Shocking Results! How New Google AdWords Advertisers Get Crushed (Test Results Are In)
January 20th, 2007 by Mike Read more about Google AdWords, Pay per click advertising, PopularOne of the single most frequently asked questions that we get from new Google AdWords advertisers is the following:
Where is the traffic?
Everybody knows that Google gets ton’s of traffic so why is it that lots of new Google AdWords advertisers fail?
We did a radical test last week to get to the bottom of what is going on.
We had been running a couple of very successful campaigns for one website for over one year. So, here is what we did.
We paused the campaigns on our existing Google AdWords account and created a new Google AdWords account. We then launched the identical campaigns on a new Google AdWords account.
Here are the results that we got after the first week. They are nothing short of devastating! Take a look at three Ad Groups and check out the results before and after moving them into the new AdWords account.
Our new Google AdWords account got crushed! Above I am sharing the results from three Ad Groups before and after the test. Our average daily click count basically dropped to zero and we got hardly any impressions. However, nothing changed!
In total, we moved over 200 Ad Groups across 10 campaigns and the results are the same across the board. Here are the results at the aggregate account level.
Lets remember that NOTHING has changed except moving the campaigns from an existing Google AdWords account to a new Google AdWords account. If this is not shocking, then tell me what is.
So, what can explain this?
The only difference is History!
In this experiment, everything was kept constant except moving campaigns from an existing Google AdWords Account to a new Google AdWords account. And, when you move accounts, the only thing that you lose is history.
Why is history so important?
- In Google, everything is based on the Quality Score which is a combination of how your ads have been performing over time relative to ads of the competition.
- Ads with a higher quality score will get surfaced more often and have a dramatically lower required bid minimum. When we drilled into the new Google AdWords account we saw bid minimums that far exceeded the maximum bids that we used previously. So, if your account has no history, you might have to spend extra to establish a good history by bidding high just to get started.
- We also know that ads with a higher click-through-rate will have a higher quality score and that ads that start appearing in a higher position will have a higher click-through-rate because people click on more links that appear higher on the page. So, once again, new Advertisers will have to spend extra to quickly drive up the Quality Score.
I have long been hypothesizing about a “Doom Loop” and a “Money Loop.” New AdWords users end up in the “Doom Loop” if they do not start out very strongly. Initially, it pays to overpay and overbid just to rapidly earn a higher quality score which will increase your click-through-rate and lower the minimum required bid for many of the keywords that you want your ads to surface for. By starting out strong, new advertisers can get into the “Money Loop” (vastly more impresseions, higher CTRs and lower bid prices).
However, what I never knew is that new Google AdWords advertiser get crushed if they just follow what existing Google AdWords advertiers are already doing. To displace them, it seems takes a far more aggressive bidding strategy than what the entrenched players are already doing.
I would love to hear your thoughts on this especially if you have any other possible explanations for what might explain the horrible results with the new AdWords account.
Recent posts by Mike
- What Does Your Order Button Say?
- Will Knol Rule The Search Engines? First Analysis Indicates...It Just Might
- Are small businesses dropping out of AdWords?
- How I Got on The First Page of Google With My Last 4 Blog Posts in a Row
- AdWords Conversion Tracking - The Ultimate Guide
About the Author
Prior to MindValley Media, Mike was the Head of New Ventures Strategy at eBay where he conceived of the strategies that led to the investment in Craigslist, the launch of Kijiji.com and the acquisition of Skype.
Mike has an MBA from Stanford and previously worked for the Boston Consulting Group. He is also a certified Google AdWords Professional.
Check out other posts by Mike
If you liked this article, here are some related posts:
If you want to see what's in the private lab...
Ask yourself... which part of your business would you most like to improve?
I can send you 7 tactics related to your specific needs, if you like.
You don't have to buy anything, just take it as a backstage pass into our private course.
Will it give you the solution you've been waiting for?
There's only one way to find out.
Mike Reining
Vishen Lakhiani
We experienced something similar with a client of ours. So we called our Google rep. essentially what we were told is that the new account had no trust– so they were going to gradually test the ad over the next few weeks to see what kind of response it got.
Yeah. It makes A/B testing a whole lot of fun.
This is exactly what I’ve been seeing in my campaigns.
One of my campaigns had to be recreated under a new account for billing reasons and it’s performance sucked even though it had the the campaign structure as before. Only by over bidding was I able to get the campaign back to track again.
Thanks for sharing your insights
Rather interesting, thank you for sharing.
