The collection and storage of personal information, both with regard to statistical demographic data and with regard to search history, has been of concern to privacy advocates for some time – but was brought into sharp focus when the government asked for URLs and search data this January. Here is an article on the initial subpoena

This request was part of an attempt to begin enforcement of the 1998 Child Online Protection Act, but sets a precedent which could be applied to the enforcement of other laws. In this case, the government did not request personal data, nothing that could link a particular search to a particular individual - but there is nothing really preventing them from doing so, or from search data being requested in a criminal court case. And as tools like Google Desktop, toolbars, and MSN’s Open Platform become more widely used, the degree to which search engines are integrated into people’s personal lives will increase.

Before you dismiss this as irrelevant to you - don’t be so sure pornography laws don’t apply to you – the recent case with photographer Barbara Nitke established that the most conservative communities in the US can determine the standards by which your Website is judged. Are you sure that your online maternity clothing store won’t be considered obscene by a community in Utah, even though you are in New York? But search information could easily be collected in other types of cases by the government – such as national security. Such a possible use was specifically mentioned by the government. Do you sell anything which might be used by terrorists? That’s pretty broad – anything from books to box cutters could qualify.

But in regard to SEO specifically:
The demographic information that helps you find your customers and clients, can also be used against them. As people become increasingly aware of this, they will become more cautious about actively protecting their privacy. They may become more hesitant to reveal personal information or sign up for memberships, wary of efforts to collect information about them, however harmless the intentions. They will clean off cookies and clear caches more often – making data collection and statistical analysis more difficult, The most sophisticated and wealthiest visitors to your site will be the most difficult to track.

In short, being aware of how the law effects search engines can help you understand and keep in touch with the concerns of your customers and clients. You will need to establish yourself as trustworthy, and secure – while at the same time being open enough that search engines can find you. It’s a balancing act.

A lot of people see the ruling regarding the subpoena as a victory for privacy. Google was only required to comply with it partially. However, the judge pointed out, information collected by search engines with regard to web surfing and search behavior is not protected, private information. What he did determine, is that Google’s users had enough of an expectation of privacy that to demand search data would risk their reputation in the eyes of users, so that it placed an undue burden on the company to provide info that was available to the government through other means. This was not a right that was upheld, but a commercial interest. Which is still a good thing, but not quite the same.

A .pdf of the actual ruling is here.

It's the most common error I see with clients who have been running their own PPC campaign, and the bigger their budget, the more likely they are to make it: They want to be #1 - and they want to be there for anyone who might possibly be searching for what they offer.

I can't blame them. It does intuitively seem like being in the number one spot would be ideal. It does, after all, generate significantly more traffic, and often a high click-through rate. Just think, if you have enough money, of all the people you could direct to your site if you can afford to be in the number one position for your most general term! An auto dealer that's #1 for "car," or a home loan site that's #1 for "mortgage" - assuming they have enough money - surely would benefit tremendously from this kind of placement.

And because of this, it can be difficult to convince a client that their favorite position, as sexy as it seems, may just be the least fruitful. It may simply be scattering their seed on barren earth - and the results it produces may be so costly that they take away resources from more fruitful areas of the campaign.

It's easiest, perhaps, to explain why a general term like "car" is not the best term for anyone in a PPC campaign. Simply put, the percentage of people searching on that term who are looking for what you offer is too small. Even if you rule out, with your ad copy, those who are looking for a review or automotive history - there's still far too much variation in what people mean by "car" in their heads, which didn't make it into the search bar. Even if you do have dealerships across the country, chances that you offer every make and model of vehicle are slim.

People typing in general terms are also generally early in the buying cycle, and doing research. They don't know yet quite what they want, and are unlikely to decide on this visit to your site.

A little harder to grasp is why the number one spot itself is rarely ideal. I guess the main idea to get across is the behavior of searchers. People type in their search - then click on the first result, often without even reading it. Only after getting to your site, and costing you a click, do they stop to notice that it's not what they had in mind at all. People who click on the results further down are more likely to have actually read the ad, and be thinking about what they want before they click. In other words, lowering position, while it may reduce the number of clicks you get - also is more likely to let your ad copy do its job to pre-qualify those clicks.

