Search Trends - Search Engine Marketing

I received an email from Google yesterday inviting Froogle advertisers with over 1000 products to try their Ad Automator from a feed similar (nearly identical) to the Froogle feed. Here's the text from that portion of the email:

Froogle merchants with over 1000 products who are also AdWords advertisers are invited to try out the beta release of our new ad-producing technology, the ad automator.

The ad automator uses sophisticated technology to automatically create and target AdWords ads based on structured data feeds, which are nearly identical to the data feeds you submit to Froogle. You’ll be able to create keyword-less data feed campaigns in your existing AdWords account while investing minimal time and effort. Simply send us your feeds, and the ad automator will automatically generate targeted AdWords ads that begin showin grightaway.You’ ll be able to:

Leverage the product copy you’ve already written for your site
Generate specific, high-CTR ads with no additional work
Increase the number of queries on which your ads appear
The ad automator provides added coverage by targeting queries that may have been overlooked in your keyword campaign. The ads produced are highly relevant and precisely targeted to the user’s search query.

Sounds like a pretty cool tool to save time creating keyword/product-specific ads in Adwords. I'm going to try it out and I'll post the results to let you know how it works.

"More than 72 percent of online shoppers use general search engines nine or more times per month, and a growing portion of those searches are shopping-related or local in nature, according to a consumer search behavior study by BizRate.com and the Kelsey Group."

Read More from Clickz

Looks like Google's not the only profitable search engine ;)

"Ask Jeeves more than doubled its profits from last year, posting third quarter net income of $10.7 million, or $0.15 per share, the company reported today.
This compares to income of $3.8 million, or $0.07 per share, for the comparable year-ago quarter. Revenues for the quarter grew 178 percent to $75.7 million, compared with $27.2 million for the same quarter last year. "

Read the article on Clickz

Its nice to hear that sex-related searches are down to approximately 5% of all searches (from 20% of searches in 1997).

People are adapting to the Internet as a medium of communication and we're just using it to do business, conduct research, make purchases, etc. The study also said searchers are still very impatient, usually only looking at one page of results and spending less than five minutes on a search.

Another feather in Google's cap:
"Google today added to its search technology holdings with the acquisition of digital mapping company Keyhole Corp, a move that has implications for its local search strategy and marks the company's expansion into subscription-based services. Financial terms were not disclosed."

Read the article from Clickz

This article (an overview of a presentation at SES) addresses the issue of tracking sales from people who research a product online but buy it offline. He offers a few suggestions for tracking some of the sales that occur through that channel. I'm sure the percentage of purchases made offline after online research varies considerably for each industry and each company within an industry, but this is an intriguing concept, and one which seems to elude most online marketers.

I'm not sure you'll ever be able to track every single offline purchase that occurs after online research, but it makes sense to do everything possible to track those sales so you can know how effective your ad dollars have been.

I read an interesting article today that talks about the new social networking site, LinkedIn will be able to be profitable. I read another article a few days back in which the founders (who happen to come from eBay) say they plan to apply the ebay business model to LinkedIn. The idea being that a few "power users" will account for the bulk of the company's revenue. In the case of LinkedIn, those power users are people like recruiters who are willing to pay a premium for access to the LinkedIn network database. I have a hard time believing this will be the next eBay, but I can definitely see it reaching profitability through ad revenue and premium services. Based on the founders' past success and the current growth of the LinkedIn network, it seems to be on the right track.

Wow! A search engine specialist jobs related to search engines. Does that mean we're really our own industry now? I think it's a great idea, and now that everyone's talking about search, it makes sense. The only problem with JobsinSearch.com is they don't have any jobs...at least not any in my country. Oh, did I mention I live in the US?!?! That's always the trick with starting a new site like that, but hopefully they can quickly get some actual US job listings in there and maybe become a decent meeting place for job seekers and recruiters. For now, unless you live in the UK or plan to relocate there, it's all just a nice idea ;)

From an article in MediaDailyNews:
"In its first report since going public, Google, Inc. delivered strong third-quarter earnings Thursday. The Mountain View, Calif. company earned $52 million--up from the year-ago $20 million--on revenue of $806 million for the quarter ending Sept. 30, 2004. That's up 105 percent from $394 million in the third quarter 2003. "

I'm not surprised that Google's making serious money, is anyone here surprised? I'm sure they'll continue to show strong growth for a long time to come. Google's stock isn't doing too shabby, either, with serious gains on news of their Q3 results.