I’ve also heard that AdSense accounts potentially perform better if the publisher is also an AdWords user.. except those were just speculations, and no test data to back it up.
So newcomers are at a distinct disadvantage then, just as a new site would be at a disadvantage in SEO.
Have you any tips or ideas for mitigating this?
Interesting. Have you tried deleting a campaign and recreating the exact copy in the same AdWords account and check how the results are like?
Also how do the results change after the first week?
I had a conversation with a Google rep about this on Friday, in fact. I hadn’t read your post (obviously since you hadn’t published it yet). We are changing some campaigns around and wanted to change the “bill to” information without changing the account around. Evidently, Google is not built to do that - instead, you have to start a new account and lose history, etc. Without going into details, I am certainly not willing to lose history on these accounts!
Google’s “company line” about this is, “The only way to get better from bad history is to set up a new account. So, for the words that have a bad history, this will be great.”
Right. How about 85% of the keywords I have kept in that campaign kick butt? So, I am going to screw these up for the 15% that don’t do as well because G cannot figure out how to change some finance stuff on their end?
One thing you might try is starting out slow and building up. Part of the history of new keywords is based on the performance of your account as a whole. So if you build up with a small list, and develop an account history, it might help lower costs for a larger list of new keywords.
My guess is that AdWords makes good money off new advertisers in this way. Their bidding system on a whole is a big money-maker at the expense of the advertiser.
It is almost ridiculous simple! You don’t have to be a rocket science!
The adword ads are posted in the right side of the page.
Every man on this planet start to read from left to right.
The adword ads are pure and simple ignored (nobody see them!)
Every visitor that come on google page look first at the first site indexed, than leave and enter on the second site, and so on till he arrive at the last site on the page. If you have lucky the visitor after he is bored after 10 big sites indexed who draw his attention …maybe he is turning head with 45 degree and see…the adword ads…Remember he is very bored and he is let his money on the first or second site indexed visited. So what he is doing next? Click on all adword ads that he is see, just for fun or maybe to enrich his culture…The adword ads today is just like “the fifth wheel at the car”.
Now, imagine what it will be if the ads will be in the left side of the page.
The CTR will be maybe triple, because the visitors will see first the adword ads and than the indexed sites. Probably with so big CTR the CPC will be much higher even for small keywords. The indexed sites will be in direct competition with adword ads.
It will be very profitable because the secondary keywords will be very trafficated and will be at last profitable.
The losers (in this actual form) are:
1. The advertisers, because the CTR in this actual form (right) is very small and expensive. Even relevant keywords have small CTR.
2. Google himself, because of this actual form they lose a lot of money. Instead to monetize the traffic they will give the traffic for free to the indexed sites and than the scraps to adword advertisers.
With the left side form they maybe do, not 1 billion $/year, but maybe 2 or more.
On my opinion is a stupid marketing mistake (almost historical) that cost both sides (Google and advertisers) a lot of potential money that could be made.
The Yahoo is in the same boat, same stupid mistake.
This is why tip #3 on my list of 11 tips for Google AdWords success is: “Bid high initially.” Now, those tips were written in 2005, before Quality Score existed. However, they still apply. CTR is still a big factor in Quality Score. Keep in mind, too, that there are actually 2 quality scores (3 if you count the QS for the Content network). There’s a QS that judges your landing page and impacts your minimum bids. Then, there’s a QS that is a factor in ad rank. The CTR is a big factor in that QS.
IOW, the “formula” for ad ranking on Google used to be:
Ad Rank = CPC x CTR
Now, it’s more like:
Ad Rank = CPC X QS
where QS is made up of number of factors: keyword’s CTR, ad text relevance, keyword relevance, etc. But, Google doesn’t disclose how they weight those factors. I believe, though, that the historical CTR of a keyword is the single most important factor. Look at ad rank this way:
Ad Rank = CPC X (keyword’s CTR, ad text relevance, keyword relevance)
Now, in your experiment, since ad text relevance and keyword relevance would have been identical, the only factors to consider are CPC and CTR. Since you have no CTR to begin with and it takes awhile to establish a good CTR, it’s crucial to bid high.
BTW, welcome to 9rules!
Like Ohad, that was my experience as well. Thanks for highlighting this. I thought I was doing something wrong!
Like Richard, what ideas do you have for mitigating this for Adwords newbies. We always tell those starting out that you can get quick targetted traffic using Adwords, but now it does not seem to be true. Help!