A recent example - I took a client from the top spot on most of their keywords, costing them several dollars per click, and lowered them to #2 and #3. They saw their cost per click drop nearly in half - but were uncomfortable about the fact that the total number of clicks also dropped dramatically. However, those who clicked were far more likely to remain on the site - with half as many clicks, their page views went up from 37k per week to 40k. I have had several clients who found that once they moved down to the 4th - 6th position for most terms, their conversion rate went up. Yes, they were getting fewer visitors overall, but a higher percentage of those visitors were making purchases - and at a lower cost-per-click, the cost per acquisition went way down.

One other danger of being in the top spot - especially for a general or popular term - is that this spot is more likely to be a target for click fraud. It's simply easier for affiliate-type fraud, to click on the top spot. And for competitor-type fraud, well, everyone wants to take a shot at the big guy.

There are some cases, however, where being in the top spot makes sense. That's for the company's own name or brand. Potter Barn could benefit from being number 1 for all terms with "pottery barn" in them. It's not always as cost effective as a lower spot, but it is good branding - and if someone is searching for a specific brand name, especially if they offer their own product for service on their website, chances are good that they will be happy to end up on that company's website. I would, however, put a cap on the CPC for those terms, as it's not unusual for someone to try to make it prohibitively expensive for a well-known company to buy their own brand name, by making their max bid absurdly high.

But it can be scary for the client to let go of that top spot, especially when they see traffic dip - so just make sure you have your tracking in place and can demonstrate the benefit in terms of the bottom line... and be prepared to move them back into the top spot if, by some fluke of fickle search behavior, it turns out #1 really is the most profitable place to be.

Yahoo announced that they will no longer allow bidding on competitor's trademark terms. This announcement has generated a little bit of confusion as to where Yahoo stands on allowing trademarked terms as keywords. I was asked, for example, whether this meant that agencies would have to start getting permission from clients to use the client's trademarks. Not according to my understanding of it.I feel qualified to answer this - since I was an editor at Yahoo, and was on the special team that was dedicated to reviewing trademark terms. As it stands, on Yahoo you can advertise under any trademarked name as long as you have content on your site supporting it – such as a comparison of your product to your competitor’s. There is nothing in the change which indicates they will not allow advertising under the trademark name associated with the website doing the advertising. This is designed to be sure that Geico can’t advertise under the keyword “progressive” – even if they have a page of facts comparing themselves to Progressive. This is designed to protect the interests of big name corporations who don't like PPC because competitors can use it to show up in search listings when a potential customer is looking for them. Geico, whether directly or through an agency, will still be able to advertise under “geico.” The place that trademark terms get sticky is on Google – where you can buy the term, but not put it in the ad unless you have written permission from each trademark holder – and while that is meant to prevent this kind of competitor advertising, often it ends up with situations where someone selling a product can’t use the name of that product in the ad. For example: MLB (major league baseball) is copyrighted. AcmeTicketReseller.com sells MLB tickets, but can’t say “MLB” in the copy. Or AcmeShoes.com - they are unable to use “kenneth cole” in the copy, even though they are selling kenneth cole shoes. If a shoe site sells 200 brands, getting written permission for each is somewhat prohibitive. These hardly seem like the kind of uses of trademark terms which threaten their integrity - since they make the trademark holder money, but that's how it's handled. On the other hand, on Google, Geico *can* advertise under the term “progressive” – as long as they do not say “progressive” anywhere in the ad copy or imply in any way that they are Progressive. Now, this seems to me to be exactly the kind of thing that the big corporations want to prevent - their direct competitor showing up when a user is searching on their brand name. Yahoo does not restrict trademark terms in this way – and it sounds like the limitation will ONLY effect competitors, not resellers. This has actually been a change they’ve talked about for over a year, as it was under discussion while I was still working there. The lawyers were evaluating it while I was there, trying to balance the choice between free speech and attracting large-name companies’ business. I personally would have preferred to see them lean in the other direction - as some legitimate consumer comparisons will end up getting shut out by this. (addendum - per Yahoo, sites that offer a comparison can still advertise under a trademark term as long as they don't sell a product that competes with this... this will mean that some sites will get sneaky and try to hide their relationship to the product sales - but should protect legitimate consumer comparisons - BUT it means they will rely on editorial review heavily for these decisions, and editors, being human and having quotas to meet, do make mistakes.)

