Saturday, December 23, 2006
It's been eating at me. The unrealized potential of all these semi-interesting, but not-really-that-edgy, local city blogs. Toronto's a cool place in my book (hey, I've lived here for 20 years and have become attached). But if you aggregated 10-20 of the better lesser-known blogs from around here, you'd get more interesting coverage (and photos) than what's churned out by the relatively friendly and informative crew at Torontoist. Are they just too alike in their vision? Lazy? Have full time jobs? Don't actually get invited anyplace interesting? All of the above?
Among other things, this crew did not report on:
- The big warehouse fire at St. Clair and Ossington. And the second fire that happened in the same spot the next day. The blaze still burned up to 48 hours later, and traffic was completely snarled for the first 24. I wish I'd had my camera. In pouring rain, under a roof overhang attached to the Portuguese restaurant across the street, a crowd of laborers stood watching the last of the fire. They had nowhere else to be.
- The coat check girl at Acqua. Enough said.
- The absurd parking spots you wind up getting if you're dumb enough (like me) to rent a spot from some well-meaning but ultimately self-serving tool on Craigslist. The first spot I got was on level P5 of a condo and took me an extra ten minutes to get in and out every day (20 mins added to a total commute time of maybe 28 mins). The second was backed up against a giant hole in the ground (condo project) and frequently hijacked by construction workers.
- No fascinating coverage of graduation ceremonies at U of T, with lengthy speeches in Convocation Hall. (OK, maybe that's not going to make the final cut here.)
- Despite massive coverage of bicycle and other methods of green transport, no available downloads of "head cams" chronicling lengthy bicycle or rollerblade trips through the city. Maybe too shaky to be as soothing for sleep-deprived viewers as those old slow-motion "city drive" films used to be (remember the ones that they showed at 2 a.m.? Was it on Global or CityTV?).
- Annex-area bias? Excessive coverage of Tony Ianno and Olivia Chow; no mention of the dirty tricks campaign in a High Park - Parkdale provincial byelection, where Sylvia Watson's group tried to spread innuendo about a Cheri di Novo sermone that was "supportive of murderer Karla Homolka". Also no coverage of the absurd Watson "handlers" and "security" swarming around her pedestrian photo op at a local Bloor West bakery. di Novo won. Hate to be so pre-post-modern about things, but salacious political coverage has to be a must for any city blog.
- Not enough real-world restaurant reviews. What's overhyped, what's better than expected, even in one area - say Kensington? I don't think you have to be Yelp or Insiderpages to keep adding to the pool of reviews available online. There are dozens of great places to eat in 70 areas of the city. Joanne Kates can't try all of them.
The bottom line is, blogs matter. They matter because people are still searching for tons of things that show zero results. I'm convinced that insofar as citizen journalism fills in these empty spaces, on the whole it will enrich our lives, and provide a knowledge base that could lead to new forms of community, and faster decisions on matters of importance to all.
- What about gonzo retail journalism? Speaking of Kensington... how about meet a shopkeeper at some place that sells tons of lobsters, as opposed to just grabbing someone's generic lobster snaps from Flickr?
Lately, I sense they've been picking up steam at Torontoist. As mentioned, they're experimenting with a daily photo drawn from the pool of available ones at Flickr. They've begun profiling major intersections, which is leading to interesting chatter in the comments. If they keep this up, there'll be no need to start a rival blog!
I certainly don't have time to start one, but I'd be interested in co-starting one. :) Drop me a line if you have any ideas. Don't worry, I've already got a great domain name for it! Way better than Torontoist. (There is no such thing as a Torontoist.)
Thursday, December 21, 2006
Thanks to John/Rick for drawing my attention to this great post from Fred Wilson.
Bottom line: you can prototype a great web service for under $1 million. You don't need to burn $5, $10, $15 million getting it built, as in the 1.0 days. Today, the raw materials are just a lot cheaper.
However, follow-on financings - for the winners - are every bit as important as they were. So you throw fuel on the fire later. (No, maybe this doesn't apply to plentyoffish.com. But for anything remotely complicated involving some customer service, it does.) It sounds like investors have a greater opportunity to lose less on the losers, and bet more on the winners. That's more efficient, while still involving some risk. Sounds like a recipe for innovation.
In part this coincides with the exciting growth of innovative means of financing startups. New angel funds and the like are springing up all over. We can feel the buzz just in the Toronto area alone. Several employees of TrueLocal (with Jake Baillie as the managing director) just announced they'll be overseeing a new investment fund, backed by an investor. And at least a couple of prominent online success-story entrepreneurs are seeking to design investment funds that focus on initial investments of about $1 million.
Meanwhile, reports on investment banking show that there is a lot of cash in the bank, seeking too few deals. One presumes many of those investors will be seeking to put their cash to work in follow-on investments in proven startups. We've already seen this with companies like Zillow. We should expect quite a bit more of it in 2-3 years. Let's face it. There's still a lot of risk in many of these unproven online ideas, as compared with traditional high-tech and software investments. Smart venture capital is still important; you don't build and scale a Zillow on a shoestring.
Wednesday, December 20, 2006
On Dec. 10 I alluded cryptically to the scary-low search referral share numbers that site owners are often reporting, flying in the face (as usual) of the official stats showing Google's share to be anywhere from 45-62%.
