The only way around it that I see is either a new domain or to request a review. How many people can just change their domain at a stroke?

The professional way to work would be to update their t&c's to say what style of advertiser can or cannot advertise and approve/reject on that basis. What they are doing is outrageous and unethical beyond belief. How many people must have spent their weekend trying to improve quality to get bids down, unaware that it is a fruitless endeavour once your account/domain has been hit.

This is what I have seen based on new test adgroups on original domain versus new adgroups on a new domain in the same affected account.

I hope their advertisers leave in droves, enough to really really hurt and reverse this unprofessional behaviour. Adcenter can't come soon enough.

6:54 pm on July 15, 2006 (gmt 0)

New User

10+ Year Member

joined:July 14, 2006
posts:10
votes: 0


Possibly it is too soon to judge the domain theory.
the world infamous bot works in mysterious ways, you may
yet wake up 2 days from now only to discover the bot
came to visit you in the night and its back to square one again..
8:41 pm on July 15, 2006 (gmt 0)

New User

10+ Year Member

joined:Mar 28, 2006
posts:38
votes: 0


Well, I decided to be good instead of attempting to "game" the system.

I did not set up new landing pages per keyword, but rather, I added what I thought google wanted to see on my existing landing pages: more internal links, some external links, much more exposition, less repetition of text, privacy policies, etc. I rolled back in content that had previously been removed to increase conversion.

Hell, I even tweaked all the pages so that they passed W3C validation.

And now it's RonCo time - "set it, and forget it."

I hope to recover, but I don't expect it. This has most definitely been a rude awakening. Now to figure out what comes next...

[edited by: spc90210 at 8:46 pm (utc) on July 15, 2006]

This 51 message thread spans 2 pages: 51
 

Join The Conversation

Moderators and Top Contributors

Hot Threads This Week

Featured Threads

Free SEO Tools

Hire Expert Members