Next week is “Black Family Technology Awareness Week” – a much too long way of saying, pay attention, race is a part of the internet. The week is promoted as a way to “provide Black families with technology access and training, and to promote the importance and value of technology in the educational and career preparation of Black youth.” It is a recognition of the vast economic influence of black youth, particularly in a time when Hip Hop and African-American trends are gold in media and advertising, but also of the fact that overall, Black Americans trail whites in Internet usage and access. Internet use by the Black community trails that of non-Hispanic Whites by nearly 20 percent, and those who do have access are less likely to have broadband – something to keep in mind if you are developing a site which targets these consumers. There are several portals and directories which are focused on African-Americans – most of which simply offer up Google results wrapped in African colors, but some of which, like the small searchblack.com, have independent results and focus on black owned and operated websites. On this site, the most popular search is for hair salons – people looking for places that specifically work on black hair. Sites like this, whether for Blacks or Hispanics or other ethnic groups, are a useful niche for websites which offer something that is truly targeting a need of that community, but not as useful for general sites trying to capitalize on a possible overlap. One of the most well-developed directories is BOBO Business, which lists black owned and operated businesses online. They strengthen their own presence by writing press releases for members. Another way to leverage ethnicity on the Internet, is to keep in mind that you have a culture yearning for a voice online. Sites like MySpace allow independent musicians can get their work in front of potential fans. Blogs create a political, and marketing, sphere of their own. After hurricane Katrina, the dissatisfaction with the governments response, and the feeling that prejudice played a factor, was expressed in blogs and chain e-mails all over the net. One popular post included many photos, and a song, and a message of support – saying the government may not care, but we do. Viral marketing may operate in a similar manner – a musician can have a MySpace profile, and a website, and encourage fans to e-mail others about them. If the message is passionate and exciting – that e-mail may end up being circulated for months. Just the other day, I got one forwarded to me that was originally an April Fool’s Day prank last year – and it included a link to a post on a website. If your client provides a service that targets African-Americans, or any ethnic community, remember that there are ways to specifically market to that group on the internet – from specialty search portals and directories to blogs to targeted e-mail campaigns. No, this is not something that applies only to Black websites, but to any business. Race and ethnicity are a factor on the Internet. Remember that, and find the resources that you can use to your benefit.

AOL sure knows how to throw a party. Whereas MSN’s recent lunch looked like something that was planned the day before (sorry MSN, but it’s true) – AOL had a premier venue, first rate gourmet food, and a polished, exciting multimedia presentation that was on topic.

Zunch, along with other major online advertising firms in Dallas, was invited to a lunch with AOL last week, where they were rolling out the new programming, talking about advertising opportunities, and oh, sure, addressing that Google thing.

Yes, we were mostly interested in that Google thing. And addressing it was something like this: “We know there’s an elephant in the room, and everyone wants to talk about it – so before we get to the meat of the program, we will. Yes. It’s an elephant. Next topic.” All they did was acknowledge that there is indeed a partnership. But the presentation wasn’t about search – it was about programming.

I haven’t paid much attention to AOL in a few years. Now, however, it looks like AOL has grown up. It’s opening up programming to non-subscribers, and embracing interactive media in exciting ways. The concept of marketing fully integrated into programming has never been closer to being realized. And well, it’s hard to argue with numbers like 112 million new uniques a month on the network

For example: you go to an AOL channel, like fashion. On that channel, they have programming that is determined in part by user feedback and preferences. They run user commentary. The page has spots for advertising on the top, on the screen playing the video program, within the video program… you can skin the whole thing… taking over the design of the page for a limited time while the program runs –putting your brand name in the face of the viewer… and they even have the ability to click on the image in the program and be led to sources to buy that product. Yes, this latter is limited right now, but has tremendous possibility.

Other opportunities include blog marketing and product reviews.