Is this the next Dot-com Bubble? I don't think so. I think it's pretty easy to see the difference between Google and Pets.com. (In case you don't see the difference, one had a sock puppet that looked like a dog, the other is making 8 figure profits every quarter). Google's hot and I'm sure it won't hurt the industry, but I think investors aren't as stupid as they were 5 years ago. Well, they might be just as stupid, but hopefully learned not to make the same mistakes.

I personally feel like fraudulent click traffic is a huge problem, more so with certain smaller PPC search engines than others, but it definitely exists and costs advertisers a lot of money. There's a recent article by Tom Zeller of the New York Times taking about the problem.

I think it's one of those things most people in the industry know about, but since it's so difficult to track, people just kind of ignore it. I like the fact that it's finally getting some attention in the mainstream press. Outside the search marketing/Net marketing crowd, I'm sure it's a relatively unknown phenomenon. Some search engines are a lot better about filtering out junk/fraudulent traffic than others. The main problem causing all the headaches for marketers is the whole PPC business model--namely the ppc affiliate programs. PPC affiliates are given a very strong incentive to cheat the system and create bogus traffic through their links.

Andrew Goodman thinks the level of click fraud is being blown way out of proportion. I think he must have read a different article than the one I read. That article doesn't blow anything out of proportion, it simply ackowledges that this is a problem. Sarcastic little remarks about ad nazis don't make it any less of a real problem--a problem that search engines don't seem to care about because not enough advertisers are making a fuss, mostly because they lack the ability to properly track fraudulent traffic.

For now, my best solution for beating fraudulent traffic is to track results from each search engine (and each keyword if possible), so I can know what the ROI is on my advertising dollars. Then I kill any ad spends that aren't performing at an acceptable level. I've run PPC campaigns that yield a 0% conversion rate, and whether that's fraudulent traffic or just bad advertising, it makes sense to cut off the bleeding. Hopefully market forces will cause the search engines to take more interest in cutting down on fraudulent traffic, but as long as they keep getting advertisers willing to blindly pay for whatever traffic they can send, they'll keep on sending "whatever traffic".

I read an interesting post this morning on Search Engine News Journal talking about search activity related to the two presidential candidates. This post refers to a press release by a search marketing company that did some research into the issue of keyword popularity and search marketing among presidential candidates.

Top findings include the following:

• Searches on the key phrase “John Kerry” outnumber searches done on the key phrase “George Bush” by a 4-to-1 margin. If you also account for those searching on key phrase “George W. Bush,” “John Kerry” is ahead by a 2-to-1 ratio.
  
• Internet searchterms “Republican Party” and “Democratic Party” are even, but searchterm “RNC” shows up one-and-a-half more times than searchterm “DNC.”
  
• The key phrase “homeland security” produces twice as many Internet searches as the searchterms “gun control” or “poverty.”
  
• The searchterm “war in Iraq,” generates twice as many Internet searches as the key phrases “civil rights” or “health care.”
  
• In looking at search engine placement, neither of the official presidential campaign sites was among Google’s first page results when searching upon the key phrase “homeland security.” Johnkerry.com could be found on the second page, ranking 16th. Georgewbush.com could be found on the fourth page, ranking 41st.
  

I think it's always intersesting to look at differences in keyword searches for various topics related to politics, current events, etc. I'm not sure why it matters that neither candidate's site doesn't show up for homeland security. The sites that should show up for that term are showing up right at the top. Like any press release, you have to take it in context of who's writing it. Coming from a SEO company, it makes sense that they would make the point that the candidates' sites could show up higher for various "related" keywords if they had sites/content that were properly optimized. However, I don't think people have any difficulty finding the official website of each candidate or any opposing views they may want to explore. That's the beauty of "natural" search, it provides the correct results naturally, as a result of relevant content and links pointing to that content.

An article in USA Today brings up a few of the security and privacy risks of using Google's new desktop search. They focus on the risks of using a shared computer, especially a publicly available computer in places like libraries and copy shops. I have to agree. There are always risks associated with using a public computer for sensitive email, purchases, or whatever else. Google's desktop search simply makes it easier for prying eyes to access that information. It's not that Google's tool is capturing any data that's not already on the computer, it just makes finding such information as easy as searching on Google. My advice? Be VERY careful what you use public computers for, and try not to leave any personal, private, or otherwise sensitive information accessible on the computer when you leave.

The big three has always been AOL, MSN, and Yahoo, but more and more it's becoming Yahoo, MSN and Google. AOL's still got enough of a userbase to shake a stick at with 23 million subscribers, but they are losing ground in the competitive portal game. The "new" big three all have their own search technlogy (or will as soon as MSN search goes live), but AOL is still using Google's search, which is fine, but it means they have to share a nice chunk of every search dollar with Google.