Seeing those numbers, my question was: what would happen in a world where Yahoo and MSN search market shares were in the single digits? Or wait a minute, are we already in a world like that?
Some of the rationale for being so weirded out was seeing private site data, of course, but nothing too sweeping or conclusive. But then it was confirmed for me by Richard Zwicky of Enquisite, who showed me some stats aggregated across a whole bunch (many thousands) of sites, confirming Google search market share (mostly North American data but I think there may be some skew towards Canada and Australia?) at a whopping 78%!
Rich Skrenta chimed into this debate in a big way today in a post called Google's true search market share. Looking at as much data as he can, he pegs Google's U.S. share at 70% or so.
As for the #2 and #3 players: what's your identity now, if it isn't to "beat Google"?
Thinking it over, the answer: "to get better in areas other than search" isn't gonna cut it, just yet. Let's not let these guys throw in the towel!
Nope, I think at least Yahoo can still claw back some of that share. But only by implementing my idea of making the home page just a search box for a month. OK, two weeks.
Then, you gotta put search boxes everywhere. Partner with grocery stores to power "store search." Buy up the airports, hotels; give away televisions, cars. Whatever you gotta do. Issue those junk bonds. But you've got to get some of that share back. One leading search engine is just too boring.
In case you missed it this morning, here's my full writeup of the Jason Calacanis keynote interview with Danny Sullivan, for SearchDay.
Monday, December 18, 2006
Who'll be keeping it real in '07?
A few weeks ago I found myself co-presenting an intro to SEM (he on SEO, me on PPC) at WebmasterWorld Pubcon, with Wil Reynolds. What struck me about Wil's presentation was all the goodies he offered the audience. Absolutely free. He'd break from his presentation and go show them how they could get a certain type of info without paying a penny. Many had no idea these tools (eg. MSN adCenter Labs) existed.
Now that may not be remarkable to some. Working on a project with a programmer like Raja Bhatia, I know just how much leverage we have nowadays, on so few dollars, to get a lot done.
What Raja, and Wil, offer, is domain expertise. They don't come cheap in the end. But they'll not be selling you an overinflated research tool or proprietary plug-in. They'll work on an overall job, at a remarkably low price, and in the end (hopefully, if you get it) realize full value on saving you so much money and creating so much value.
That's the way I've always sought to operate, too.
Later, in the lobby of Pubcon, I talked to another SEO expert who had some surprising insights about the bid process for full-service SEO jobs. He might have bid $70-80k on a job (sounds costly, doesn't it), and find himself coming in third, behind two better-known firms who had bid $150k and $160k respectively. Clients would assume that because his quote lagged so far behind, there must be something wrong with his service. Well hey, if you guys aren't careful and don't hire us soon, this guy (the SEO) and I (the much more affordable Paid Search guy) are going to double rates... again... until we catch up to those "heavy duty solutions" at the big agencies. :)
I compare Wil's generous sharing of free tools with the brochure that just fell out of my conference memorabilia file from circa 2004. I'm sure this will be one less drinkin' buddy as I out some company (without mentioning them by name), but I'm astounded to think that you could make some mysterious "9 billion query database" part of a bag stuffer or exhibit. That a spike in searches for "Valentines" around Valentines day could be something leading up to a sale of a proprietary tool, when you could get the same info for free using common sense or Google Trends. I think it's that type of sales pitch Wil has grown tired of. And so has anyone else who works for a living in this business. Someone who claims to have a proprietary 9 billion query database... who really doesn't? (Did you read my Winning Results book where I explain that the best keyword research is always to run a real time AdWords campaign because you get real Google data, not phony secondhand metacrawler puffery?)
Some of us don't feel threatened by all the stuff that's being made free, to help us achieve goals for a lower and lower cost. To achieve those goals requires evolving domain expertise, integrity, and a thoroughgoing, full service approach. You won't need to find me in a booth to offer it, but hopefully, you'll find me. And vendors like Wil. And sorry, but Raja's already booked up. :)
Friday, December 15, 2006
No, I don't mean Blogger makes bloggin' fun. It's still giving everyone headaches, me included. It's you, the incredibly tolerant and talented Traffick readership, who still make it worthwhile. (Although you still don't write enough comments. You know, maybe you're aiming too high. We'll accept dumb comments around here!)
So to prevent myself from posting too much more this month, I thought I'd cap off the year by trying to review a year in search engine industry news (and related stuff), by picking one and only one Traffick post from each month that will hopefully be reminiscent of the highlight or "a" highlight of the month. In months where it's just too hard to pick one, I'm allowed up to two honorable mentions.
January: Traffick looks at an early version of Google Video, and offers its thumbs-down and "what the?" Later that year, Google waves white flag and buys YouTube for 10 figures.
Hon. mention: Chris Sherman quips about an analyst's projections for GOOG stalk, I mean stock.
Feb.: In the true spirit of forgiveness, Google welcomes back blackhats from BMW after three days in the penalty box. I'm guessing a maker of cheap, ugly cars might've faced a longer hiatus. But then what would Google know about cheap, ugly cars?
Mar.: Holy light news month. But still managed, using a minimum of words, to stick it to the jokers at Kinderstart, for suing Google.
Hon. mention: Signs of Web 2.0 jumping the shark: irritating 1999-like jargon.
Apr.: Not too much going on, other than Yahoo dragging its heels on getting rid of poor publisher partners under the YSM platform, and (yaaawn) Google posting blowout earnings.
May.: Although it might not have come through fully in my lengthy posts, I had the time of my life at the mesh conference. And I didn't even attend any parties! Amazing job by Stuart MacDonald, Matthew Ingram, and Mark Evans, organizing this groundbreaking conference. Evans has since left the National Post to take a position as VP Operations at b5 Media. Based on the month-long hangover from Search Engine Strategies Toronto (it was April 25-26) and our afterparty at The Boiler House, this will also be known as the month I officially went crazy, so it's lucky Gnarls Barkley was around all summer to commiserate.
Jun.: Some gentle props for MSN adCenter Labs.
Jul.: Google announces a hiring spree in Michigan. I believe this is newsworthy.
Dishonorable mention: Yahoo announces a delay in the Panama rollout schedule. Later, when the launch happens, we are surprised to find that they barely missed their timetable, so it's a shame they had to take so much flak from Wall Street about it.
Aug.: We return from a much-needed summer vacation to discover Danny's leaving Search Engine Strategies and Search Engine Watch!!!
Cheeky diversion: My Googlebomb of woefully out-of-touch Globe and Mail business reporter Patricia Best succeeds. In it own self-indulgent, unassuming, inconsequential way.
Sep.: Yahoo starts sharing forthcomingg Panama features with select folks, and at month end alludes to ad testing and a quality index in its advertiser newsletter. Same month: lots of site owners grappling with pages getting into Google's "Supplemental" index. Chatter about "BigDaddy" is ongoing.
Oct.: A gaggle of influential bloggers get a full run-through of Panama at Yahoo Search Marketing's Burbank headquarters. We hold a conference call to catch up on old times in the purple haze, and to reflect on how the new features will affect advertisers.
Nov.: Everything happened this month - the busiest month for posts and news this year. Thanks to Nick Fox for clarifying many details about Google's evolving process for assessing ads quality. Google Slap? Not something most advertisers need to worry about. Users are happier.
It's a dry heat: My first Pubcon won't be my last. What a blast! Unfortunately, if you know how to search the web, you'll notice that what happens in Vegas doesn't stay there. Goodman, Fishkin, and RustyBrick on a revolving bed? It's just in bad taste. Gord Hotchkiss jokes about people always misspelling Bellagio: priceless. Danny's keynote: timeless. Great fresh sessions on topics like viral marketing: guileless (OK, not really - often shameless). Rooftop banter with complete strangers: witless. And everything in between. This is what marketing conferences are all about: meeting new people and making new connections.
Dec.: Danny Sullivan, along with Chris Sherman and two others, forms a new company, Third Door Media, announcing a first Search Engine Expo conference for Seattle in June 2007.
The Notorious Young JC: Calacanis gets invited to keynote-chat with Danny at SES Chicago, and comes out with "SES is bullshit," and calls Valleywag "slime." He also comes up with some admirable insights about protecting the quality of online content. Whispers in the background cast aspersions on the quality of most of the content he oversaw at Weblogs, Inc.
Thursday, December 14, 2006
You've taken a few knocks from the naysayers in recent days. So you published a counter view humorously titled "The 12 Ways of Click Fraud" giving equal time to a typical Google critic. Hey, I'm as fond of critique as the next guy, and I love a good (or bad) joke. But these throw-all-against-the-wall type of quasi-exposes don't pass my BS filter easily. I cannot stand to have innuendo masquerading as fact on this issue, and we're just supposed to believe it because the target is fat and happy. Pssst - fat and happy are supposed to be proof you're doing something right...
If you don't mind, I'm going to respond in detail to the "12 Ways" assertions here. How scrapey of me, I know. Hey everyone, sign up for Andy's RSS feed so you get more of this kind of good stuff.
Mike's assertions in ital, my rejoinders in bold or bold ital...
1. Google earned $ 1.52 per month last quarter for every internet user over the age of 4 that use the internet 1 day a month (109 million) Less than 50% of internet users use the Google search engine. When you consider that many users do not click on PPC ads at all, the ones who do must click on a larger number.
Don't cite reach numbers to reach conclusions. How many impressions (searches, roughly) are served, and what's Google's effective CPM rate on that? That's generated by x% of users clicking on paid listings on a SERP. They do click on them. As for search market shares, Google generates probably 65% of total searches. Whatever numbers you're using about Internet user behavior above are (a) probably outdated; (b) don't lead to any clear conclusions about click fraud or even clicks.
2. There are many other things occupying the time of internet users besides search. Video has become very popular and in the first year of YouTube.com’s life, over 9,000 years of video had been watched. Email, Instant Messaging and sites like MySpace would also take time away from clicking on a PPC link. I absolutely love frisbee, and leaf raking. My cat enjoys sleep. Don't quit your day job.
3. I question the steady increase in PPC revenue quarter over quarter. Other forms of internet advertising have always experienced a drop in interest over time. One of the main reasons I don’t click on PPC ads is because I find them to be ineffective. Any that I have clicked on turned out to be disappointing.It's possible the engines have engaged in some earnings management to keep the growth looking smooth. Google actually implements initiatives to toughen editorial standards on advertisers, leading to short-term softening in revenues, but that's only so they pre-empt a potentially larger growth quarter in the following quarter (thus smoother looking growth). As for your personal experiences, they aren't borne out by the data, which is unsurprising. My clients track ROI and show real users clicking on real ads, buying real products. Your personal experience, therefore, is not real enough for me to care about.
Eventually, paid search revenues will level off. Of course. It just hasn't happened yet.
4. What’s with Google’s recent increase in click fraud PR surrounding PPC? Between Matt Cutt on a debunking quest and Shuman Ghosemajumder’s recent meeting with Andy Beal, the Google PR machine seems to be working overtime. Shuman tossed in a red herring by introducing the term “invalid click” into the click fraud discussion. Invalid clicks should be an internal metric used only by Google. Invalid clicks have no place in the click fraud discussion.
I'm guessing they're tired of self-interested parties lying about the issue, to Google's detriment and embarrassment. Over the holidays, Googlers' families can now say "you sure told them!" instead of "gosh, you're really in that search game? I read that article in TradPub and man did they make you guys look like criminals!"
Google's called them invalid clicks for some time, likely as a euphemism but also because fraud must be proven. Google doesn't charge for a variety of clicks that are no good to advertisers but it would be presumptuous to call this fraud in all or even many cases. Law professor Eric Goldman, an expert in this area, has argued against the use of the term "fraud" to describe non-converting clicks, as it confuses real legal issues (as in, you must prove fraud and convict someone - or it is at best, alleged fraud). I'm sure Goldman has a smarter way of describing it, but I'll leave it there.
5. Bid rates for competitive keywords has been on the decline. The only reason for this would be a lower ROI on those keywords. This sends savvy AdWords customers further down the “long tail” in search of quality leads and better value.I won't even accept the premise.
6. When AdWords customers travel further down the tail, it bolsters the click fraud network. The long tail hides a well written network. Quality bot networks would access the fraud site through Google’s own search engine results. By achieving this connection, the PPC transaction is given more weight and undergoes less scrutiny than other sites. To score high enough in the listings, the fraud site has to use terms in the long tail.Nothing hides anything. The long tail is a valuable part of search inventory, and advertisers often use it to their benefit. They are not getting click-botted into submission (we measure ROI, remember?).
7. This long tail also enables search arbitrage. This practice involves buying AdWords at a low bid price and directing the clickthrough to an AdSense website with higher paying PPC links. Mix some of this practice into your click fraud network to water down the clicks coming from the bots.
Google is explicitly targeting many forms of search arbitrage through the quality score initiative that includes stringent checks on landing page quality. This process disables a wide variety of keywords and ads in accounts intended to achieve arbitrage-type benefits. Users dislike arbitrage; advertisers don't like to compete with arbitrage advertisers. So Google's taken systematic action against this. That is manifested in prohibitively high minimum bids in accounts that point users to low-quality landing pages, such as certain kinds of ad network dominated "arbitrage type" pages.
8. Search arbitrage encourages multiple PPC service abuse. The fraudster uses an AdWord account to send leads to a Yahoo, MSN or Ask optimized page. In this case the money is paid to Google and removed from the other company. This can also work in reverse to further cloud the issue.
I'd agree that click fraud on Yahoo publisher partner links has been a more significant problem than many realize. Yahoo will have to give advertisers more transparency into the workings of the network, and will need to police bad publishers. Google's actually gotten out in front of the issue with the above-mentioned landing page quality initiative, making it more difficult for those publishers to even receive traffic from cheap AdWords clicks. In short, Google's ad quality initiative is actually saving Yahoo Search Marketing advertisers money, without their necessarily being aware of this.
9 . Advanced bot networks have been found producing SPAM email from infected computers. Some have even installed anti-virus filters themselves. This is in an effort to keep the bandwidth and processing power for the bot. Some operate in a peer to peer method. This allows the network to be run by one machine and be restarted by a totally separate machine. We catch email bots because they leave millions of spam email as evidence. Click fraud bots leave no evidence.No question, we are dealing with some nefarious bots. The question is, how much fraud is flowing through to the end advertiser's bill, in the end. You haven't done anything here again but simply raise innuendo levels and spread FUD.
I'm getting hungry so I'll let you have that one.
11. Click fraud is really only a concern for those AdWord accounts in which Google is paying a publisher. It makes very little sense for click fraud to be present on any site that Google owns. Unless someone is trying to max out a competitors account, there is no financial gain. From my best research, Google allows you to opt out of Google but does not offer a “Google only” option for AdWord customers. By mixing in the publishers (AdSense accounts) to the pure Google network, they hide the true rate of click fraud. The baseline metric missing is the pure click through rate on the Google properties alone. That would give a good indication of inflated conversion percentages in other areas.
Are you telling me you've never actually logged into AdWords? It's absolutely incorrect that you cannot opt out of network traffic. You can! You can also bid lower on it if you wish, or exclude unwanted publisher sites either through the interface or by contacting a Google rep. All of the above info and disclosure is available; and all of the opt-outs you seek are already in place. They weren't three years ago, but this isn't three years ago.
12. The way that Google handles new accounts make me question their ethics. Many pitfalls for new accounts are outlined in this article http://blog.merjis.com/2006/10/04/googles-evil-way-with-content-match/ including $ 5.00 minimum cost per click and initial set up including the content network. The content network is where the highest rates of fraud are reported. Since Google creates very little content, nearly the whole sub-area is AdSense accounts. It is those accounts in which the fraud lives and hides. AdWords does allow customizing of which sites your ads will be shown but only as an advanced feature.