With so much programming, you have channels to reach several specific demographics. They focused on the following, but there are many more:

• Music (with unique concerts and unplugged shows as well as videos, mp3s and artist information – much like Yahoo Launch) – with more live music programming than any other site, they have the potential to draw a lot of repeat visitors
• Celebrities – entertainment news and gossip, the kind content that gets people talking
• Television – the closest thing to true TV on demand that exists. Choose which programs you want to watch, when, from Warner Bros. inventory – with interactive features like information about the actors and the episodes
• Gaming – reviews and clips of video games and platforms, a great place to reach young males with disposable income
• AOL news – where people spend a remarkable 28 minutes per visit on average. A good potential venue for more sophisticated advertising, perhaps – messages that are more complex than the usual eye-blink a user spends taking in a page
• Movies – not only the basic movie information look-up, but fun programs like “Unscripted” where celebs interview one another using a combination of viewer’s and their own questions
• Life Coaches – aimed largely at a slightly older and mostly female demographic, a group that is known to have powerful brand loyalty once you win them over
• Professional Blogs – not just the personal blogs they have on their main site, but some of the most well-respected and often visited blogs out there. Blog marketing is just beginning to develop – and there is a lot of potential for businesses to utilize this type of venue
• Fashion – some of the best usage of interactive marketing here. Those shoes the model is wearing? They cost $346 at Nordstrom.com
• Kids programming – get ‘em while they’re young! Advertise on websites with content parents trust and kids enjoy

While much of the type of advertising they offer is currently outside of our scope, I certainly see the possibility for Zunch to grow in this direction – especially as the difference between in-programming spots, skins, blog marketing, and search marketing continue to be blurred.

Yahoo has decided to cut the length of their creatives down to the same length as Google - 70 characters.

The strongest advantage of this, is that it will allow advertisers to run the same ad on both networks, giving them the opportunity to more directly compare results and to present a more consistent message.

The biggest disadvantage is that Google's ads are too short, particularly considering that Yahoo has editorial rules in place that will make it even more difficult to maximize that small amount of real estate.

Yahoo's longer ad length allowed the advertiser to more clearly qualify the click - presenting a better idea to the user of what lay beyond the click at the advertiser's site. I never thought that anyone read a full 190 character ad, but relevant bits would jump out to their eye. A 70 character ad scans more easily, and can be absorbed in less than an eye-blink - which is as long as you've got to get your point across anyway before a user moves their eye on to the next item on the page or navigates away.

But there's a huge difference between the two. 100, 110 characters would still allow for a little more messaging, while keeping the ads concise both verbally and visually.

Yahoo has also stated that they are not changing any of their editorial rules, which makes no sense with such a radical change. You must include the entire search term in an ad – shoot some of my search terms are that long. You can’t use ampersands, or abbreviations. We aren’t allowed any of the neat “tricks” that allow us to squeeze a little more information into our ads as we can on Google.

Yahoo also doesn’t allow advertisers to run multiple ads simultaneously, which puts them at a disadvantage. Advertisers will use Google to test creatives first, and apply what they learn from Google to Yahoo, secondarily.

So, this is going to be a little bumpy at first. Until editorial standards are changed to more closely reflect those at Google, it still won’t be a simple matter of using the same ads and being able to compare results from one engine to the other. Ads will still have to be re-written, sometimes dramatically. Still, I think that shorter is better in the online world.

(Not that you’d know it from my blog posts!)

article here

Not everyone is a good candidate for PPC. It can be hard to learn when to turn a potential client away.

Reasons to tell a potential client not to do PPC:

1. They don’t have a completed website.

You can’t develop a decent PPC campaign without a website. If you are building the website internally, then you may be able to work with the developers on it and actually tailor the site to make a smoother PPC campaign. However, if all you have to go on are the client’s descriptions of products, it’s difficult to assess the feasibility of PPC. The client’s descriptions may be overly optimistic. They may describe something as a “gourmet gift basket” when it’s just sausage and cheese.

2. The product is a commodity, or otherwise not well differentiated or branded.

If there’s nothing outstanding about the product – whether it’s better or cheaper – it’s damn hard to sell it through any advertising medium. But if you have Acme Little Guy selling Acme way-cool widgets – and they are up against Big Guy Widgets, who is one of the names everyone thinks of for widgets, and up against Discount Widgets who are cheaper, and Universe of Widgets that has more kinds of widgets than have any sense existing - well, the chances of diverting some of that traffic to Acme are slim. You have to have something unique. This is a problem many internet retailers have. They see that there are many companies doing well in a vertical and think that they can jump in the market, without finding their own niche.

3. The site has no focus.

Any business needs to have a clear focus. People want to know that they are at a hardware site, or a gourmet food site. They don’t like to go to a site looking for widgets and also see caviar. The widgets and the caviar might both be quite good, but it throws people off. They don’t have confidence in the product. Yes, some big names out there have a broad focus, but they almost all started small, with a narrow product line in a brick and mortar store and built a name for themselves.

