Yearly Archives: 2011

101 Questions You’ll Need to Answer for an Accurate Quote from A Website Developer

It’s taken us since 2004 (that’s 82 Internet-years) to come up with a super-comprehensive list of everything we need to know to understand your business, get you an accurate quote and build a site you’ll love that markets your company according to your business principles and vision.

And we just can’t keep it to ourselves.

So, whether you end up getting on board with us or going elsewhere, here’s a handy checklist of questions your web developer should be asking–and that you’ll need to be able to answer–in order to build the website best suited to you and your business.

Hosting and Domain Name

  • Do you already have a URL (registered website address) you plan to use? If not, do you need help selecting and registering a good URL?
  • If you have a current website, who is your hosting provider? Are they meeting your needs with respect to storage space, bandwidth, databases, email and price? Do you need to move to a new hosting provider?
  • Do you have full access to FTP, databases, domain names? (If you don’t know what these things are, we can help.)
  • Can you provide usernames and passwords for FTP, databases, domain name registration and hosting?

Logistics and Planning

  • Are any other companies or contractors working with you on the development of the website?
  • Do you have a budget you are trying to meet?
  • Do you have a completed site architecture (an outline of all clickable items and menu-items/pages with a hierarchy, if applicable) or a wireframe (a rough sketch of the location of major elements on the home or inside pages) for the new website or will this be part of the scope of work?
  • When do you need your website up and running?

Marketing Goals

  • What is the purpose of the website: are you selling something or is it informational or brochure-like in nature?
  • What types of actions do you want your visitors to take on your website–buy a product, click on a button, fill out a form, call a phone number?
  • Describe your target audience in terms of age, income, profession or other demographics.
  • What business values or image do you want expressed on the site?
  • What makes you different from your competitors and why would people want to do business with you rather than a competitor?

Design, Look and Feel

  • If you already have a website, what’s good about it (a certain functionality, a look, the content etc.)? What frustrates you about your current website?
  • What websites have you seen that you like (does not have to be industry specific)? Does another website (perhaps a competitor) have a feature or look that you’d like to have, or that you really hate? Can you describe the style of website you want?
  • Are there specific company colors that need to be implemented such as logo colors, brochure or press release materials? Can you provide the Pantone numbers for them?
  • What are the three things that are most important in the design of your new website?

Website Functionality

  • Will front-end users need to log in to your site for any reason (e.g. to see proprietary content or to gain access to their account)? What kind of content will be in this protected area?
  • How many web forms does your new site need and what is the purpose of each? How will the data from the form be handled (e.g. sent to your email address and/or stored in a database etc.)?
  • Do you need any social media features built in (such as clickable Facebook, twitter etc. icons or feeds)?
  • Are there third-party applications that will need to be integrated and if so which ones?
  • Do you need any subscription services? Or do you use a third party for any part of the subscription content delivery or payment?
  • Do you wish to use any “content-on-demand” features (hidden elements that are made visible with certain actions)?
  • Do you require an online chat feature?
  • Do you need an updatable events calendar?
  • Do you need an internal site search feature?
  • Does your site need a blog or a forum?
  • Will you be offering advertising on the site? How should it be implemented (e.g. in a banner, in the sidebar)?
  • Do you have any other specifications or need specific functionality that has not been addressed?

Image and Text Content

  • Do you have a logo or will one need to be created? If you have a logo, can you provide the original artwork files (usually in .eps, .ai or .pdf format)?
  • Do you have any specific photos you plan to use—do you own the rights to these photos and can you provide them in high resolution files? Or will we need to find/create images for the website?
  • If video or audio is to be part of the new website can you provide us with the proper files? About how many video or audio files are required to be implemented?
  • Do you have any other media, such as PDF documents, that need to be incorporated and will you be creating these or will we?
  • Will you need a favicon (small icon that appears in some browsers address bars) created?
  • About how many pages will the finished website be?
  • Do you have a tagline you wish to use or do you need help creating one for your site?
  • What information must be on the home page?
  • What information (e.g. features, sections or contact info) must always be visible or emphasized throughout the site? Do you want contact information prominently displayed?
  • Do you have the content for the website or will content creation be a part of the scope of work?
  • