Morgan Stanley's star internet analyst, Mary Meeker, gave a keynote at AD:TECH in San Francisco, available in PDF.
READ IT! It will give you more validation that you are in the right business for the next 10 years if you are focusing on the internet, especially in advertising and reaching consumers and businesses around the world.
¶ 4/26/2005
I'm always surprised at how many Google AdWords advertisers don't know about the
{keyword} macro for tracking the actual keywords that were typed in before the user clicked on your ad. It's simple to use. If your normal destination URL for your AdWords ad is:http://www.example.com/landingpage.htmin the AdWords UI when you set your destination URL for the ad you simply append
?kw={keyword} onto the end, so it looks like:http://www.example.com/landingpage.htm?kw={keyword}
Then when the ad is clicked, you can check your log file, and you'll see that the request portion of the log line for the click has the keyword you bought that matched what was typed into Google search right before the user clicked your ad:http://www.example.com/landingpage.htm?kw=search+term+you+boughtYahoo! Search Marketing (Y!SM, aka Overture) has a similar feature that's actually even easier to use. Go under "Account-->Account Set-up" and turn "Yahoo Easy Track" to ON. Then your logs will contain the matched and actual keywords and the match type for each click.
Update: I've corrected a mistake in the original post about what the Google {keyword} macro will insert into your request. It will give you the keyword that matched the user's search, not what the user typed in. Y!SM's Easy Track feature will actually give you both, however.
¶ 4/25/2005
One of the intriguing things about Google's test of a mutant CPM model and new site targeting features is that it is an actual beta! In other words, they are announcing the details in advance, while a small number of users get to test the new features.
Why would Google do it that way? Usually, they get it ready and then spring it on the world and call it a beta for the next several years. Pre-announcing is not really their style. Maybe they are starting to eye Yahoo's more diverse piece of the ad world pie. If that is the case, they're sending a message with this beta.
Additionally, by pre-announcing, Google is giving time for all the media planners and buyers to include them in their summer campaigns.
¶ 4/25/2005
Lost in the coverage of Google's massive, insane, fly me to $300/sh quarter, was AOL's pay-per-call introduction. Yes, of course, certain blogs mentioned it, but isn't it a big deal? I mean AOL is now doing pay-per-call!
I'll tell you when it will be a big deal: when Google does it! All that rumor/speculation over Google's dark fiber acquisition - maybe that has something to do with Pay-per-call? Maybe not.
The point is, if you want to prosper, you need to go to where the puck will be. Try Ingenio and AOL pay-per-call today, because someday the big G will have it. Then we'll find out if pay-per-call is real.
PS. The main problem with pay-per-call is currently the fact that the acronym (PPC) is already taken. Maybe Google will name it something new?
¶ 4/24/2005
Google Gmail has tacked on a 1-line RSS feed reader at the top of the inbox called Web Clips. Most important fact: you can turn it off in Settings. I found it very annoying because, unlike the contextual ads, the content is not relevant to any of your mail.
Anyone else get the sense that Google wants to "do something" with RSS? I hope they do something interesting and relevant, rather than just slap a Yahoo-like RSS reader on top of Gmail.
¶ 4/23/2005
I like Seth's response to this survey on cookie deletion from Atlas OnePoint: "Vindicated!".
For whatever reason, Eric Peterson from Jupiter (who the study that claimed 39% of web users were deleting cookies monthly) is rather circumspect about this new study. He's casting aspersion on the motivation and ethodology of the new survey.
I've looked at a lot of logs, and the data I see is a lot more in line with Atlas OnePoint's than Jupiter's results from a questionnaire. I also trust the examination of log files more than I do the questionnaire approach of first explaining what a cookie is, and then asking how often people delete theirs.
¶ 4/19/2005
And I won't spoil it for you... It's a long article fantasizing about a Google dominated future. Give it a click, if just to see the headline
¶ 4/19/2005
PPC experts Andy Beal and Andrew Goodman both comment on their blogs about click fraud. Essentially, they have a view that the click fraud meme is over-hyped, and that eventually people will get on with it, and accept it as a part of doing business. It won't "take down" Google.
I agree that is what will happen, but I think it will be a bumpy road. The reasons Google is more vulnerable than say Yahoo are:
- All of their revenue comes from ads.
- Their outward stance on click fraud is one of secrecy and denial.
- Their culture looks for technological solutions first.
In other words, managing outside perceptions is going to be important for dealing with Click Fraud. Outreach with advertisers and agencies, along with changing some of the ways the Google accounting system works will be important. A simple thing they could do is establish an independent auditing council.
In the modern history of advertising, 3rd-party auditing has been an accepted way to deal with the fuzziness of counting audiences. Even though PPC ads can be counted for in a pretty direct way, that doesn't mean that there isn't a role for auditing.
If Google wants to be proactive on click fraud, and minimize its risk to their business, they might consider independent auditing as one part of the solution.
¶ 4/18/2005
Pretty much the usual coverage on click fraud from the Washington Post, including some good quotes from click fraud maven Jessie Strichiola. The fun part of this article is the battling product managers. Compare:
"If there is anybody who says this is not a real problem, they are kidding you," said John Slade, a product manager with Yahoo.
"I would characterize the problems due to click fraud as small. We have a software system that filters out fraudulent clicks even before advertisers get billed for them," said Salar Kamangar, a Google product manager. ... "This is not a significant problem..."
At least one product manager is past the denial stage. The Post article also mentions cost-per-acquisition as a potential solution.
Update:SEW contrasts the quotes a bit differently, pulling the old quote from Google's CFO. I think he's the only one at Google who's been forthright about the potential of the problem - but then again, due to Sarbanes-Oxley, maybe he has to be. Also, Andrew Goodman responds to the WaPo article. He seems circumspect about the click fraud problem, and urges reporters to find a balanced view of the PPC world.
¶ 4/17/2005
Good clickz article covering the fact that advertisers and publishers have found a model for the advertising on internet that actually works. Publishers of all sizes are creating content to sell ads - and they're focusing on areas that have higher value. Inventory (i.e. web content that sells something) is purposefully being created:
- Targeting high value topics
- Targeting specific buyers
- Targeting specific sellers
In other words, it's a marketplace responding to demand. Sounds great, right? Well, it is great and it is a better model than the dot-com bust days. The market is actively searching out what sells and creating more of it.
The only downside is the eventual regression to the mean. Right now, it's easy money for all involved. Ad networks, publishers, search engines, agencies, SEO and SEMs are all doing well. So well that they are attracting more investment, more people, and more capacity into the market. This in turn creates more supply.
The internet scales easily and more supply can engender more demand to a point. However, eventually ROI will decrease as more money and people chase the same goals. It's a classic boom time. The hard part is predicting how far into the boom, and how close to the bust we really are.
¶ 4/15/2005
Joe Holcomb, Senior Vice President of Marketing, BlowSearch has a post that blows the lid off of click fraud. It's a long post, but it's a much needed wake-up call for the PPC industry. Here are a few of his key points:
- Competitor based click fraud is minuscule as compared to automated click fraud.
- Google is no better than anyone else is when it comes to detecting click fraud.
- Pay per click advertisers seem to feel that there is an “acceptable level” of click fraud. It’s looked at as a “cost of doing business".
- The bottom line is that the click fraud detection technology that exists today is pathetic and the fraudsters know it.
I've spent a lot of time analyzing several sets of log files from different companies, looking for click fraud. My observations lead me to believe that the overall points in Holcomb's post are accurate. The main problem is the current stance of the networks (i.e. Google, Yahoo) who are trying to make people believe they've got the click fraud issue under control. However, they aren't telling you what percentage of click fraud they think is acceptable. They are not open about what is really happening, and how much they can do about it.
This current stance by the networks - attempting to say the impact of click fraud isn't significant - only works because many advertisers are still making money. Advertisers accept the unknown impact because it's not preventing their campaigns from working. It's really not a top priority issue for most.
Click fraud may be analogous to spam. Back in 2001, spam was around but not a major issue. By 2003, IT teams were running around with their hair on fire trying to combat spam. Even though it was frustrating, email users worked through the problem, and now know how to live with it. Fortunately, the spam problem didn't have a group of email vendors trying to sweep it under the rug, the email firms didn't have a conflict of interest in combatting spam, and everyone agreed to fight it together. That's not really the case currently with click fraud.
However, while Holcomb is saying that the major search engines are dissembling and not doing enough, he does warn that some of the blame is on the advertiser side.
More than 70% of the time advertisers have no idea what their results are with their pay per click advertising. They are not tracking their ROI even though the tools are available to do it. That’s foolish on the advertiser’s part, and in my opinion, irresponsible sales and marketing by the PPC providers for letting it happen. Education is the key to success. Educate your advertiser and you have a happy advertiser.
This is dead-on correct. Even if Google really were focused on rooting out click fraud (Holcomb says they are not), that would not be enough. The advertisers have data that Google doesn't have. The advertiser can understand what happens after the initial ad click. Without that information, it's very difficult for Google to detect mildly sophisticated click fraud. Holcomb puts some burden on the advertisers to demand more accountability:
Advertisers are certainly not forcing PPC networks into disclosure with their wallets, so why should they disclose anything? If you are a PPC advertiser then you’re being complacent. It’s not the PPC’s fault. You haven’t stopped advertising long enough to force them into changing their ways. You haven’t made them accountable. I personally think that you should.
In other words: WAKE UP!
PS. I wouldn't take too much solace in Holcomb calling the competitive click fraud problem is miniscule, since I've seen it cost significant amounts of money. He's saying it's relative small compared to automated fraud. Competitor click fraud is slightly better understood and definitely easier to detect than automated bot click fraud.
¶ 4/12/2005
An example of how useful the Yahoo API can be. Get your Yahoo local results directly into Excel! Build your own list... I tried pretty hard to do this with SOAP and the Google AdWords API, and barely got it to work - I had to use .NET VISTO and it only works on Excel 2003. Maybe the REST approach is just easier...