4. It’s a poorly branded product where the competitors are well branded, or in a vertical where people are looking for a brand-name.

This is partly the issue of commodification – as in the case with Big Guy Widgets in #2, but also a case of brand loyalty. Women, especially, are brand-loyal, and will rarely jump to an unknown brand. Things that are tied to traditions, as well, such as holiday or religious items can also face this issue. You are unlikely to get people to try a new brand of summer sausage for Christmas, when most of them buy one from Hickory Farms for their dad every year. To change people’s buying habits requires a strong branding campaign. PPC can have a role in this, but it’s not going to be a big money-maker at the outset.

5. There are no good, specific keywords which designate the product.

That is, the product is hard to describe, or not quite what people usually mean when typing in the relevant keywords. A good example of this kind of thing is gift baskets. A gift basket contains so many different items, that to get specific with search terms can be quite difficult. People who are looking for “smoked gouda” are unlikely to want to purchase a gift basket in which gouda is one small component. Another example is new technology. A combo widget-thingamabob might be a fabulous idea, but if few people know they exist yet, then few people are searching for them. People searching for “widgets” and people searching for “thingamabobs” are going to be difficult to convince to buy a combination item.

What is a good candidate for PPC? A well-branded, discount product with a unique benefit… that comes with something free. People love free.

OK, so not many of those come through the door. I’ll go into more detail about good candidates next week.

What factors make a client a good candidate for a PPC campaign?

The ideal client offers well-branded, discount product or service, in a wide selection, with a unique benefit… that comes with something free. For example: Nike, promoting their latest shoes with a new patented suspension system, in more colors than are available in stores, on sale – with free shipping.

No, not many of those come through the door – but each of those elements are what is needed to set a campaign apart.

1) Well-branded:
It doesn’t matter if you’ve heard of it, only if the target client has heard of it. If you don’t buy widgets, you have no reason to have heard of Acme Widgets… try doing a little research. Use the internet, and ask your uncle who buys widgets for his watchamacallits. They don’t have to be the biggest name out there, but if they are a name people know – that gives you the strongest possible advantage over other advertisers.
In a PPC campaign, unless the brand name is not known, or is similar to a common non-branded search term (for example: if the company’s name was “red tomatoes” and they sold make up,) then we can expect to see click-through rates of over 10% and conversions of over 5% consistently on branded search terms. One of our clients saw click through on their brand name averaging 40%, and conversions averaging 15% on that term. This is a simple matter of making sure that the client is there, at the top of the page, when people look for them.

2) Discount:
Comparison shopping is easy online. If you offer the same or similar product as anyone else, you better be sure you are selling it cheaper – or at least for the same price. This is especially important when the competitor mentions price in the ad. This is more important in women-oriented verticals, and verticals where the product is widely available. The client doesn’t have to offer the lowest price on everything, especially if some of the other factors mentioned here are going in their favor (especially brand and uniqueness,) but they should at least appear to be competitive.

3) Wide selection:
Again, comparison shopping is easy. So is finding exactly the color/style/size or the precise service someone is looking for. If your client doesn’t carry it, chances are, someone else does. A wide selection, or variety of services, allow you to not only advertise that selection in the copy – but to use more tail terms, and get broader coverage.

4) Unique benefit:
Are your client’s widgets the shiniest? Do they come in more colors? Do they widge more effectively than any other widget on the market? Are they highly rated by critics? Bestsellers?
In short, as an advertiser, you should always be looking for what it is about this product that sets it apart. Exclusivity is almost as much of a sales driver as brand or selection – so if your client carries something that can’t be gotten from anyone else, you can capitalize on that.

5) FREE!
People love free. Whether it’s a free pair of sunglasses when you order tanning lotion, or a free night’s stay when you book at least 2 months in advance… but the best offer of all is free shipping. Free shipping is the single biggest reason why a user will click on one listing rather than another. So if it’s at all feasible, try to talk your client into having a free shipping offer of some kind, and be sure to mention it in the copy.

Earlier this month, Tony commented briefly on a news tidbit that there is a class action suit against Google for overcharging the daily budgets. (Original article here.)

Now, I know this is a shock – but I’m gonna come to Google’s defense on this one.

