Calendar
Network
Feeds
- 24
- Sphinn It!
Went Hot: August 21, 2007 - 12:59 pm
Posted By: Lyndon 683 days ago
Topic Type: News Story (Jump to http://www.ppcthink.com)
Category: Google AdWords
29 Comments
Went Hot: August 21, 2007 - 12:59 pm
Posted By: Lyndon 683 days ago
Topic Type: News Story (Jump to http://www.ppcthink.com)
Category: Google AdWords
29 Comments
Sphinn Sponsors
Be a Sphinn Sponsor - Click HereTop 10 Most Sphunn
In What's New
- It’s All About Local Organic Domina... (18)
- Free PDF SEO Guide Ebook by Hobo-We... (16)
- Avoiding Twitter Spammers (15)
- Where the European SEO industry rea... (14)
- Online Genesis: Chapter 1 (And Part... (13)
- Twitter Unveils New UI; What This M... (13)
- Mark Prints for Self on Other’s Blo... (11)
- Some Rules To Follow As An Internet... (11)
- Real time search engines; the hype ... (10)
- SEO Checklist for Website Server Mi... (8)
Latest Comments
- I am glad you have been saying this too :). Dont grumble :), just stick to …
- i agree with u Nice your post is very useful. Thanks for all of the sites y…
- I remember something about eating the forbidden fruit of commercialism or s…
- I've always been saying this, but noooooooooooooo, *someone* always tells m…
- :) Thanks for the post. I'll be reading this for a while.…
- @ear…
- I'm not going to fill out the seven required fields to download the white p…
- - read more -
Save the date for:
• SMX Singapore - July 2-3, 2009
• SMX São Paulo - August 4-5
• SMX East - October 5-7, 2009
• SMX Stockholm - 12-13 October, 2009
• SMX Mexico - November 11, 2009
Learn more about search marketing through free online webcasts and webinars from our sister site Search Marketing Now.
Upcoming Webcasts:
- How Large Offline Marketers Drive Superior Search Marketing Results - July 21, 2009
- Trademarks, Brand Terms and PPC Advertising: Updates You Must Know - July 22, 2009
- Attribution Management Buyers Guide - July 28, 2009
Comments
Hmmm, maybe I'm missing something but basically that article says if you set up your campaigns/ad groups wrong the wrong ad could get shown to users. How would that be considered "letting Google screw you"?
"They’re overriding the specific intentions of the advertisers"
I don't see any eveidence of that anywhere, although, maybe that is due to the word "intentions". I set up all my campaigns with the intention of making money. When they don't, that must mean Google's overriding my "intentions".
Other comments such as;
"Originally, if you had an ad group containing ads for a particular brand or product with appropriately targeted keyword phrases specific to that brand or product, as one might expect those ads in that ad group would display for those searches. Unfortunately for both advertisers and searchers alike, that is not necessarily true any more. "
are presented as fact but are not backed with any information/evidence that validates them. Summarized that paragraph basically says Google will show whatever ad it feels like regardless of your ad group structure. If that really does happen I would love to see an example, theorhetical or otherwise.
Interesting theories (although I don't buy into them) just need to see more substance.
Jeremy, maybe you were one on these people who set up their adwords perfectly first time. If so bravo. I didn't and got screwed. Adwords can take you down a number of ways. You have to watch a number of variables and be aware of things that are quite complex.
Most blog posts are opinion rather than essays or white papers and so do not necessarily have a ton of evidence to back them up. You go on the reputation of the blogger, if you don't know the blogger, do a bit of digging.
Lyndon is absolutely correct. Google is serving up ads for higher bidded, not as relevant, broad match keywords rather than show exact match keywords that are "exact matches" with higher quality scores and lower bids.
The only way to stop it is to use negative keywords.
Lyndon,
Maybe I responding more to the title than the post. The title indicates that "Google is screwing you", which is not the case. People are screwing themselves by not learning the ins and outs of the system they are using.