I've actually been thinking in terms of the big four instead of the big three for the past couple years now. AOL has to lose a lot more ground before I consider them outside the top tier of search/portal advertising.

Read the article on ecommercetimes.com.

Once again Andy Beal really does have the lowdown when it comes to search engines. The news is starting to show up all over the place, but Andy let us know this morning about the launch of desktop search from Google. We all knew this was coming. It was just a matter of when. Well, I guess the time is now, at least for the beta version. Of course, this is Google we're talking about, so this thing could be in Beta for years. You know how they like to make sure it's perfect before taking a project out of beta. (Microsoft seems to be at the other end of the spectrum when it comes to beta testing and quality control.)

Anyway, you can check out Andy's Review of Google desktop search, or go download it yourself and try it out.

"Internet media company Yahoo Inc. reported on Tuesday a quarterly net profit more than tripled from the year earlier, boosted by stronger online advertising and the sale of its stake in Google Inc."

Article from Reuters.com

Search Engine Strategies is coming up in December in Chicago. They recently announced the schedule. I personally don't know if it's worth the high price tag. I went to SES in Dallas a few years back and was kind of disappointed in the level of information that was discussed. I'm sure it could have come a long way over the past 3 or 4 years, and we all know the industry has changed quite a bit since then, but I'm going to save my money and just go to WebmasterWorld's conference, which I've heard provides more hands-on useful information at about 1/3 the cost of SES. If you've got the cash and time to go to both, then go for it. I'm sure it won't hurt, and those types of events are always great for networking within the industry.

The second part of a report from Atlas DMT shows some interesting results. Consistent with previous test results I've seen, it shows that conversion rates decline significantly for clicks on lower-position paid placement ads for high volume keywords. The interesting twist is that conversion rates were actually higher for the lower position ads on low-volume keywords. This isn't some dinky little test, either. They analyzed July and August 2004 traffic representing 41,460,000 clicks and 408,000 keywords. That seems likemore than enough to provide a valid sample for this test.

Obviously there is going to be variations by industry and for myriad other reasons, so you'll want to run your own tests to see if it holds true, but I was surprised to see these results turn out like this.

In their report, Atlas offers the following advice for marketers:

• Balance your efficiency and volume goals. When setting target values for return-on-adspend (ROAS) or cost-per-acquisition (CPA), you should factor in the volume of conversions. You may attain the best ROAS at rank 10, but may be disappointed at the total number of conversions generated at that rank.
  
• Pay a little extra for a higher rank. Conversion rates for your highest volume keywords will drop by rank. If rank bids are close together, as often happens, it will likely make sense
  to pay the additional few cents to move up in rank. Analyze your own data to identify what bid increases are justified by rank based on the resulting increase in conversion rate.
  
• Test lower ranks for low volume keywords. Conversion rates for your low volume keywords may very well increase as rank falls. We observed this trend across both search engines. This insight presents an opportunity for marketers to lower their costs while sustaining conversion volumes for their low volume keywords.
  
• Look at your own data. Many factors contribute to the likelihood of a conversion; therefore a study such as this is primarily useful for forecasting and comparing your results to industry benchmarks. What we have shown here may significantly differ from the trends revealed by your own data. Every marketer will find unique trends specific to their circumstances. Finding the sweet spot for conversions for your ads is about measurement and experimentation.
  

Read the report for yourself

I came across an informative article by Kevin Kantola about optimizing PDF files. The search engines that matter these days are all indexing PDF files, so why not optimize them if you've got a whitepaper or some other marketing piece in PDF format that you want to show up in the search engines. PDFs still aren't the preferred format for optimized documents (ever hear of HTML?), but it can't hurt to optimize any and all documents you want people to find and have access to.

MSN's Search technology preview is up and working again. It's been off and on since my last post about it. It appears that they've updated their index very recently, with recent blog posts and other new content showing up. Could it almost be time for MSN to unleash this beast? I think we're getting close.

Everyone who cares about this stuff is well aware that google is currently updating PR in the toolbar. The InsideGoogle Blog has a list of links to forums covering the response to the updates.

MSN cracked open the door for another sneak peek into their new search engine. It's available at the same URL as before: http://techpreview.search.msn.com

Does this mean MSN is finally getting close to changing over to their own search engine? I'm not sure, but at least it's showing some progress. I think they feel left out because they've had this project waiting in the wings this whole time and Google and Yahoo keep cranking out new products like local search, personalized search, and whatever else. They must feel like they're being left behind (did they just barely realize that?). Well, it is Microsoft, after all, so don't count them out just yet. It's obviously going to change the search marketing industry significantly whenever they decide to flip the switch with this new search engine.

