search engine marketing

Google’s Instant Search: Instant Irritation?

Phillip Nones Uncategorized , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,

Google's Instant Search is a Non-StarterHow many of you have been noodling around with Google’s new Instant Search functionality since its unveiling earlier this month? I’ve spent the better part of a week working with it, trying hard to keep a “completely open mind” as to its benefits.

I’ve finally came to the conclusion that … I can’t stand it. I’m a pretty fast typist, and generally know what I’m searching for. I really don’t need Google “pre-anticipating” search results for me, and find the constantly jumping search results window extremely off-putting to the point of distraction.

I gave Instant Search a full week … and couldn’t take it anymore. I’ve now elected to turn it off completely.

Wondering if I was the only one with this view … it certainly didn’t take long to find out that there are a great many people out there who feel the same way. You can use Google search (either the “instant” or “traditional” will do fine) to find any number of blog posts and user comments about Google Instant Search that are just one notch shy of mutinous — and hardly genteel in their choice of language. (A few examples can be found here and here and here.)

If the comments by disgruntled users are to be believed, Bing/MSN may find itself with a nice little bump in search volume market share by the end of September.

And if that actually happens, Google Instant might die a quiet death – which wouldn’t be the first time Google laid an egg in its “throw-everything-against-the-wall-and-see-what-sticks” approach to product development.

But if Google Instant does gain traction … there are some negative implications for search marketers as well. Many companies seek to structure their online marketing campaigns by determining the optimal amount of spending on search advertising, display ads and social media. The key to success in this endeavor is undertaking a process that examines the millions of cookies and billions of clicks that are made by web users, along with factoring in other elements like geographic location and time of day.

All of this information is weighed against the cost of various ads and the likelihood of success as they are served to the user. That’s determined by running regular models of millions of keywords and word combinations, judging the relative costs to determine the optimum frequency. For some of the most aggressive marketers, these models are run once or twice daily.

The advent of Google’s Instant Search scrambles all of that, because it makes the process even faster and more hectic than before. As those of you who have experimented with Instant Search know, you start seeing “suggested” search results with just the first one or two keystrokes … and those choices continue to change with each new keystroke made or movement of the cursor down the list of Google’s suggestions. For marketers, the result is a lot more velocity on the ad side – and more price changes.

As proof of this, within the first few days of Instant Search’s launch, sites that Instant Search recommends after the first one or two letters are typed into the search box – “Mapquest,” “Ticketmaster” and “Pandora” are three useful examples – were experiencing significant increases in traffic, whereas their hapless competitors were not.

If that’s what is happening with the big boys, where does this put smaller businesses? The answer is obvious: They’re going to get squeezed big-time … and as a result, their search advertising costs are going nowhere but up.

Mighty sporting of you, Google.

$100 cost-per-click on Google AdWords? It’s already here.

Phillip Nones Uncategorized , , , , , , , , , , , , , , , , , , , , , ,

How much is one clickthrough to your web site worth? If you’re a legal firm specializing in bringing mesothelioma cases to court, it turns out it’s worth a lot.

In fact, the search term “mesothelioma” was the highest-priced keyword in the U.S. during the third quarter of 2009. That’s according to a recently-released AdGooroo Search Engine Advertising report.

Just how expensive? For Google’s AdWords program, the highest price paid for a #1 ranking on that search term was a whopping $99.44 per click. (Over at Yahoo, the high figure for this paid search term was a little less rich: $60.68 per click.)

One wonders how many times the advertisers have actually had to pay out this king’s ransom. Whether it’s for a few or many clicks, it’s clear that some legal firms recognize a highly lucrative revenue opportunity in filing mesothelioma lawsuits related to asbestos and lung cancer.

Interestingly, the highest paid search term in Bing’s paid search program isn’t “mesothelioma,” but rather “auto insurance comparison.” At $55.20 per click, the dollars aren’t as high, but it would seem like the potential payoff isn’t nearly as great, either. After all, there’s a pretty big difference between a multi-million dollar legal verdict and the value of an automotive insurance policy.

But beyond the eyebrow-raising stats of these extreme examples, the larger issue is how much more costly search advertising has become in recent times. A few short years ago, it was common to talk about search terms costing an advertiser 50 cents or $1.00 per click. Now, those same terms are far more likely to cost $2.50 or more.

Google, being the 500-pound gorilla in search engine marketing (SEM), has certainly contributed to the price inflation. That’s one reason why many are rooting for alternative search options like Bing to succeed (dream on).

More fundamental to the increase in keyword click pricing is the realization that advertising to people at the precise time they’re searching for particular goods and services is a far more effective way to engage customers and prospects and drive actual sales.

And that’s even more the case compared to trying to get their attention or otherwise “intrude” on them when they’re online for other purposes. The abysmal clickthrough rates experienced for banner advertising bear this out.

But paying $100 per clickthrough? That does seem excessive – even for ambulance-chasing lawyers!

Click fraud: How much is really out there?