If you have never driven a car, jumped in one and drove it into a brick wall did the car screw you? No, you screwed you by not learning to drive before turning the key.
AdWords is far from perfect but in my opinion people are too quick to pass the blame to Google for their own mistakes.
edit/add
@ JstaTad
Are you saying they do that universally? I'm looking at a number of ad groups at this moment that would kind of knock that theory out the window.
Ad group structure and keyword matching/bidding are more important now than they ever have been. The tools are there...use em'.
You can Count on Rose to stir things up. Apparently this is a conversation that needs to take place.
This is nothing new people. Dial out your adwords campaigns, don't use broad match unless your client has deep pockets, stay on it, use negatives, yadda yadda. Nothing new here, move on.
I agree with Jeremy. It's mainly the headline here that I have a problem with.
I read the article a second time thinking with the original author's headline in mind rather than this one and while it wasn't really useful to me I found the article much easier to digest.
Without that headline it wouldn't have received much attention, I get that, but maybe it wasn't worthy of such attention. In my view it doesn't do much good to complain that Google should "change the variables used in their algorithms" when you can yourself alter the effectiveness of their system.
It's self-service, so if you run a bad campaign then you're screwing yourself.
I think the article wasn't totally clear on what is actually happening: while it's not new that you need to add negative keywords and use mostly exact match in order to exercise complete control over your campaign, I have noticed that the broad match keywords are generating increasingly irrelevant search queries. As the article says, the correct response to this is to increase your negative match keywords and use more exact match. However, it would be nice if the Google algorithms that broad match search queries to keywords were a little bit better. I'm not complaining, though, since it keeps my competition a step behind me!
Since I wrote the original article I can clarify and elaborate on it. This is a complex subject that is widely misunderstood and has been written on by others as well.
The core issue is that Google has changed how Expanded Broad Match decides which ads to display for a particular search. This hits those using brands, product names, and product numbers particularly hard. Campaigns that have run well and profitable for months or even years can suddenly tank.
It is one thing to not understand the system and create a bad campaign; it is another to have campaigns and ad groups that ran wonderfully over an extended period of time and then precisely because you did an excellent job of targeting your ads your results crash and burn.
I will add some additional articles with specific examples and links to supporting documentation. This is a brand new blog and I wasn't expecting to generate such controversy. Yes, the sharpest PPC managers are aware of at least some of the causes behind this issue. Even they may not have seen all of them.
My intention is to offer simpler explanations, more detail, and step-by-step information so that even those who are not PPC experts can recognize and improve their results.
P.S. That isn't the headline on the article in my blog; it is the headline that Lyndon used when he posted it to Sphinn. Lyndon is a master linkbaiter and it certainly did get your attention, didn't it?
One thing that I believe that they are doing and not telling anyone about is using the adbot to read site content beyond the landing page and then broad matching to that content.
We had a non-profit who had a site about cholesterol. Our ads were showing for words we were not advertising on and were not synonyms to words we were advertising on. The two most unbelievable were "big black girls sex" and "fat girls sex".
We did site searches for those keywords and found them on a one page PDF deep within the site - not in exact order either. We also had an herb with a very unique name pulling our ad that was mentioned in the sites forums and nowhere else.
We also have automotive clients who are not advertising on their competitors names, but Google continues to show new ones each day. I sure didn't volunteer for that.
I personally believe they are ripe for a class action lawsuit with this stuff.
As for the headline - that's just how the game in played now - you likely wouldn't have read the article with a boring title. It's link bait at its finest.
I agree with Jeremy on this completely. Google isn't screwing you. You're screwing yourself. If you jump on a bicycle and crash into a fense, are you gonna say the fense screwed you? No - learn to ride the freakin' bike :)
Ooooh, a juicy 12 comment thread, yum. You guys are such literalists, it's all about the metaphor.
I'm no expert on adwords, don't claim to be. The way I saw the article is the Google is making it harder than necessary. To use your fence analogy, the fence moved and I call that being screwed.