I'd agree that a rookie opening a new account without knowledge is probably going to be in a world of pain. But where you get this $5.00 business is beyond me. Yep, the network's been a problem over the years, but we've been through a lot of waves of policy and implementation on this front. Real advertisers have observed all developments closely. As someone who has never logged into an AdWords account, I guess I can't count you in their number.
…. and a partridge in a pear tree.
I'm getting thirsty now. So I'll give you that one, too. Merry Christmas.
Wednesday, December 13, 2006
The votes are in, and a sizeable percentage of commenters appear to believe that Andy Beal's exclusive on Google's click fraud detection process is just telling one side of the story.
As you know, over the years we've been hard on all search engines (especially second-tiers and Yahoo) when it comes to low quality traffic in general, and in particular glitches and poor traffic in the content networks.
That said, I've long been with Andy in that I could never see where these various "accusers" were getting their numbers.
In my seminar at post-SES (Incisive Media training) last Friday, I asked the group of 23 workshop participants what they were hearing. What the press (as opposed to industry insiders or colleagues) was talking about when it came to the search industry. I'd expected various top-level stories and the nonsense about Larry's plane hammock. But no, they only recalled one basic thing about media coverage of search marketing: click fraud. Hmm.
There are several components to the click fraud debate, and now, to the "click fraud remediation industry." If you assess how various practitioners are dealing with this, there are the self-promoters and fearmongers, and then there are those, like ClickTracks, who take a low-key approach to educating the marketplace about how you might document the forms of click fraud that slip through the cracks.
The info about how much fraud exists at various levels of the filtering process simply rings true to me, and I manage or oversee a lot of different campaigns. The stuff that finally gets detected by an advertiser, that hasn't been caught by Google's algorithmic or human reviews, is a tiny proportion of overall clicks (if it exists at all).
So then the tactics of how to deal with the (rare instances of significant) fraud have to be weighed. Is submitting a special type of forensic report going to "impress" Google and make them more likely to refund you? I haven't heard many instances of this. All it seems to be is sales efforts by the third party fraud detection companies to find potential customers who have potentially large fraud claims, getting a % of a fraud refund as a cut of the action. As yet, I have no evidence that those third parties have a superior causal effect in terms of Google's refund decisions or amounts when compared with your own actions backed up by your own analytics; indeed, it's just as likely that they might be counterproductive.
I had an odd exchange with one student in Friday's seminar on this topic. I basically wanted to point out that whether or not you use a third party service, the end result is that you submit your fraud claim to Google, and they decide whether to give you a refund. No matter how fancy your forensics, you know what? Your request does little other than to trigger a request for Google to do its own forensics. The attendee at the seminar contended that "we don't need to be at the mercy of Google" if we "opted out of the lawsuit settlement," which I'm afraid conflated a request for a refund with a class action claim. I'm also afraid that we are, in fact, at the mercy of Google when it comes to a click fraud refund decision. You submit your complaint. They decide. There is no third party body helping or adjudicating claims. Hope that clarifies things.
So does this post, too, qualify as a "release from Google's PR department"? Well, say it if you like, but what do you do to act on the sentiment? Set up an autoresponder to ping Google daily with fraud claims? Probably not. So... what do you do then? Continue managing campaigns to the best of your ability? Or... wait a minute, do you even manage any campaigns?
So, I've noticed that most of those who like to talk about fraud either sell something related to this, or do very little if any advertising at all. Those who are "thinking about" paid search, amazingly, seem incredibly eager to discuss the topic - as it excuses their inaction on deploying a new type of (highly effective) advertising.
Yep, there are isolated cases. Those need to be rectified. There remain disputes over what counts as an actual click. But the number of cases and the dollar cost of invalid clicks (those that actually get charged to us) may well be much smaller than we've been led to believe.
If I offended you, please be offended enough to make this Diggbait. Thanks for your time.
Tuesday, December 12, 2006
Y!SM just announced they've launched the new online sign-up system for the Panama advertising platform. Previously, Panama was only open to existing advertisers who were transitioning from the old system (Direct Traffic Center) to the new one. The company continues to incrementally migrate existing customers (done by Y!SM invite) to the new system and their plan is to have all accounts on the Panama system by Q1 2007 (no set date yet).
For more information or to sign up for a new Yahoo! account, go to: http://signup.marketingsolutions.yahoo.com/.
Current customers can request an early account upgrade at: http://advision.webevents.yahoo.com/newsponsoredsearch/invite.
Sunday, December 10, 2006
You know I agree with John Krystynak's take - because earlier I accused wild and crazy Peanut Butter Man, Brad Garlinghouse, of "incrementalism." Shortly after that memo, Yahoo announced a reorganization. Krystynak thinks it's missing the mark, as outlined in his post.
I sent John a note with some further points:
- Agree on Semel. Technical chops are hard enough for a semi-tech-savvy younger more experienced manager to build up. Semel is none of the above, and efforts to learn would be like teaching me Swahili - it's bound to be too little too late.