One of the early investors in Google and current board member, John Doerr, said Tuesday that he doesn't believe Google will enter the Web browser market, despite widespread speculation that they will.

Read more about it from News.com

When Bill Gross makes an announcement, people listen. The next big thing to come out of IdeaLabs is a search engine with quite a few unique twists. The name is familiar to online search veterans, but Snap appears to be anything but familiar in what it offers.

Here's what Bill says about Snap:

We built Snap to save you time. We think we can cut the time it takes to find what you’re REALLY looking for in half! Today we’re unveiling Snap, with three bold innovations to improve the way you find things on the web.

The first is USER CONTROL, where you get to change the order of search results, refine search results instantly, and hone in on exactly what you’re looking for. We think you’ll love it. Give it a try. It works on any search, just refine by entering filter words in the field above any column, or click on any column to sort. Try it right now, say for digital cameras or laptops.

The second new feature is USER FEEDBACK. We take into account what happens AFTER people click on search listings at our site AND others, to use as feedback on the relevance, and get you better results up at the top. This helps you avoid dead end searches, and saves you time. We also can figure out, based on millions of users, what people are really looking for so we can put custom formats on search pages where previous users signaled their “intent” by their follow-on searches. Try it right now, say for cars, Walmart, or mp3.

The third innovation is complete and total TRANSPARENCY. We want to reveal every action and transaction at the site, so you know what we are doing and what other users are doing. We even are revealing OUR REVENUES. Our conviction is that you get better results because transparency prevents advertisers or others from gaming the system.

We hope you like all of these things. Please keep in mind this is just the tip of the iceberg. We’ll keep improving technology while keeping you up to date along the way. We really want to hear what you think about Snap – how it saves you time, and what we could do better.


I haven't had much time to mes with it yet, but the concept is somewhat intriguing. Go check out the beta version of Snap and see what you think.

Yahoo continues to rev its search engine with the announcement of the beta version of My Yahoo! Search (requires myYahoo login). Yahoo's personalized search lets you save web pages to the "My Web" section and block out the ones you don't want to see again. It also lets you add notes so you remember why you saved that particular page to "your web". You can also share web pages with friends and then go back and search through your saved search results.

Everyone's been talking about personalized search lately, but I think Yahoo will be the company to take the concept mainstream. They've got a huge userbase who are loyal and already have to login to check email and get their customized homepage in MyYahoo anyway, so it will be a smooth transition to get people to use the personalized search features. I'm sure we'll see a lot of changes and evolution in the way these personalized search tools work, but it seems to be moving ahead full speed.

I'm sure there are a lot of people out there who don't trust Yahoo enough to store their search information like this. I personally see a lot of value that can be derived from this type of personalized search. It will take a while before people fully understand how this works and get used to the added functionality. I don't think people are going to start saving web pages on every search result, but this can be a huge help for in-depth research projects. It makes it easier to pick up where you left off, rather than having to start over every day.

Sempo is conducting a survey to try to get a finger on the pulse of the search engine marketing industry. If you're reading this blog, chances are you are somehow involved in the industry, so go take the survey.

Local search is all the rage lately. From what I've seen most local search is simply an online version of the yellow pages, with some "local ads" above the listings. I think local search has the potential to be big, but they've got to start providing some real value if they want people to care enough to keep coming back. I guess they've got to start somewhere, though, and Yahoo seems to be making a charge at the local search market as they unleashed Yahoo! Local out of beta today. What does that mean? Not much. But it does go a long way in illustrating two different philosophies on Beta testing. Yahoo is all about getting out of Beta so they can claim they are the first ones there and get as much press coverage as possible. Google, on the other hand, believes in keeping their products in beta for a very long time, so if anyone complains about quality of results or relevancy, they can simply say, "Oh, it's still in beta testing." (Case in point: Google News).

Beta or not, I actually like Google's local search better than Yahoo's.

Google recently expanded their shopping search efforts with the luanch of the UK Version of Froogle. I've been a big fan of Froogle since the first time I saw it. As a marketer, it gives me a another chance to get my products in front of potential customers using different methods than traditional Google site optimization (and without spending tons of cash on adwords). As a consumer, it gives me the search power of Google combined with the utility of a shopping search engine. What more could you ask for? Sometimes you get funky results that aren't quite what you're looking for. That's probably why they still say it's in Beta...they have a perfect excuse in case anyone complains. Same thing with Google News--how long's that thing going to be in beta?