I make no apologies for the title, it's a dry subject and needed a little spicing. I've seen some big time SEO types post stories with snoozeville titles and they get no votes.
Can't you abuse this in the other direction?
Bid on just the cheap keywords and ensure you have the keywords that Google decided to fix artificially high somewhere in your copy?
Note I am not an Adwords expert in any way, though am fairly well read.
It seems that this is a backdoor to compensate for those keywords you wouldn't otherwise be able to target.
Jeremy said "if you set up your campaigns/ad groups wrong the wrong ad could get shown to users". What I'm writing about is NOT due to incorrectly set up campaigns/ad groups. The better targeted they are the harder this issue can hit.
You said, "Summarized that paragraph basically says Google will show whatever ad it feels like regardless of your ad group structure. If that really does happen I would love to see an example, theorhetical or otherwise."
You're correct; that IS what I'm saying and I have plenty of examples of that behavior as do many others. I've provided a multitude of links in a follow-up article I just posted at http://www.ppcthink.com/2007/08/22/which-google-adwords-ads/. They are from a wide variety of advertisers in various PPC related forums and blogs.
I'm writing about this issue (which as SEO commented is not really new) because of the serious impact on many who are not yet aware of it.
I concur with Lyndon. The main issue is that there is no reason for advertisers to expect what has run consistently and profitably for months or years to suddenly spike spending that can put a small company of business.
Most advertisers using Google AdWords - and that includes many "Authorized Google AdWords Companies" do NOT understand how the system works and can be financially damaged in a big way by these changes.
As Lyndon points out the issue is that they "moved the fence" in the dark and advertisers are running into it. (I have horses and if you add a cross-fence to a pasture they already know they WILL run into it! - and so will advertisers).
People expect consistency and the longer something has worked one way the more they count on it continuing. I am not saying that is reasonable expectation - I am just pointing out that what changed is beyond what any logical person would expect.
For the best specific examples check out the third post in the SEF RollerCoaster thread (Feb 24, 2007) and what Tonerman posted in the Up the Creek Thread at SEW (Jan 26, 2006) that I summarized in the SEF thread Google Broad Match Turned into Expanded Match (Feb 24, 2007). (The links on my blog post are in descending date order.)
Here's something that hasn't changed:
Broad Match
Phrase Match
Exact Match
Being lazy and setting up a campaign the wrong way and then managing it loosely doesn't equate to Google screwing you.
Nice headline Lyndon - want to do some linkbaiting for my clients?
@flyingrose
We'll have to disagree on the "google shows whatever ad they feel like" theory. When campaigns are properly built Google will show the ad you tell them (via your setup/matching/bidding options) to show.
"I concur with Lyndon. The main issue is that there is no reason for advertisers to expect what has run consistently and profitably for months or years to suddenly spike spending that can put a small company of business."
Markets, and the tactics needed to market at an acceptabel roi, change constantly - and that's not unique to AdWords or PPC. In traditional media rates and specs change all the time. Succesful marketers must constantly adapt and can't expect that what worked yesterday will work today, or tomorrow.
If a business sets a budget in AdWords that could potentially put them out of business due a change in click volume that's their mistake and responsibility.
"People expect consistency and the longer something has worked one way the more they count on it continuing. I am not saying that is reasonable expectation - I am just pointing out that what changed is beyond what any logical person would expect."
Some of the links you posted reference info from years ago. Google's been "moving the fence" for years. Proper campaign setup has been (and will continue to be) the best defense against the "moving fence".
In my experience most of the people who had major issues with AdWords (poor quality scores, high minimum bids, etc) are the same who think AdWords is a "set it and forget it" marketing platform. It's not, it's a rather complex system that required near constant maintenance and attention in order to maintain maximum levels of profitablity.
@JstaTad - I would have read the article regardless of the headline. Even though I don't agree with what's being written it's still an article that sparks thought, conversation & healthy debate. Nothing wrong with that. The healine on the other hand is just plain inaccurate and misleading.