I’ve also read that new websites (fresh domain) fetch a higher minimum bid that established domains. A friend of mine did some testing but it’s not conclusive. Any idea?
I have not done any testing on whether the age of the domain makes any difference. However, we have done tests on “starting strong” with high bids and that can have a huge positive impact.
It most certainly also helps to start establishing a good account history. When we launch new campaigns and new Ad Groups in an existing AdWords account that is performing well they seem to do much better and get off the ground much faster than if you start a new AdWords account.
Excellent post. This is actually something that I’ve experienced also. They do take 1-2 weeks to start trusting a new account. Usually 1 week solves the problem.
This is actually nothing new for most SEMs. Google has always “penalized” new campaigns/adgroups/keywords as the need impressions and clicks to build up quality score. You will see the same thing if you add a keyword to an AdGroup and try to get into a top sponsor position or create a new ad for an existing keyword.
I haven’t seen much written about this however so it’s great that you shared this story for the broader group of small business marketers who likley have little idea what’s happening.
If you plan to keep this account active, please share how long it takes you to start getting the same results you had before. Fascinating stuff - thanks for sharing.
We’ve gotten burned by AdWords simply by following their advice. AdWords raised our minimum bid on keywords because of the quality score. However they admitted that the keyword as well as the landing page were highly relevant. Personally I think Google is turning into the next Microsoft. We’ve also posted on the subject on our blog: http://www.itsontarget.com/blog
[...] MindValleyLabs had conducted an experiment where they paused several very successful campaigns which had been running for over one year and created a new Google AdWords account. [...]
Interesting post!
What I don’t quite get through is how the average position could be so similar and yet the impressions collapse. Are we saying a new advertiser is immediately hampered by only being shown sometimes?
[...] Thanks go out to Mind Valley Labs for running this test, proving the importance of account history in Google Adwords. This is something a lot of us have suspected for a long time. Read the full post: http://blog.mindvalleylabs.com/marketing/shocking-results-how-new-google-adwords-advertisers-get-crushed-test-results-are-in/ [...]
Hi Mark,
That is exactly what Google appears to be doing. While Google will show your ad in roughly the same position, it seems that they only show the ads a tiny fraction of the time.
This is due to the lack of history because the ads and the account have not accumulated any quality score. How long it takes to get out of this will have to test and share the results once we find them.
[...] First of all, Google is pretty harsh on new AdWords account as we covered in our leading post last week where we shared our shocking results. Go back and read the Shoking Results here. [...]
[...] I recently read that advertisers should expect to bid higher at first to get desired results. This is good advice for all of you starting new Adwords campaigns. The folks over at MindValley Labs blogged earlier in the month that new advertisers “get crushed” when employing the same bidding strategies as an established advertiser, and they backed it up with test results. [...]
I am new in SEM Business.
I want to thank you for this valuable information.
pls keep rolling……
Hi all,
This post has been a real eye opener. I had earlier read something simmilar but never bothered to try it. Bidding hig initially really helps in getting better a better quality score. But how much does the CPC really decrease? Any comments?
cheers!
http://woracal.com
http://rich-desi.com
[...] If not, check out my famous post on how new Google AdWords advertisers get crushed. [...]
Hi
I’ve been working with Google Adwords for a number of years now for many clients. I’ve known that Google Adwords don’t like new accounts and/or domains for a couple of years now.
In fact, for the last year I have actively turned down work for new clients if I thought they had unrealistic expectations of an Adwords campaign. I’ve always told them the first 3 months’ money is to be considered testing and you won’t get much return from it. The next 3 months we will build on that - then from 6 months on you will be able to achieve the high returns you need.
But I was finding even when I explained it like this I found they quite often started a campaign only to suddenly panic and cancel the whole contract 2-8 weeks later. I didn’t want to say they would be disadvantaged because of the discussions trying to explain that would get into.
Google Adwords is promoted as instant traffic, but as you know you need to really work and analyse the results these days to do well.
I now mainly work with clients who want to improve their results, not fresh-faced businesses who have all their hopes pinned on the campaign from day one. As an independent I don’t have the financial backing to be able to educate these people, nor the time I need to spend explaining the results. With a small budget I need to spend my time on optimising and improving. Many start-ups are wanting to spend only £50 per month and that doesn’t go far, especially if they aren’t getting results.
