This was once known as customer behavioral profiling, although some I/O psychologists would probably have you believe otherwise that there are far deeper sentiments that consumers attach to the everyday and ordinary that surprisingly have monetary value. The most common example of predictive modeling in the financial services is credit scoring.
In the USA, credit scoring is a basic data analysis of a person's (or business') ability to maintain good credit standings and pay off debt. The higher the rating, the lower risk the person is to taking on more debt, or so the working practices go. Such factors like how many lines of credit a person has, what the revolving balances are, how much is paid off at the end of each billing cycle, are there any missed or late payments, current balance on your primary savings account, etc. Oddly enough, you actually have to take on more debt and faithfully pay it off to raise your credit score. The top major credit bureaus that amass credit data on individuals are Equifax (ScorePower), Experian (PLUS score), and TransUnion. If you are responding to an online ad about getting your free credit report, it's very likely that it's a consumer program managed by one of these companies.
So who uses this scoring methodology? Banks, of course, so they can lend you money for a home loan or extend credit to you for a HELOC; Utility companies (Gas, Electric, and Water services often require $25 deposits for customers who have poor credit); Apartment managers; Cable service providers; Private and public companies who look at a job applicants' credit history.. the list goes on.
Consumers are kind of screwed to begin with since each "hit" on their credit report influences their scoring as well. There isn't enough transparency with how scores are calculated to see if the report was accessed for financial reasons (buying a new house or car, applying for a platinum credit card) or for living reasons (relocation to a new city, getting gas service hooked up to a new apartment). And, there are hardly any resources for consumers to realistically "fight the system" when bad credit history is slapped onto their virtual account. I digress.
Predictive modeling helps companies identify their prime target for new or existing products and services based on current or historical purchase behavior the customer has exhibited. Looking at aggregate data (multiple consumers, multiple purchase points), you'll start to see trending in what products or website areas are more popular with consumers who bought X and are considering the purchase of Y. Amazon.com
already does this and their metrics are a value-add for everyone looking at it. The consumer is given a feed of popular products purchased by other registered site users. The company's marketing team has a lot of supporting data statistics to provide to affiliate merchants and advertisers, and guest users can see where else on the site to look for things most similar to their search.
Let's say that you already have a website because if you didn't, I'd have to write a series of blog posts explaining how to do it or refer you to one of several tutorial sites that can help you build one from scratch. At some point you'll be pitched by an SEO vendor to improve your website's rankings with various search engines. WhiteHatSEO.org offers some guidance for "ethical" SEO practices in its member codex. Here are a few of their tips.
- Don't use shadow domains.
- Don't use doorway pages.
- Don't participate in link schemes designed to increase clients site's ranking or page rank.
- Don't use hidden text or hidden links.
- Don't employ cloaking or sneaky redirects, which is used to deceive users or present different content to search engines than it display to users.
- Don't send automated queries to search engines.
- Don't load pages with irrelevant words and phrases.
- Don't create multiple pages, subdomains, or domains with substantially duplicate content.
This all good, but if you're a sole proprietor or a company looking at SEO for the first time, how would you know who was legit and who wasn't? And, given how diverse SEO methods are these days, which ones are the right ones to use? Is this a task that can be handled in-house or does it need to be outsourced?
Key SEO goals are intended to make sure that:
* Pages are indexed and a good site map exists
* Pages are well-structured with headers and subheaders, content-appropriate meta tags, etc.
* Descriptive page titles are used
* Keyword-rich inbound links are used
It looks a lot like a new spin on basic webmaster duties.
The easier you make it for a user who knows almost nothing about your business to find your website in the myriad of millions of sites, the better. Unlike the big yellow book from the telecom company that shows up every year to unreceptive household doorsteps, SEO allows you, the web domain owner, to be in control of advertising, product or service positioning relative to next best complements, and manage who visits and how. In that respect, SEO is more cost-effective than print advertisements. Opting for paid SEO is entirely up to your company's customer acquisition strategy.
At some point I'll follow-up with a post about black hat SEO and why ethics has nothing to do with whatever technique is used but how it is used that can get sites delisted from search engine indexes.
Remember: The only good SPAM is when its fried up with rice.
Related links:
White Hat SEO tips for bloggers
Questions to Ask a Paid Search Vendor
Questions to Ask SEO Consultants
Glossary:
White Hat
Shadow domains
Doorway pages
Link Farms
Here's a cheer for all the consumers out there. *Yay!* Finally, it seems that someone has been reading all the FTC complaints and companies will have to comply if they are to stay compliant with these new proposed operating rules with regard to how a customer unsubscribes from a list. I'm sure at least one of these scenarios has happened to you when you tried to subscribe from that bacn list.
The new rule provisions address four topics:
(1) an e-mail recipient cannot be required to pay a fee, provide information other than his or her e-mail address and opt-out preferences, or take any steps other than sending a reply e-mail message or visiting a single Internet Web page to opt out of receiving future e-mail from a sender;
(2) the definition of “sender” was modified to make it easier to determine which of multiple parties advertising in a single e-mail message is responsible for complying with the Act’s opt-out requirements;
(3) a “sender” of commercial e-mail can include an accurately-registered post office box or private mailbox established under United States Postal Service regulations to satisfy the Act’s requirement that a commercial e-mail display a “valid physical postal address”; and
(4) a definition of the term “person” was added to clarify that CAN-SPAM’s obligations are not limited to natural persons.
More details - http://www.ftc.gov/opa/2008/05/canspam.shtm