I keep hearing about the current problems with expanded broad match (and I've seen some examples that people have posted that, if true, were simply beyond the pale) but I ran extensive search query reports across about 8700 keywords over six client accounts, and I found exactly two queries that I felt merited adding new negatives to the campaign. It's not that I don't believe it's happening, I just haven't seen it happening to any of my accounts. Which is fairly weird.
Netmeg: Believe me, they're true. If you understand how neural nets work what is happening would be less weird to you. Neural nets are always "learning" and change based on historical data. The more historical data, the more anomalies. I have far more than six client accounts and vastly more than 8700 keywords in each of them.
PPCbook: Broad match HAS changed; it has morphed into expanded broad match and that is at the core of most of this; however, it is not the only cause.
Jeremy: "We'll have to disagree on the "google shows whatever ad they feel like" theory. When campaigns are properly built Google will show the ad you tell them (via your setup/matching/bidding options) to show."
Maybe they will; maybe they won't. It isn't that simple any more. It definitely reduces the risk (but does NOT eliminate it) if you never use broad match; however, before the days of Google Analytics I tagged every keyword in my major account to indicate broad/phrase/exact for Index Tools and tracked traffic and conversions by type. Roughly half of all traffic, sales, revenue, and profits came from broad match. Never using it is certainly safest; however, it also leaves half your potential income on the table.
I assure those who doubt that just because YOU haven't seen this yet does NOT mean it doesn't exist: only that it either hasn't bit you yet OR you haven't identified it yet.
It is far easier to see in a very small account than a large one. When a keyword phrase that historically spent $15 a month and generated a profitable sale suddenly spends $362.68 in half that time and generates no additional sales in an account that only has a $500 a month budget that WILL get your attention. If a $362 increase wouldn't stand out in your accounts you could easily overlook it until you notice ROI sliding.
You can see all the details on that particular example in the thread I linked into the follow on article titled Watch Spending When Using Google Broad Match. Here is a sample of that detailed information: Spending on that one phrase went from $14.79 A MONTH (less than $0.48 per day) to $362.68 ($17.27 per DAY). Spending increased by over 3,590%!!! The 13 clicks and $14.79 spent in January generated one sale while 212 clicks in the past 21 days generated no sales and cost $362.68. Clicks went from less than ½ click per day to over 10 per day – an increase of 2400%! (See the links in the follow-up article I posted for more details.)
This used to be an excellent example that anyone could see. The word involved is "luggage". Ads for very specific high dollar brands of luggage were being displayed for a search for JUST the word luggage. That is what caused the example I cite above and until recently if you searched for the word luggage you would see four to six ads for specific very expensive brands displayed for that search.
No intelligent advertiser is intentionally paying upwards of $2 a click to display ads for specific expensive brands of luggage for that word when they have generic ads for those keywords - if they want that traffic at all.
Over time advertisers have recognized the issue with that keyword and blocked that behavior with negative embedded keywords which is what I did as well. And that is VERY time-consuming in a large account that sells many brands and many products under each brand. (Frank Watson aka AussieWebmaster at SEW posted that fix and has helped more advertisers than we'll ever know.)
Hopefully when we notify Google about these issues they may fix some of them over time. If you've already blocked the behavior in your accounts you won't know if and when they fix it. I know they haven't blocked the luggage issue because there is still an ad for a specific high dollar brand showing for an advertiser who sells many brands right now.
I'll be adding more follow-up articles and more specific examples with full details in my blog. This is a huge issue with many facets. I've seen many of them over time. They appear overnight and for niche marketers can have serious consequences.
I completely agree that a huge responsibility rests on our shoulders as PPC managers to set up sound strategies in regards to Expanded (Broad), Phrase, Exact and Negative Match keywords.
However, I do think AdWords is shirking some of their responsibilities for improving 'search quality' when they start making Broad Match more and more ridiculously broad.
JstaTad's example above for the cholesterol website is classic. Here Google has the most elaborate, meticulous and accurate algorithms in the industry when it comes to their organic search results. Do you think a cholesterol website would ever show up in the organic results when someone typed in 'fat girls sex'? Most definitely not. But then again they aren't making money off of organic either.