Phillip Nones Uncategorized , , , , , , , , , , , , , , , , ,

One of the knocks against pay-per-click advertising is concern about fraudulent clicks being made on online ads that cost advertisers money and drain their account budgets needlessly. And while Google, Yahoo and various online publishers have long held that their SEM operations can detect patterns of fraud and then credit-adjust advertisers’ accounts accordingly, that hasn’t mollified the skeptics at all.

And now SEM critics have new ammunition in the form of two click fraud reports issued in July by Anchor Intelligence and Click Forensics, two of the industry’s leading traffic auditing and traffic quality management firms. Researchers at both companies have discovered that “scripted” programs that click on ads increased in volume during the second quarter of 2009.

Click Forensics estimates that the overall average click fraud rate was nearly 13% over the quarterly period. According to the firm, this also included an ominous rise in “collusion fraud” on advertising networks. That’s when publishers rotate IP addresses (botnets) to click on ads on their own sites to generate inflated commissions from unprotected ad networks. Many ad networks have difficulty differentiating these attacks from valid clicks.

Based on these results, Click Forensics estimates that the amount of money lost yearly due to click fraud exceeds $4 billion. And while a large chunk of those dollars are presumably reimbursed to advertisers in the form of discounts or rebates, it is impossible to know what portion that amount actually represents because SEM program providers don’t share that information with the outside world.

Anchor Intelligence reported even higher rates of attempted click fraud during the second quarter 2009: nearly 23%.

Where are the nefarious attacks coming from? Richard Sim, Anchor Intelligence’s vice president of product marketing, says, “Vietnam stands out in terms of the fraud as a percentage of all traffic. Nearly one out of every two clicks from Vietnam was registered as click fraud.” That’s nearly double the rate of attempted click fraud found by Anchor Intelligence for the next highest ranked countries – Canada at ~28% and the U.S. at ~26%.

What this says is that click fraud is very much with us, despite all of the best efforts that go into trying to root it out. This should be taken into consideration by advertisers when planning and executing an online advertising program. And it wouldn’t be a bad idea to factor in a 15% or 20% “degradation” factor on all advertising goals and results when evaluating clickthrough rates and calculating ROI.

The good news is that, even with this reduction factor applied, when you compare search engine advertising against alternative forms of promotion, it’s still one of the better buys in the business.

Searching for effective lead generation and conversion.

Phillip Nones Uncategorized , , , , , , , , , , , , , , , , , , , , , , , , ,

In the current business climate, companies are relying more than ever on new sales opportunities to replace business that has been lost with current customers. And it’s pretty clear by now that “search” has emerged as the form of online promotion that generates the best lead generation and conversion results — outstripping other e-promotional tactics such as online display advertising and newsletter sponsorships.

This isn’t surprising, of course. Search advertising captures the interest of online viewers precisely when they’re in “search mode” for specific products and services, rather than when they’re just surfing the ‘net for news and updates.

(In fact, some advertisers have come to believe that even print advertising outperforms online display advertising. That’s because readers are more likely to browse all the way through print publications. Compare that to visiting informational web sites where viewers are far more prone to selectively pick and choose the pages that they open. A well-placed display ad on a “new technology news” page, for example, might be invisible to the vast majority of viewers who come to the home page and then decide to click through to only one or two additional pages on the site.)

But back to search. Many advertisers wonder which is most effective: gaining high “natural search” rankings that occur based on the content of the web site, or opting for pay-per-click search listings such as Google’s AdWords program with their entries on the right side of the screen.

As it turns out, both tactics have their pluses.

In fact, a new year-long study that ended June 30, 2009 of more than 25 e-tail web sites by Engine Ready, Inc., a search engine software development firm, found that visitors who clicked through to the sites from paid search ads were ~50% more likely to make a purchase, compared to visitors who came to the same sites via clicking on a natural search link.

Specifically, Engine Ready discovered that the conversion rate from pay-per-click links measured 2.03%, while the conversion rate was only 1.26% from organic search clickthroughs.

On the other hand, various research studies conducted over the past few years demonstrate the clear popularity of natural search listings over paid search listings. It’s been shown pretty consistently that around two thirds of total clicks are made on natural search listings, compared to just one-third on pay-per-click listings.

So the key takeaway is that any marketing program worth its salt incorporates search marketing as a key component. And in most cases, that effort should encompass search engine optimization for natural search rankings along with a pay-per-click advertising program.

Bing Search: Pike’s Peak … or Halley’s Comet?

Phillip Nones Uncategorized , , , , , , , , , , , , ,

Well, it didn’t take long for the marketplace to render its verdict on the Bing search engine phenomenon. Fueled by a multi-million dollar advertising rollout plus an aggressive PR push, web tracking service StatCounter has reported that Bing actually vaulted past Yahoo to become the #2 search engine … for one day.

That’s right. According to StatCounter’s data, on June 4th, Bing captured over 15% of the U.S. search share market, while Yahoo had only around 10%. By the next day, Bing’s share had dropped below 10% while Yanoo notched up a point to 11%. And by Day 3, Bing’s share had fallen still further to just under 7%.

Think it couldn’t get worse? The day after that, Bing was mired below 6% share.

Similar results were recorded worldwide.