Yes, at first, for a few minutes, I found it a little confusing. But it took about five minutes to understand that the daily budget is an average. That during the course of a month, there will be some days a little over and some a little under. That if there is significant overspending, Google will refund the charges – which means the client has gotten free clicks, since it’s not like they can take back the traffic.

The suit alleges that there are instances when campaigns have spent as much as 160% of the daily budget and not gotten a refund. I’d venture to guess that these same people had days that ran underbudget – and they are expecting a refund for each day they go over, when that’s just not how it works.

Google does a much better job of keeping the daily spend close to the daily budget than Yahoo. You can set an average daily budget on Yahoo and still get days when it spends 3x that amount, and as a result shuts the campaign off completely on other days. Though really, that just seems like poor management to me. It would be in the search engine’s interest to keep those ads running a bit every day, at the very least because this would make the clients happier and more secure.

I’d guess that there may be technical limitations to how accurately the engines can keep a campaign under a specific budget… how fast is the data processed, the costs updated, and then the campaign shut off? In the time it takes for that to happen, how many more clicks will be incurred?

It’s not like Google make a secret of how the budget works. If I advertise with a newspaper and don’t understand their written policies – can I sue them? Seriously – do people get to win a lawsuit just because they’re dumb? Oh, wait… (remembers McDonald’s hot coffee warning) sometimes they do.

I see it all the time… especially with retail sites. Landing pages that may be optimized for the keyword, but are not thought through as entry pages to the site. That is, the product is there, with an engaging, keyword-rich description. The price is there, and a button to click to buy the product. But the page is missing the incentives that are advertised on the index page.

Incentives – whether free shipping, or a sale, or a free gift with purchase – often make the difference between a conversion and a bounce. When I was at Yahoo!, they presented us a study that showed free shipping as the number one reason why a customer would choose one online retailer over another. But it’s not enough to put the offer in the PPC copy.

Some sites bury their incentives, mentioning free shipping only on the shipping info page. Or mentioning a sale only on a category page. But most of our clients do better than that… even if it’s only at our recommendation… and make special offers a bold presence on their index page. This does no good, however, when the potential customer is entering on a product page.

If you have an incentive – make sure it’s visible on every page of the site. Look at every possible landing page as the entryway to your site, and remember most people will not go any further than that page. Keep those incentives in front of your visitors, no matter where they go on your site. Catch ‘em, convert ‘em and keep ‘em – even if they walk in through a side door.

I have a radical proposition: Get rid of insertion orders for PPC advertising.

Do IOs really serve a purpose? Or are they simply an extraneous carry-over from print advertising? Someone in an accounting department would probably be able to explain to my why this document is necessary… but from the point of view of a pay-per-click advertiser, they are a nuisance and a potential liability.

What information is in the IO? The length of the campaign, the budget, the client info, and the payment method. All of these are defined when you set up an advertising account online. Perhaps this is seen as a kind of protection for the advertiser, to make sure that they do not keep their ads up longer than intended or spend more than intended… but defining and changing the date or budget parameters is simple online. Even setting a max spend not to be exceeded, or an end date…

If it were not so simple to control these factors, or if, as in print advertising, there were a long lead time – so that you can’t cancel an ad at the last minute – then insertion orders would make sense.

OK – so yes, it’s another piece of paper that requires a few e-mails back and forth to obtain and get signed. One more thing to fax back. Pretty minor nuisance. How can it be a liability?

It becomes a liability when the Insertion Order runs out, and the campaign stops – and the advertiser does not realize it immediately.

Oh, wait – maybe that’s more a customer service issue. Maybe the real problem is that customer service reps do not send out notices that an account is about to be shut off because the IO budget or end date is reaching an end. Sometimes they do, sometimes they don’t. Or the customer service rep does not assure that a new IO is prepared in a way that will leave the account with no downtime.

Uh-oh. Is this turning into another Google rant?

Yes and no. Neither Google nor Yahoo has been consistent about keeping me apprised of IOs reaching either budget or time limits.

Yahoo has been better about letting me know when one is going to end and needs to be renewed, but not so good about noticing that an account had nearly spent its limit, or was spending too fast.

Google has been wildly inconsistent.

For example. One client has three accounts. The IOs are all set to end on the same day, but they are burning through their budget at different speeds. The rep I work with lets me know, about a month in advance, that this is happening… and that one of the accounts will reach the budget limit early.