- Absolutely on the need to acquire more ad inventory. I know it's no easy matter, but it's a matter of leadership and dire consequences if you don't do it. Google risked major disaster thru the Youtube acquisition and the MySpace guarantee, but they did it "because they needed more places to put ads". Plain and simple. I don't care if Yahoo has to issue junk bonds, they have to get one of these deals done. Internally all the smart people at Yahoo seem demoralized at losing out on Youtube, Facebook, and anything else that may be up for grabs.
- Including Efficient Frontier in the should buy category is pretty much just you logrolling for friends or people you'd like to see get rich, so I will ignore that one. :)
- On YPN, the go broad plan is definitely what they seek to do - they're all over this. But getting adoption is hard. Google arguably did not have its top people on AdSense, which is why it became such a mess (especially for advertisers) - it won anyway, because it got quick adoption. Nick Denton claimed that one of the top YPN folks is a "scenester". Time'll tell.
To all this, John sent a concluding point, which he gave me permission to quote. In part:
- On the ease of dumping AdWords campaigns into the new YSM: this is an obviously key point for us in migration. Some of the Overture campaigns we manage are worse built than the AdWords campaigns, and will import more stupidly than the AdWords campaigns. However, this import from AdWords feature is already a working feature of the Panama implementation. There should be how-tos and seminars about this right now, but everything is happening in such a hurry. The default seems to be very compelling to just go with the overture-to-panama version but we would rather not do that in many cases, so, I'm looking into this and pondering the best way to tackle it. Here's how it looks at the moment. You have to say "yes" to upgrading your old YSM account if you want to upgrade to Panama, but you probably won't want to keep that account. So instead, download it all to a spreadsheet as a backup, then delete that stuff and go ahead and use the import feature to bring in your (probably superior) AdWords campaign. I hope that makes sense. :)
"The REAL problem with Yahoo is that they are so matrix managed molasses entombed that they CAN'T DO what they seek to do.
"They are too slow, too perfectionist, too gold plated, too contemplative.
"They are losing because of all that. That's why even though Decker (from Blade Runner?) is an incremental improvement, Yahoo is in need of a Jobs-ian transformation. Semel is Gil Amelio, and nothing incremental at this point is gonna change the ponder-osity that plagues them."
Now back to Goodman's thoughts...
I wish it were prettier. But if you're a search engine junkie, this stuff isn't pretty. Yahoo as a whole is still one of the better companies in the world. But for search marketers? They have spirit and a new resolve - I've witnessed that first hand. The question is, though -- do they have searches? You know, the stuff we want to rank in, and show ads next to.
Try this thought on for size. It's a rough one. But if you don't listen now, don't worry: I'll repeat it later, if I get confirmation from the real life numbers.
For years, metrics agencies have been overreporting Yahoo's search market share numbers. So, we assumed, if we got tough with the numbers, Yahoo was "only" garnering around 25% market share in terms of raw web searches.
What if you woke up one day, looked in the server logs of a few hundred representative, popular vertical websites (ones that aren't paying Yahoo for inclusion), and found... say... Yahoo's market share is actually more like 10%! Or 6%! Or 5%. Could it happen? Is it already happening? It would explain a lot, if so. It would mean Yahoo has to somehow figure out a way to survive by serving and selling a whole lot of ads against online content. And that giving up on search might be part of the plan. It would be a sad day if it happened. But the numbers can't be made to lie forever. Stay tuned.
If you're not paying for inclusion, what percentage of your site's visits come from organic Yahoo search?
Want the truth of what Shoemoney does to make money from AdSense? You won't get it from a Forbes reporter, evidently. Jeremy's followup explains a little.
Friday, December 08, 2006
On Tuesday at SES Chicago, Danny Sullivan interviewed Jason Calacanis.
I'll have a longer article in SearchDay. Basic observations, now that the dust has settled:
In the end, Calacanis' substance didn't offend me. I found it constructive. His style was easy to get past, for me, but not so much for those who don't appreciate that the play's the thing. So you have to conclude that the dismay with the man stems from some combination of fallout from his real-life actions, his self-promotional schtick, and the hypocrisy that inevitably surrounds the rhetoric of purity. Regardless, he'll probably hit a few more home runs before he's done.
- Before the keynote, there was anticipation that this "controversial character" would be a "wind him up and let him go" type speaker. Some let it be known that he'd probably pop off and be offensive.
- Calacanis was offensive, to some. Not to me. His observation that "SEO is bullshit" is far from unique (Godin said similar, and so, I think, did those old Wired and Webmonkey one-pagers that used to rank so well for the term SEO). But to 75% of this audience, apparently it was blasphemy. What Calanacis meant was really not far different from what a number of the white hatters say, but it goes farther. Being very concerned about building a content business happens to focus so heavily on the fundamentals that of course it is good SEO in itself. People: I'm afraid he's right, or at least, how can you argue with success. Weblogs, Inc. got a whole lot of traffic from that attitude.
- A number of other seemingly offensive points were fundamental defenses of the spirit of quality web experiences against deceptive ones. Not liking pay-per-post blog schemes is not exactly offensive, it's actually again, kind of a white hat statement.