I get that Google is a business and that they focus on revenue just like any other business. But I find their focus on 'Quality' these days conflicts greatly with what they are doing with Broad Match these days. Big contradiction...
Thanks for the information, flyingrose. I'm not exactly a newbie at this, and I do have many accounts of varying sizes, dating back since Adwords' inception.
I don't disagree that Google will take your money if you're not careful, but they are giving you more tools than any other platform to manage the entire process.
Don't like the broad match?
Run search query reports everyday and find more negative keywords until you have the ROI under control.
As far as the cholesterol example, can we really make a judgment about Adwords expanded broad match in this particular case until we see the entire campaign or adgroup that JstaTad was using? This is just all anecdotal evidence until someone proves otherwise.
Again, I don't doubt that Adwords is pushing the limits in the name of profits, but the tools are there to control your campaign. You just have to use them. There's a sense of entitlement to a continuously profitable campaign here that I don't understand.
ppcbook certainly has an interesting point of view and I am certainly not arguing that advertisers should be entitled to continuous profits. Google offers a service for a fee. The better that service the more profitable they will be so it makes sense for their system to operate in a logical way.
What we are seeing is anything but logical and if I were Google I would be working diligently to fix it. The core issue is that their algorithms are based on flawed logic. As any old-timer knows garbage in = garbage out.
And since the system relies on historical data, improving that logic will not eliminate the existing errors in the data. I can already hear protests about my "flawed logic" argument so I will elaborate briefly:
Assuming that clicks equals relevance is flawed logic. Here are three examples where that does not hold true:
1. Many searchers will automatically click on the first ad on any search result page. They don't read them; they just click. Some will even back up or repeat the search and then click on the second ad without reading it first. Those actions pollute the data collected for relevance. If you don't believe me test my theory for yourself. Select ordinary Internet users at random and watch them actually use a computer. I call these "automatic clicks".
2. The searcher who searches for a specific item and sees your ad for a similar product does not know whether you have what they want or not. That they clicked on your ad for that search DOES NOT mean that they will find what they want nor does it mean that you wanted your ad to show up for that search.
3. Just as content ads garner lots of curiosity clicks, search ads with interesting titles and text may garner clicks on unrelated searches. This is less likely than the first two examples but surely will happen sometimes if you write compelling ads (or foolish ones that over-promise and end up featured).
If you sell your service on the basis of relevance wouldn't it be wise to get your system to act accordingly? We know how to manually make the ads most relevant for best results and then the way Google's system functions undermines our efforts.
Much of this issue is NOT intentional on Google's part. The flaws are in the basic assumptions made. The outcome is caused by crediting computer programs and neural nets with abilities they do not currently and may never have.
My view on this is simple and it comes down to what is best for the advertisers, the Internet users, and Google:
ADVERTISERS: the better ROI advertisers can drive the more they'll spend on advertising, the happier they'll be and the more money Google will make - far more than they will by displaying the wrong ads.
SEARCHERS: showing the most relevant ads keeps Google in the favored search engine position and increases their income.
GOOGLE: the more relevant the ads the more money they and their investors will make and the longer they stay on top.
Search is the King now; however, other means of locating data including Social Networking will be encroaching on their search kingdom. It would be wise not to kill the golden goose prematurely.
@rose
"Google offers a service for a fee. The better that service the more profitable they will be so it makes sense for their system to operate in a logical way. "
I think this is the core issue / debate we're having here...I (and maybe a few others) believe the AdWords system does operate in a logical way and Google is slowly giving us more and more tools and targeting abilities so that we can better maximize our ROI.
Where a lot of folks get into trouble is using Broad Match as their default option without really understanding exactly what broad match is and how it works. I essentially use broad match as a "net" so to speak - it casts itself wide and deep and brings back some gold but there's some gravel in there too. I pick out the gold and build on it, and fileter out the gravel and use it to further refine the exclusion list. The AdWords accounts I work on tend to generate a better ROI the longer they run - it's a continuous process of adding and exclusing. If I start to ignore a campaign 9 times out of 10 it's performance slowly degrades until it's no longer profitable.