I am happy they have done this, and ask that we go ahead and complete new IOs.

I hear nothing back for a week. I am told there’s no rush. I get passed between two account reps, who do not share all of the information I communicate. One IO gets completed, but with a start date that leaves the account with a day of downtime. The others get extended – so that now, instead of three accounts with a single end date, I have three with three different end dates.

I hear nothing further, until I notice that one of the accounts is not running. My client loses money. Google loses money. Why did I not get a call or an e-mail? And why, when I did hear back (two days later, after I followed up again, from a *third* rep), did they act surprised that I wanted a new IO?

This is not the first time Google has allowed a campaign to simply end without following up.

Oh, wait… I guess it did turn into another Google rant. (Hey, don’t get me wrong. The folks at Google are really quite nice. Just not so well organized.)

But seriously, why is all of this needed in the first place? If an insertion order really is a necessity, why can it not be generated easily, by the advertiser, online? If an electronic contract is insufficient, the info could be autofilled from information already in the account, printed, faxed back… Take out a few steps and simplify everyone’s lives.

Have you ever wondered who those “editors” are at Overture/Yahoo? What exactly do they do? Why do their decisions seem to be inconsistent.

Well, I can tell you. I used to be one of those editors.

Here’s the basics:

An editor receives a full week of training when they start, mostly on the guidelines and interpretation thereof. Some of them are cut and dry, and any editor will make the same decision (unless they are just being sloppy.) Like the back button. It’s got to work. If someone goes to your site from a PPC ad, and sees that it’s not what they are looking for – they must be able to easily return to the listings using the back button.

Some of the rules require interpretation. This most often is an issue with content. Let’s say you want the keyword “tire” and your site sells rims. Is a rim a tire? Is it a tire part? The editor gets this in front of them, and they make a judgment call. My judgment call may be different from the editor sitting next to me. Is someone who types the word tire into an engine looking for rims? In fact, might decide one way today… and (having forgotten about seeing the listing before) reverse my decision a month later when presented with the same scenario.

Now, here’s where it gets fun… The editor may turn to their neighbors and get an opinion from one, two or three others sitting nearby. We did that all the time. Some editors more than others. Some editors are inclined to be more lenient. Some are more strict to the letter of the guideline. Some think more like the advertiser, some more like the searcher. And sometimes, at the end of a long week, when we are rushing to meet our quota (yes, they have quotas for the number of listings an editor must process) – we get sloppy and don’t pay as close attention. We miss content, we forget to check the back button. Listings get declined when they shouldn’t or approved when they shouldn’t. Not often, but they do. (I prided myself on my exceptionally high quality scores, with a less than 2% error rate.)

The most common reason to decline a listing is probably obvious path. You have a lovely home page, and want all visitors to your site to go there – but how do they find the exact thing they were searching for from there? If it’s obvious (shown on that page, or listed in a menu), then you’re good to go. If the editor doesn’t see how to find it, then a searcher is unlikely to as well. Unless of course, the editor doesn’t know that the keyword you want, “dubs,” are large rims, and thinks it has something to do with voiceovers… and then, even though you have 22” rims listed in the menu, they might decline. I’d hope that they’d take a moment to figure out why you want the term (asking, doing a web search) first, but – sometimes that quota looms and finding answers takes time.

Here’s the thing about editors. They’re only people. People are inconsistent, though we do our best to be otherwise.

That is, when your listing is reviewed by a person, and not a program.

Another source of inconsistency is a little thing which editors aren’t supposed to talk about called “E2.” These are listings that are deemed “low risk” and processed automatically. Low risk listings meet the following criteria: they are low-volume, they do not contain any brand names (at least none that are in the database), possible drug or medical terms, gambling, or adult. Sometimes a listing can be flagged because there is a word in the description that the system can’t interpret according to context… like “bet” and then it will be routed to an editor.

You see, the volume of listings has increased so rapidly that there is no way for editors to keep up with them all… so E2 processes many of them. And, being a program, it has limited capacity for interpretation. What’s more, when things get really busy and the editors are all working overtime, the E2 program might get tweaked to be less sensitive, and allow more terms through.

