I was thinking about my last post Mining Keyword Data for Financial Acquisitions and felt I had not really gone nearly as far as I could with the data in my keyword referrals that's being stored in Google Analytics.  In a way, I'd prefer to focus on this kind of information as opposed to regular SEO, which I find boring.

For example, looking at buyer or acquisition behavior I'd look for phrases that contain "buy", "bought", "sale", "acquire", "cost", "price", "pricetag", "behavior", etc.  Here's what I came up with from Google Analytics:

1. Keywords containing "buy" that might indicate buyer or acquisition behavior:

So…who is going to buy Visual Sciences? Someone wants to know; there were 3 visits, one on July 12, August 19th and August 26th. The landing page visited (which shows the page on my site that comes up when someone executed that query in a search engine (probably Google):

 And the "Locations" where people came from:

The locations, in this case, wasn't that helpful so I looked for the city each search came from:

It's all foreign visitors, so whoever wants to know who is going to buy Visual Sciences, it sure isn't anyone I know and they probably don't live around here; however they were all "new" visitors, meaning they're not regular readers (RSS Subscribers) or people who visit my site regularly.

Microsoft buys hopstop was a search that happened on August 28th but it resolved to a page I wrote about Hopstop:

Hmm… I think I got something here….. HopStop, which I said Google should buy earlier this year:

"..Transit planner HopStop, however, takes on these larger metros, providing coverage for Boston, Chicago, San Francisco, New York, Washington D.C., and several cities in New Jersey. For each city, HopStop will plot directions and estimate travel time by foot and public transit (bus / subway) between two or more addresses. It also comes with some other really useful features. You can choose which travel method you would prefer more, walking, busing, subway, or a combination of the three. Also, since you don’t always know the street address of your destination you can pick it from a map. "

When I went to the Etsy.com party in Brooklyn a couple of months ago, I used HopStop to get directions  – and they were pretty good.  "

Someone is wondering if Microsoft will buy HopStop… question is ….who?  Is it someone from a Financial Institution?

It looks to me like www.Pfsweb.com is a Fortune 1000 Global Outsourcing firm and it's not really clear if this is in someway related to purchase behavior or if it's some kind of study, that suggests maybe Microsoft should buy HopStop.  The individual who searched was located in Addison, Texas, at the Corporate Headquarters.

Let's say it's just that ….. in someone's mind, just knowing that might be valuable information.  For example, it's known that Google is beefing up on local search using public transportation, but it appears they are building that capability themselves…at least, it looks that way to me, else they'd have bought HopStop..but they didn't.    Maybe Microsoft would want to buy HopStop….and if I were a broker or speculator …. maybe I'd put some money on it (both that they will, and that they won't).

Now, all of this is public information, I have every right to it, they came to my blog from a keyword search.  But what if I worked at the Wall Street Journal or the New York Times, or Associated Press…think of the data I'd be sitting on. 

I wouldn't even need to be anyone high up on the food chain, I could be a "web analyst"…. I could pull this data.  Question is…would I know enough to know what to do with it … is it enough information to be actionable.

Probably not….there's probably not enough actionable data here to take any action, except if you were someone who knew enough about how markets work…that data might then lead to further inquiries that might lead to an action…and action one might not take otherwise.

I think there's an awful lot of valuable data, "gold", in Referrer Logs and in Local  Search  Query Logs too, but it takes a bit of datamining to get at it.   However, I think this is much more rewarding and some companies are specializing in datamining precisely this information.

I was talking to a reporter from the Boston Globe recently on how Search Data might be used for Financial Gain.  Immediately I thought back to the situation with HitWise, when I saw instances of people searching on terms like "HitWise Sold" more than a year in advance on when it happened, earlier this year, to Experian.

You could say…that when I traced back those referrals to my site, it often went to a financial institution … meaning that, in some sense, my search query log could be used as a possible barometer of what might be.  

I guess anyone's search query log could be used that way – if you had enough searches on your site.  A large newspaper site, like the New York Times, would be the perfect type of site to datamine for acquisition data….. I'm not telling anyone to do this … just that it could be done – and how well depends on …well.. how good you are at searching.

Here's an example – HitWise – an obvious example

The data is being shown is from February 2006 till today – September 20th, 2007.  The first spike was on August 20th, 2006.

And look at who is visiting (asking if "Hitwise is sold"):

Unfortunately, Google Analytics just can't hone in deeper by the actual visit (ie: I'd have to set my range to August 20th, 2006 to get details of the first visit for "Hitwise sold" but you can be it was probably one of the three "network locations" listed above.

So what do we have here….. someone from a financial institution … Lehman Brothers…..is wondering about HitWise being sold, way before anything is formally announced about HitWise being for sale.

Of course, I knew, because I was watching the data coming in …. but I bet, you could go a lot further than I just did.

Just a thought on the last day of September, 2007.

Having a gas reading Sebastian Wenzel's post at WebAnalyticsBook.com on Web Analytics 4.0:

"….Analytics 4.0 (OK I stop this number nonsense) is going to be an web analytics industry, that is not tied to the internet and works across media channels. In the future, I expect web analytics to merge with other industries e.g the Digital signage industry.

Digital signage is a segment of out of-home advertising in which content is displayed on a digital screen rather than printed sign. The content can range from text and still images to full-motion video."

Not sure Sebastian is joking or not…. I sure hope 5 years from now I'm not going to be called Web Digital Signage Analyst..although, who knows?   I mean, I can see us collecting metrics on it, just like anything else we can get data on, but I'd probably see Digital Signage as merging with Web Analytics than the other way around.

"…A company in Sacramento, CA called SmartSign Media is using interactive technology to gather demographic profiles about passengers as they drive by and respond with advertisements tailored to their preferences. Data on how many cars listen to hat FM radio stations is collected and ad messages are adjusted in response to the demographic profile of the cars’ passengers. HumWare Media based in Colorado uses polling where consumers can respond to a question posed on a digital sign…” [Source]

The technology looks interesting though I'm wondering how a sign can serve a different image to each demographic when you have a mix of demographics in front of it at any one time. 

Yes, I think there's probably ways to do it …but we begin to create a world that is so customized that no two people will end up seeing the same thing (ie: a sign) in the same exact way…and I wonder if that's what we really want, after all.