I would also add that I personally like the way broad match works. It's my "net" so to speak and allows me to build and refine kw lists without having to manually add thousands of potential keywords to find the ones that work. When I flip on broad match I expect it to push the limits in an effort to find what works and what doesn't.
@Jeremy - have you read the posts I've made on what is happening even with phrase and exact match triggered ads? Or the one about the new report seeming to indicate that even exact match ads are being triggered by "other unique searches"?
Is it that you don't believe what I'm writing or that you don't object to these issues and having to spend an ever-increasing mountain of time controlling what Google AdWord's system is now doing that used to work just fine?
Computers are fast and efficient for many things; however, "to err is human; to really foul things up requires a computer" and the AdWords system is too time-consuming to control manually. Everyone is looking for ways to automate it.
Note that Yahoo! Search Marketing and Microsoft AdCenter have barely made a dent in Google's lead. The reason is their systems are too time-consuming!
Microsoft should have easily conquered far more market share by now. My original intention was to open accounts for every advertiser I had ever assisted. Their system is so bad I still rarely use it.
The new Panama edition of Yahoo! Search Marketing should have been an improvement. Instead it is another time-pit. Again, instead of spending thousands there for many advertisers every month most of what we were spending had to be turned off because their transition was such a disaster. The consensus among the brilliant is it was easier to completely start over than try to fix what transitioned.
With all their resources why don't any of the search engines get input from truly brilliant PPC visionaries and usability experts like Steve Krug? Who ARE they having beta test these systems? There is no excuse for how horrendous they are.
AdWords used to work! Accounts could be counted on to run reliably. You could stabilize one in a week or two for a tiny business and it could run practically on auto-pilot with good results for months or longer.
When Expanded Broad Match was introduced the Google train started running off the track. I for one would prefer to put it back on instead of arguing that a train wreck is fine.
Call me resistant if you will; however, continual improvement is my path and what is going on right now is NOT improvement. Google created an awesome cash cow whose milk has benefited many. I'm not ready to just leave it be and let it sour yet.
Rose,
“@Jeremy - have you read the posts I've made on what is happening even with phrase and exact match triggered ads? Or the one about the new report seeming to indicate that even exact match ads are being triggered by "other unique searches"?
Is it that you don't believe what I'm writing or that you don't object to these issues and having to spend an ever-increasing mountain of time controlling what Google AdWord's system is now doing that used to work just fine? “
Haven’t read them all, but have read some you’ve linked to/wrote about + many others I’ve come across in my own time online.
I’ve ran thousands (that’s not an exaggeration) of search query reports since they were made available and have yet to see exact match being triggered by terms other than the actual term being bid on. So I guess that means to certain extent, I don’t believe those reports. I maintain a lot of AdWords accounts and am usually able to validate or discount things I read online based on the data in one or more of my accounts. Before replying to this thread I actually ran a few more SQ reports on EM terms with tens of thousands of clicks and millions of impressions. Nothing appears to be out of order.
As far as spending more time managing AdWords…that’s ok by me as it’s really not a surprise. In relation to managing other forms of media it’s still very, very efficient when setup and managed properly. If you’re doing it right it’s not a part time set it and forget platform – it needs near constant attention.
“Computers are fast and efficient for many things; however, "to err is human; to really foul things up requires a computer" and the AdWords system is too time-consuming to control manually. Everyone is looking for ways to automate it. “
Finding the perfect balance is tough – too much reliance on automation tends to stifle campaigns and creativity but trying to completely exclude automation drives management costs higher than they should be.
“Note that Yahoo! Search Marketing and Microsoft AdCenter have barely made a dent in Google's lead. The reason is their systems are too time-consuming!