Note that it's Excel 2003 that handles the XML lists well - even with the Yahoo API. Check that you have the right version of Excel, or it probably won't work.
¶ 4/06/2005
Via Research Buzz. Yahoo is improving Y!Q contextual / related searches. That's good. The early version was a bit broken, and Y!Q was just a nice hack, really. But Yahoo seems intent on making it a valuable tool.
¶ 4/06/2005
Eric Peterson of Jupiter Research has a blog thats chock full o' good web analytics stuff. In a recent entry, he staunchly defends the Jupiter view that cookie deletion is causing problems for measurement of visitor behavior. He also chastises those that criticize their numbers without understanding the methodology...
Basically Jupiter claimed that up to 39% of users deleted their cookies monthly. SiliconValleyWatcher.com, however quotes the report as saying that:
17 percent of consumers delete cookies weekly, 12 percent monthly, and 10 percent daily
And Eric defends the Jupiter survey (which was a series of interview questions asked to users) with some Nielsen/NetRatings panel data (which measured what happened on the users actual computers) that shows that 25% of Google users were issued a new Google cookie at least once during the month of July 2004. Eric also engages in some speculative extrapolation - imagining that techie sites like CNet probably have higher cookie deletion / re-issue rates.
I can see why experienced marketers were cynical about a survey that asked consumers about cookie deletions. Consumers have been largely unaware of cookie technology up to this point. And for comparion purposes, a recent report said that 37% of consumers couldn't differentiate between paid ads and search results when using Google and MSN. All of a sudden you expect people to believe that a significant number of these users have the savvy to delete cookies regularly.
But in looking at the data given above, and the Nielsen data released this week, it seems that the cookie deletion data isn't quite as alarming as the headlines suggest. Maybe people are strongly reacting to some of the numbers they've heard paraphrased rather than the actual data - and Jupiter's summary isn't a heaping pile of sober analysis, either:
Accuracy of Web site measurement has long been taken for granted, relying on the fallacy that data collection is complete.
(Emphasis mine). I'd give marketers more credit - most realize the difficulty in modeling user behavior with the stats they have today.
However, 10 percent daily deletion sounds reasonable, and 17 percent monthly isn't that many overall. Recall that most affliate marketers only get credit for sales within less than 7 days from the time the lead was originally cookie'd. If one out of 7 cookies gets deleted every 30 days, it's not going to impact analysis too much.
Perhaps the best thing to come out of this is awareness. Spyware detection and increasing consumer awareness means that cookie deletion is only going to get worse...
¶ 4/05/2005
Here's a way to draw attention to click fraud. Sue Google, Yahoo, TW, Ask Jeeves - all at the same time. Covered in the Wall St. Journal. Not bad.
The plaintiffs, led by Lane's Gifts & Collectibles LLC, a Texarkana, Ark., retailer, allege that the Internet companies knowingly overcharged for advertisements they sold and conspired with each other to continue doing so. The plaintiffs are seeking to have their suit, which hasn't received widespread attention, certified as a class action.
...
The search engines have antifraud systems and sometimes issue refunds for bogus clicks. But they decline to comment in detail on the scope of the problem, exactly how they are fighting it, and any specific instances of click fraud, in part because they don't want to tip off fraudsters. That has fed some advertisers' fears that the problem is bigger than the search companies acknowledge. Estimates of click fraud run as high as 20% of all clicks on search ads.
You know what they say: "Sue them all, and let god sort it out..."
¶ 4/05/2005
For all the general carping and criticism that Google faces these days (including some at this blog), they have a pretty good track record of putting new code out there. Now they've taken the Keyhole satellite data and put it into maps.google.com. That's what, a few days before the 6 month anniversary of buying Keyhole. That's a pretty rapid integration...
¶ 4/04/2005
Yahoo put up an API for term extraction. You give it a sample text and it pulls out the relevant words. This could be really handy for keyword research or SEO. You could even implement your own contextual advertising program, if you were ambitious...
One potential downside - this is the API to the technology that Yahoo uses to extract terms when you do a Y!Q "relevant items" search. When I played with Y!Q way back in February, I found that it didn't extract the terms I really wanted it to...
¶ 4/04/2005
Marketing Sherpa debunks the idea that "rich media ads are hot" - ie. advertisers using them more and more. And they get bonus points for mentioning that they were one of the outlets promoting that idea 6 months ago. Here's data from a survey about rich media ads:
Dec 2003 Dec 2004
We do not spend on this 42% 39%
Increasing spending substantially 15% 11%
Got great rich media results 34% 34%
Got moderate results 38% 48%
Disappointing results 29% 18%
In this article, they claim that rich media ads (i.e. Flash and movies) usage isn't really growing very fast. They also have a list of "worst practices" around rich media ads. Definitely worth a read - but it's only "open access" until Apr. 9.
¶ 4/04/2005
April Fools is now apparently a corporate sanctioned holiday. Is there a regulation somewhere that says "hip" internet companies have to do an April Fool's joke?
Well, if ever anything has "jumped the shark" it's these things... April Fool's corporate jokes are about as hip as Burning Man is these days...
¶ 4/01/2005