But call me biased (OK, yes, I admit it) I prefer this system, even with its flaws, to Google’s… Because I know when my listing is submitted that it’s either online or not. If it’s declined, I can make a change and resubmit. With Google, I may have a listing live for months, only to one day get an e-mail saying that the ad has been shut off because of a guideline violation… You get a higher quality ad, most of the time, when it’s been proofed by an editor before it goes live. (Some of the ads on Google are such a mess, I’d be embarrassed to use them to represent my business.)

So if you get something declined and don’t understand why, maybe it went through E2. Maybe it went to an editor who was tired. Most likely, on the other hand, it simply didn’t meet the requirements. Keep in mind, most of them are there to improve your click through and ROI – because doing so is in Yahoo’s best interest. You getting traffic is what makes them money. You making money is what keeps them making money. The better you know the guidelines, the better you can make your ads work for you.

Google’s Smart Keyword Evaluation Tool – what exactly are the advantages, to the advertiser, to Google, to the consumer? It’s supposed to disable terms that are underperforming, so that they are taken out of the rankings to improve the user experience. I’m simply not convinced it’s working at all as intended. I don’t even think anyone at Google really understands how their own tool works.

We recently had a client who was offering an online broadcast of March Madness college basketball games. It was approved by the NCAA, and, in fact, the URL was on their site. Can’t get much more relevant. This was an unusual PPC campaign, as there was a very narrow window of consumer interest. We started off with a test campaign using a core of college basketball and March Madness related terms, and then expanded to specific team and school names as the brackets were announced.

Traffic was slow on the test campaign… and some of the key terms got disabled. Quickly. Terms like “march madness basketball” – which were getting an acceptable clickthrough rate according to Google’s own definition. Look in the FAQ, and it says .5% for first position. Of course, part of the problem there may be that the numbers you see in the client center, and the numbers Google uses to determine a keyword’s status are not the same pool of data. You are looking at the total, including performance on partner sites. They are using Google-only search numbers. Why, if they are able to separate these numbers for their own use, and they are making decisions which effect your campaign, can’t the data be displayed separately for the advertiser as well? And how does it help Google, the advertiser, or the user if terms that are performing well on partner sites are not allowed to continue to be displayed on those sites?

There was a huge spike in March Madness-related search terms once the games got underway. Pre-season there was very little. So, by starting the campaign early, we ended up being penalized for an early lack of interest and were allowed to use some terms which had a high degree of relevancy and for which there was a great deal more traffic than when we started the campaign. The “smart” tool is not smart enough to factor in seasonal differences.

Then, as we added terms to the campaign, we encountered another problem. Terms were put on hold or even disabled – with no history whatsoever. How can any algorithm determine that any term will not perform well for any given client if it is never given a chance to be displayed?

During this campaign, I had lunch with a few people I work with at Google, and asked for an explanation. I was told it may be because of the performance of “similar” terms in the campaign, like plural versus singular. Problem was, there were no similar terms to most of the new ones which had any more history for such an evaluation. The Optimizer I spoke with seemed just as baffled and frustrated that accounts she uploaded new terms for often ended up with several ‘on hold’ or disabled from the start. She said, vaguely (something I have heard, in just as vague terms) that the number of terms on hold had to do with the total number in the account. If you upload too many new terms at once, several of them may be inactive. This also makes no sense. Why can I start a campaign with 2000 terms and be fine, but if I start with 200 and add 1000, I have several that are inactive?

The standard advice is to choose terms which are more specific. This, however, does not explain how a term like “gonzaga bulldogs college basketball game” – which is very specific – would be disabled as soon as it’s uploaded. Apparently, the Smart Keyword Evaluation Tool bases some of its decisions on the history of a term across accounts. So, if several other advertisers have not had success with a term, it might be disabled from day one when you upload it. This does not seem to make much sense. There is no way that an algorithm can evaluate relevance without history for the particular client that is advertising… and if a term is disabled for poor performance, a change in ad copy should allow that term another chance.

I think that this is a tool that was created as an attempt to shortcut the time necessary for human review in relevance. It is, however, shortsighted and flawed. It does not take into account conversions – a term with a low click-through but high conversions would seem like a good one. It does not take into account the differences between different advertisers for the same keywords. It does not take into account seasonal spikes in a term’s relevance. The only way to tell whether a term is relevant for the user is to either review it manually (with editors, as Overture does) or – as has been Google’s strength in the past – to allow the users to decide.

And so, I say to you, oh Google-powers-that-be, let the keywords run! Let the people decide!