Sam
http://www.real-work-at-home.co.uk
Guys, I’d love to see you do this without duplicating a campaign that’s still in the system… because I have launched dozens of new sites with new accounts in the past year, and we never have a problem getting into rotation. Those click-through rates are appaling, BTW - maybe you should try some ad testing to bring it up.
I have been studying adwords for a few weeks from the Google Learning Center and other sources. There is so much information to grasp. I had been think about just jumping in and learn as I go. After reading about new advertsers being crushed and seeing the statistics I will not jump right in but keep on reading and learning. You see I do not want to be crushed. What is a proven method to determine quality score?
I have a few questions:
1) If you bid high enough to get into the top 2 or 3 positions, I find that the CTR goes through the roof and I spend 5X more per day than normal, often times exceeding my budget. Does it matter if you set the campaign to spread your budget out evenly over a day vs. using the acelerated method? If not, couldn’t a new advertiser use a crazy high bid amount ($3.00 per click) and set their daily budget to $50 and spread it out over the day, this way you get very high CTR’s but never spend more than $50 per day, and then at the end of a week or two, you can get as much traffic as you want?
2) I was experimenting one day by increasing bid amounts by 1 cent, several times per day. By the 3rd day, my traffic increased by ten fold. My conversion rate was low and my budget draining, so I lowered the bid amount and then the next day all the traffic ws gone. I dont know if it was something I did or if a competitor dropped out for the day and I got their traffic. Perhaps Google saw all of these separate bid increases as a good thing (there were probably 15 increases over 2 days) and the increasing CTR and AVG position, and then it decided to open up the flood gate?
Yes, you can start out as a new advertiser by limiting your daily max. spend and entering a very high max. bid value. That way you get a top position but avoid bleeding through money.
In fact, that’s what I would recommend new advertisers should do. Then, tweak your site and ad copy to maximize your conversion rate increase your ROI so that you can afford to bid more and increase your daily budget. Either way, there is no way around having to start with high bid prices to help a new advertiser break into the market.
Our strategy is also to bid high to start with, then to refine bids down as clicks come through and ad quality builds in. If you are starting out with several ad groups, you can switch your ad groups on stepwise, so you budget is concentrated on only a few ad groups that are in startup mode at any given time.
We also try to switch from braod match to phrase match keywords as soon as possible, since phrase match keywords usually have better CTRs. The AdWords “Search Query Performance ” report tells you the actual keywords that are triggering your ads, which helps to identify good phrase match candidates.
I’ve heard an explanation along these lines that get more to the dollars and cents of everything….. Billing. In the end google wants to makes sure that you’re going to be a consistent on time paying customer. You notice is it when opening a new account. Whereas my one acct gets billed once a week for a few hundred dollars, the new account got billed every few days for tens of dollars….
[...] Перечитывая Ижевского бомжа наткнулся на коротенькую заметку, которая почему-то не получила должного отклика. Возможно потому что он поленился первести полностью материал на который ссылался, хотя тему в дальнейшем раскрыл хорошо. Придется восполнить этот пробел, тем более что тема врядли потеряла свою актуальность. Итак: [...]
[...] Welchen Impact die Account History tatsächlich hat, zeigt ein Experiment auf dem Blog MindValley Labs. Eine Kampagne aus einem älteren Account wurde 1:1 in einen neuen Account kopiert. Bei gleichen maxCPCs ging der Traffic von durchschnittlich 1006 Klicks pro Tag auf durchschnittlich 29 Klicks zurück. Ist das in diesem Ausmaß tatsächlich gerechtfertigt und sinnvoll? [...]
[...] Here are some actual test results. [...]
I’m curious as to whether or not any of the keywords were in review or not. It seems quite odd that the impressions dropped even though the average rank remained at about the same level. I can’t see a drop from 5 to 6 causing a drop in impressions of several thousand. I have seen similar situations where the impressions dropped because the ads weren’t being served out due to the fact that the ad groups / text ads were under review by Google. Considering that this was a set of campaigns that just went live, and a lot of them, wouldn’t it be plausible to assume that Google sandboxed it in some way?
Secondly, to complete this experiment, did you then raise the bid price and see the rank climb back up to 5 from 6, etc? If so, then did the number of impressions increase once again? Did you do a visual check that your ads were appearing where Google said they were or is the rank value what you picked up from AdWords?
The domain aspect is a myth.. I’ve proved it over and over again..
Brand new domain means absolutely nothing.. You have to have keyword history in the same campaign you’re running the new domain in..
Try it yourselves!!