What’s behind the primal shrug that Bing seems to have met in the marketplace? Certainly, all the PR hype was successful in getting people curious enough to click through and do a bit of tire-kicking. But it’s obvious that most weren’t particularly impressed by what they experienced, despite the fact that Bing does provide some user-friendly features not available over at Google.

But that’s not nearly enough for success. Google’s users are, by and large, quite satisfied with the search experience. It’s what they know. It’s comfortable. And unless there’s a compelling reason to switch — to change deep-seated habits — most people simply aren’t going to play ball … whether you put millions of dollars in advertising behind your pitch or not.

The folks at Google might have been shaken a least a bit on June 4th when their market share of search dropped to 72%. But they needn’t have worried. Four days later, Google’s share was back up to 80% — where it had been to begin with.

Next case, please?

More Action on the Search Engine Front

Phillip Nones Uncategorized , , , , , , , , , , , , , , , , , , , , , , , , ,

Bing logo designWolfram Alpha logoDespite the fact that Google has proven itself to be all but immune from threats posed by competing search engines, hope springs eternal. Within the past couple weeks alone, two new challengers have emerged, accompanied by much fanfare in the business press.

Microsoft takes yet another swipe at Google with its new Bing search engine. Based on an earlier one called “Kumo,” some industry observers — though not all — believe it is a pretty good competitor. Reviewers are particularly pleased with the presentation of refined versions of search queries. Bing also features a rollover display of each link’s content, allowing you to see how useful it will be before clicking through to the site.

The search engine also appears to index more recent “breaking news” items, whereas with Google, those results are not shown unless you click through to Google News — an extra step.

The big question is whether Bing will be able to wean web users away from their habit of searching on Google as their default choice. Certainly, Microsoft is putting some serious promotional dollars behind the launch — upwards of $100 million according to Advertising Age magazine. But based on the tea leaves, a wholesale change in search behavior seems unlikely. Search habits aren’t going to change dramatically unless there is a dramatic improvement in the effectiveness and speed of search activity. Fom what we see of Bing so far, we’re talking about improvements nibbling around on the margin rather than big sweeping change.

But “big sweeping change” just might be the recipe for Wolfram/Alpha, the other new entrant in the search engine sweepstakes. That’s because W/A isn’t actually a search engine in the classsic sense. Instead, its developers refer to it as a “computational knowledge engine” that uses complex algorithms to search databases to come up with answers to questions, rather than presenting a list of sources where the answer might be found. It can report some really cool factual results just based on the user typing in, for example, a date range, several city names, or an animal species.

The key difference between Wolfram/Alpha and Google is that W/A does not index web pages. Instead, it draws answers from a wide range of information-packed databases. So if you want to know the number and magnitude of hurricanes hitting North America in the past 15 years, you’ll get a specific answer rather than being presented with a series of web links wherein you might find the answer to be hiding.

Some observers see the potential for W/A and Google to team up rather than compete against one another. After all, what they do isn’t directly competitive, but in more respects complementary. And in an interesting twist, it turns out that Stephen Wolfram, the ~50-year-old computer scientist and developer who created the software platform upon which W/A is based (called “Mathematica”), once supervised a summer intern by the name of Sergey Brin — who would go on to develop Google with partner Larry Page.

Sergey and Stephen teaming up once again would be quite the coincidence … or would it really?

Search … and destroy? Nah.

Phillip Nones Uncategorized , , , , , , , , , , , , , , , ,

New statistics published earlier this month by Hitwise show that Google continues merrily on its way to even greater heights of dominance in the search engine field.  Despite the Don Quixote-like efforts of other search engines like MSN, Ask and Yahoo to take a run at Google’s position, the latest stats show that Google’s search engine is as popular as ever.

More popular, in fact.  The numbers reveal that Google’s share of search activity has now risen to 72% versus 67% a year earlier, whereas the others continue to decline.  Yahoo is in second position, but getting 21% of search share is about on par with H. Ross Perot’s vote percentage in the 1992 presidential election – all hat and no cattle.

More startlingly bad is MSN’s performance at around 7% of search activity, because they’ve been trying hard to make a dent in Google’s position. Keep on trying, gents.  Maybe you’ll break 10% share before long, although I doubt it.

Does any of this come as a surprise?  After all, people are creatures of habit. And when a habit gets as big as this, it’s really hard to break.

Also, most people typically take the path of least resistance. And when it comes to search, isn’t Google the easiest path?  Simple visual layout … easy to use … robust results.  What’s the point of going anywhere else?

UPDATE – 4/1/09 – As if on cue, another search engine bites the dust.  Wikia has announced it is closing down its Wikia Search project.  Introduced to great fanfare last year, Wikia was intended to be a user-generated, open search engine.  The problem?  Wikia Search was simply not generating any sort of worthwhile volume.  In fact, traffic was running about 10,000 unique users per month.  That’s just a blip on the screen — and certainly disappointing considering the success of other initiatives like Wikipedia and Wikia Answers.  Further proof that to be first in cyberspace with a good idea and good execution is a huge advantage … and to be fourth or fifth is considerably more difficult, even fruitless.