- He said a lot about his work at Netscape, recreating a Digg-like experience. I found the comments about paying the top 50 "content curators" to ensure a professional approach to editorial work quite insightful. Indeed it eerily paralleled the many specific discussions initiated and fostered by Traffick around the implicit incentives lurking at the Open Directory Project when no editorial staff got paid. I argued at some points for making those incentives explicit, to anchor the thing in reliable editorial quality. Calanacis is saying precisely the same thing. Strangely, after all these years, at this same conference I bumped into David Prenatt, then known as netesq in the old ODP debates which made us infamous around here. :) (There were forum discussions that are no longer accessible)
- Danny really shouldn't have. He played word association with Calanacis. The first few things that came out of his mouth were cautious and respectful, and you got the feeling he was laboring to be statesmanlike. Then Danny mentioned Nick Denton ("Valleywag", to be precise) and lo and behold the venom spewed forth. "Slime" was one of the words he used. It seems there is no love lost between these two, but it's hard to really figure out why, other than the fact that these two were competitors and one likes to downplay his business accomplishments and the other to oversell them. And both make a living or a hobby out of offending people, it seems.
Thursday, December 07, 2006
One of my favorite Dilbert cartoons is the one where Dogbert is accused of becoming "an annoying person who misinterprets everything you say." He replies "Yes, I'm more assertive." Interlocutor replies: "I said annoying, not assertive." And then something or other about climbing into a clothes dryer.
Anyhoo, there is a usual tempest in a teapot unfolding with a site getting into trouble with Google, and ranting and raving about it even though Google took appropriate actions and even tried contacting the site owners. So far, there are 121! comments on Matt's post about this.
One of the comments, by Shelly, reads:
So what you’re saying is that if a site has links to subjects that Google doesn’t approve of, and these links are hidden, you remove the site from Google’s results? When is Google going to realize if you can’t legislate morality, you can’t algorithmically control morality, either?
Shelly must be one of those people who believes hidden text referring to animal porn provides an optimal user experience, and that Google has an obligation to help hackers disrupt legitimate businesses. What a bunch of heavy-handed goons that they won't bend over backwards to accommodate whatever it is Shelly seems to want, or might be thinking. Because, not to put too fine a point on it (as Shelly clearly refuses to): a bunch of randomly-selected search results showing up willy-nilly is perfectly analogous to free speech, and equals freedom for all.
It does look cosy in there. I think I'll climb in.
Nick Denton weighs in with the definitive Lloyd Braun post.
At least, I think it's Nick Denton. BWTHDIKA.
Two different actors played Lloyd Braun on Seinfeld (a bit of Seinfeld trivia). The second (Peter Keleghan), handsomer than the first and the architect of the doomed name tag strategy for Mayor Dinkins, also played the arrogant news anchor on the nichey Canadian TV series The Newsroom. Neither actor, apparently, is the real Lloyd Braun.
Tuesday, December 05, 2006
As expected, the second shoe has dropped on Danny's new plans. His company, Third Door Media, will hold its first conference in Seattle in June.
Some of the live-blogger posts will be more comprehensive, but here's some quick takes.
According to one rep from Incisive Media, a large proportion of this year's attendees are first-timers. This has meant, I believe, a high degree of satisfaction with sessions that are geared towards beginner to intermediate search marketers. Matt van Wagner, who co-presented Paid Search 101 with Dana Todd, discovered that the entire audience for that session really was in the right room. Tons of folks are really just looking for an introduction.
Somewhat typical, if surprising, was the furniture retailer I bumped into at the Yahoo! party (and when am I not at a Yahoo party?). He mentioned he was spending a scant $100/mo. on paid search, just dipping his toe in that water. "I just hate giving a dollar to Google or Yahoo," he said. That, of course, assumes there is some other legitimate way to consistently drive traffic to your company's site. He was considering it, and wondering what the right amount of money would be to run a few tests to get a campaign optimized. That's the right type of thinking - but it also indicates that it hasn't happened yet, which augurs for continued growth both in conference learning and in spending on paid search.
The session on Ads Quality in which I spoke along with Joshua Stylman and Jonathan Mendez, was fuller than you would expect, in one of the larger session rooms. Many advertisers who weren't in that room and who aren't paying attention will take some time to crack that nut.
Currently Yahoo is holding a networking lunch & learn to intro their Panama platform here. As I walked past, the room was already starting to fill beyond capacity. Because this is a "sponsored" type (or at least large-company-sponsor related) lunch that is in parallel with the standard lunch, you wouldn't necessarily expect attendees to attend (interruption marketing vs. lunching with pals)... but they're doing so. They want to hear the message.
I just finished moderating a panel on domain type-in traffic and the like. The CEO of Moniker.com, Monte Cahn, spoke extensively. Shorter presentations were offered by Jon Lisbin of Point-It (from the SEM agency side), Andrew Beckman of SearchAdNetwork, Josh Meyers of Yahoo, and Hal Bailey of Google. As moderator, I vow to be moderate. But a key takeaway from the Q&A is clearly this: Jon (backed up by at least a couple of audience members) argued for site exclude capability and challenged both Yahoo and Google to do better at this. With Google, it looks like parked domain traffic and the like are offered through "search partners" and not the content network, meaning that exclusion is only something you could negotiate outside the platform, going through your AdWords rep. But Jon's agency rep noted that two major sources of this traffic were "unexcludable." Hal Bailey claimed a very recent decision had been made (one wondered if it was right on the spot) to make "any site excludable." FWIW, Lisbin was smart enough to show a screen shot of the communication from his agency rep, so basically, he had them dead to rights.
