Optimization is the effort to try to make something the best that it can be.

You've likely seen television shows featuring chefs who help struggling restaurants by providing an objective opinion at how the kitchen and the front staff work. Those efforts might involve shrinking a much too large menu to something that can be managed.

They may require a look at the books, and how much is being charged for menu items. They usually involve updating decor to something bright and clean and simple.

When we do search engine optimization, we're doing more than making sure that permanent redirects are set up correctly, or that search engines spiders can crawl through the pages of a site that we want indexed without any impediments.

While we want pages to include keywords in the right places, and create persuasive and engaging titles and meta descriptions that might convince potential visitors to click through to a page when they see them out of context in search results or in social media shares, there's more to SEO than that.

As we begin to work with a client, we aim to get a sense of the objectives of their site, the audiences they hope to reach, and the language those audiences use to search with when trying to find what a site owner might offer.

During that process, we may find that sometimes a site doesn't reach out to the right audience, and doesn't describe all the goods or products or ideas their pages could be about.

As a partner, it's our responsibility to give our clients the best chance possible to shine, and to be seen by people who care about what they offer.

We might recommend changes to the information architecture of a site, or for the creation of new content, or both. We might suggest new ways to reach out to audiences through the use of blog posts or social media activity.

We bring ideas for new recipes with us, new decorations and changes in the way that staff might interact with clients. Often we bring a different and fresh objective, and that's something that can make all the difference.

Pages I enjoyed this past week:

Stories provide a way for someone to communicate an idea within a framework that other people can appreciate and understand. This is true in sales, in software development, and in SEO. That's part of the reason why case studies can be so valuable. It's at the heart of helping people understand why we might recommend certain changes to their sites. A story can help describe and define the value in following a certain path, and described in Why We Need Storytellers at the Heart of Product Development

In telling the story of you and your company, there's some great advice in the blog post 12 Simple Ways Marketers Can Humanize Their Brand. Most of it applies well to more than just marketers.

What happens when a developer leaves behind his comfort zone to take on a whole different set of challenges in the marketing field? In What A Hacker Learns After A Year In Marketing is a great story about the efforts one man undertook so that he could broaden his experiences in the pursuit of starting his own business. My favorite line:

Good marketing is a product of the same inputs as good code; long hours, sweating the details, and the judicious application of experience doing it the right way.

David Harry tackles the topic Are you ready for the next Penguin assault? The update is about more than just the links pointing to a site, and David gives us a closer look at other issues involved in Google's update. If you haven't started looking past links, this is a good introduction to some of the other issues involved.

On the more technical side of things is a whitepaper from researchers at Carnegie Mellon University and Google that actually was presented in Korea in early July in a conference on natural language processing, titled Reading The Web with Learned Syntactic-Semantic Inference Rules (pdf). If you're interested in some of the ways that Google might be learning more about the Web while building their knowledge base, this one is worth trying to untangle.

For example, Google may looks at a knowledge base such as Freebase to find out information about a specific person, and Freebase might contain some information but contains gaps in what it knows. Google might look at patterns on the knowledge base and elsewhere on the Web to understand that such a gap exists, and to try to fill it.

A very simplified example of such a rule that might be created would be, "If person was born in town and town is in a certain country then the person is a national of that country.”

Retro Post of the Week

In 2003, three Stanford University students were planning upon building a next generation search engine to rival Google, Yahoo, and Altavista. Taher Haveliwala, Glen Jeh, and Sepandar Kamvar suggested that their search engine would be different by bringing personalization to search. They spelled out the details in their Business Plan for the Kaltix Search Engine

The executive summary from the plan tells us:

We have developed a technology for personalizing web search that will allow a dramatic increase in revenue as compared to current search engines by providing more effective search and more targeted advertising. We propose a company (Kaltix) that will develop a next-generation search engine for consumer use.

The plan talks about a recent acquisition of Altavista by Overture and of Inktomi by Yahoo!, and how Microsoft had earlier that year hired 50 new employees for their MSN Search team, and committed much of the faculty at Microsoft Research to Web search. At this point in time though, none of the search engines offered personalized search.

We're told that one of the values of offering personalized search is that searchers would be less likely to switch from one search engine to another if one of the search engines took advantage of their previous searches to offer more targeted and better results. That seems like an assumption that hasn't changed much in the decade since. So this is the opportunity that Kaltix is trying to meet:

A search engine that tailors results to a particular individual's information need, utilizing the user's query, personal preferences, and search context, will have a significant edge in the competitive search market by offering an improved user experience. Furthermore, considerations such as increased traffic, targeted advertising, acquisition of user preference information, attachment, and user transfer, make a compelling business case for personalized search.

Such personalization, we're told, becomes affordable based upon technology developed by Kaltix, as partially described in a number of whitepapers developed by the team, and listed in the business plan, including a topic sensitive PageRank.

In the business plan, we're told that a functional prototype had been built, but that it didn't quite scale to what the major search engines where doing yet. It was only capable of using a 50 million page web crawl, while the other search engines were crawling 200 million pages a day. It would also need to be able to handle at least 1 million hits per day, the top search engines of the day were handling over 200 million hits daily.

To be able to reach 200 million pages crawled a day and to be able to take care of 1 million queries per day, the plan estimates that it would take 6-12 months and require $100,000-$200,000 in capital for machines and founders' salaries.

The business plan for Kaltix was published on June 16, 2003. On September 30th, 2003, a Google press release came out announcing the acquisition of Kaltix by Google.

Kaltix became part of a next generation search engine, and chances are that ideas they introduced to Google helped Google launch its personalized search.