Microsoft should have easily conquered far more market share by now. My original intention was to open accounts for every advertiser I had ever assisted. Their system is so bad I still rarely use it.”
Yea…their systems and policies need work. Not to mention quality support is missing…regardless of spend.
“The new Panama edition of Yahoo! Search Marketing should have been an improvement. Instead it is another time-pit. Again, instead of spending thousands there for many advertisers every month most of what we were spending had to be turned off because their transition was such a disaster. The consensus among the brilliant is it was easier to completely start over than try to fix what transitioned.”
Fully agree…I’ve never recovered spend in Yahoo since the release of Panama. It’s a major flop in my opinion – horribly planned, executed and supported.
“With all their resources why don't any of the search engines get input from truly brilliant PPC visionaries and usability experts like Steve Krug? Who ARE they having beta test these systems? There is no excuse for how horrendous they are.”
I’m not sure who they use but I would again agree…they should open the pool party to some new guests.
“AdWords used to work! Accounts could be counted on to run reliably. You could stabilize one in a week or two for a tiny business and it could run practically on auto-pilot with good results for months or longer.”
And we should have all been getting while the getting was good:-) The writing was on the wall from the early days that the system would get more complex and require more time and resources to effectively manage. I still view the AdWords platform as highly efficient and cost effective when compared to managing other forms of media.
By this time next year I'll bet the AdWords system will be even more complex. We're just at the starting point in my opinion.
“When Expanded Broad Match was introduced the Google train started running off the track. I for one would prefer to put it back on instead of arguing that a train wreck is fine.”
I didn’t and don’t see it as a train wreck, I see it as Google doing some more of my work (testing) for me. I know what to expect when BM is being used.
Great points and great discussion, let the debate continue:-)
Thanks, Jeremy, for this latest response. I agree with everything you've said except for Expanded Broad Match. It is good for huge advertisers and an exceptional burden to niche and smaller businesses.
There is a simple and obvious solution that would make us both happy: Google can let us opt in or out of it. AdWords is one of the most brilliant methods - possibly THE most miraculous method ever - for growing a business. Ensuring that small advertisers can generate positive ROI benefits everyone and especially Google.
That is the bottom line: if advertisers can generate a profit they'll spend more and more money there. Even the spend of the tiny advertisers adds up to millions. Google can either help them stay or push them out - possibly unintentionally.
They would do well to not allow what happened last holiday season to happen this year. We're planning ahead and if necessary many small advertisers will opt out of November/December spending altogether.
I am something of a contrarian and a very hands-on PPC specialist. Since I work only with small businesses (and help many microscopic ones) and implement strategies only used by a few others I'm aware of such as Mary O'Brien and Frank Watson, I encounter different issues than agencies would.
If you're not familiar with Mary, she used to teach the official Yahoo! Search Marketing Seminars and now gives the awesome two day ppcSummit seminars. I highly recommend them to everyone - the more you know the more you'll get out of it and even those who know absolutely nothing will learn and benefit.
You are more likely to know Frank as he is the moderator at Search Engine Watch who posts as AussieWebmaster. Frank is already here and Mary will be soon.
I'll have to agree with Jeremy that I don't see expanded broad match as a disaster. Not even close. I manage all kinds of campaigns - fortune 500 to small ($30/day). I can't find a single case of expanded broad match causing problems. My issues mostly come from domain ads and error page ads.
One thing I can agree with Rose on, a simple solution is to give the advertiser more control, let us Opt Out. Just like the domain ads issue, it will help those who 'manage' campaigns to spend even more.
Since this Sphinn I did a follow-up post on the Expanded Broad Match problem that links to twenty blogs and forum threads discussing the issues. That post is at http://www.ppcthink.com/2007/08/22/which-google-adwords-ads/ and if that isn't strange enough I started mass deleting broad match keywords with some highly unusual results.
I posted actual AdWords Account Snapshot graphs showing what happened at http://www.ppcthink.com/2007/10/05/deleting-broad-match-keywords-increases-roi/ and there are already twelve comments discussing the results.