Yahoo's Meyers noted that new control features would be coming down the pike as Panama rolled out, but gave few specifics and made no tangible promises. Both Yahoo and Google maintain that this traffic is "search-like," and have no plans to move it into the content network or its own category.
My comment would be: at least, if you say you're going to offer fuller-featured site exclusion controls, do what you say, Yahoo and Google. I think I speak for nearly all advertisers when I say: we're waiting for you to come through here.
Back in awhile.
Monday, December 04, 2006
Seth notes that Kraft Guacamole lists avocado as the ninth ingredient. Guacamole is (pretty much entirely) avocado.
Is it inevitable that we buy stuff like this? Of course not. Are we conditioned to buy stuff like this? Nah. Not only conditioned -- steered.
The secret isn't only advertising. It's distribution. If the other products aren't readily available, people don't know about them, or aren't willing to look. At least, that's the way it used to be. I'm an advocate of the belief that the concept of "search" will actually change the level of complacency out there even in the offline search for what you want. The tide of consumer empowerment is changing a lot of tastes (because it can) - but in the meantime, the big guys resist, in the name of profit.
Granted, I didn't look hard, but in my grocery store, I didn't find "Kraft Guacamole". I found whatever the guacamole was that they sell in their large display of dips, including 80 kinds of hummus and so forth. I found whatever they had, took it home, then looked at the ingredients. It said "avocado."
I actually had trouble believing it; kept looking for the other ingredients. So I'm at least conditioned to think that someone is trying to screw me.
Guacamole, with hardly any avocado in it? I have a good friend who spent much of her career at Kraft. I'd love to know what she thinks of this. No, what she *really* thinks of this.
How will our supermarkets handle us, the Long Tail Generation? Or are "we" just an elite of demanding snobs? Barack Obama says (on Leno, last night?) it's comical to think people outside of big cities don't want to, say, have Dijon mustard with their hot dog. It's this crazy media stereotype. Then again... he's Barack Obama.
Friday, December 01, 2006
Danny reflects briefly on his last post at Search Engine Watch, as he moves over to Search Engine Land, a new venture.
One thing I hadn't heard much of before was Eric Ward's instrumental role in getting the word out about the early Search Engine Watch (Eric is famed for his "invention" of the linking campaign, having done this to great effect for Amazon in 1994).
On this special occasion of Danny's shift, maybe it's a good time to share the story of the Traffick.com linking campaign circa 1999-2000. I had no idea that Danny had so much help from Eric. :) Some people know this because they asked followup questions in the bar after asking the usual "how did you get into this business" questions.
At a certain point, again before most folks had even heard of linking campaigns, we decided that we admired Search Engine Watch's linkage. We noticed a ton of libraries and so on had linked to Danny's site as a great "resource" about search engines. It was really one of the few around, so of course a must-link for most libraries. And we reasoned: libraries are a pretty reputable source. They must have really high PageRank. A lot of them are .edu. It's the sort of reasoning you might use now as an SEM, but today, the difference is, you won't be able to act on it, because no one's going to respond to that kind of "link to me" request.
So anyway, I contacted a few hundred of these librarians (yes, I mean the same ones we saw linking to SearchEngineWatch.com) and said "hey, this Traffick.com here is also a good resource." Not exactly a brilliant, original strategy, but for me, still the best linking campaign I've ever conducted (OK, also one of only a handful I've ever personally done). A large number of university resources pages were gracious enough to link to us.
Little did I know, Eric was at least partly responsible for a lot of the great linkage we were trying to piggyback on. Though there would have been no way to personally contact the tens of thousands of reputable sites already linking to SearchEngineWatch.com. That's how popular the site was, even in the early days. It continued to grow from there.
Becoming a fraction as popular as SEW was enough to get me into the business full time. Turning to consulting meant that content was more of a hobby than anything. It still is - and I still write too much. :)
As for comprehensive search coverage, Search Engine Land will be one of the few blogs worth adding to your feed reader, in an age of info overload. It looks pretty much like business as usual with Danny & Co., which is a relief.
Another point Danny makes in his farewell post - citing a post he made for us Search Engine Watch Forum moderators last week - is characteristic of his open, transparent approach. He cautions against anyone making up rivalries where they don't exist: By no means do I want anyone thinking that staying on here, or perhaps doing other things with SEW, is somehow something I won't like or perhaps "disloyal" in any way. I don't know if anyone was even thinking like that -- but if so, don't!
Indeed. Life is too short, and that would be silly. The fact is, the site, Search Engine Watch, and the conference series, Search Engine Strategies, have gone through ownership changes for the third time now (to recap: from Danny to Internet.com/INTMedia/JupiterMedia; from Jupiter Media to Incisive Media; Incisive Media now being taken private under new ownership) and are now no longer spearheaded by Danny. Although Danny has been the central public figure in these properties, he has not owned them since 1997. The owners, with or without Danny (clearly, now without) will have to decide on the best way to serve the customers they've always served, and the best way to turn a profit. That likely means focusing on what happens next with the conference series, as that has been the key revenue driver as compared with the online properties. (And in my personal opinion, Search Engine Watch won't have much of a future after Danny's departure.) Continued management of SES conferences globally doesn't sound like it'll be mutually exclusive or impossibly competitive with any new conference series launched by Danny. Time will tell.
That's Shoe #2, waiting to drop. Danny's moved to a new set of editorial roles - but what about the conferences?
View Posts by Category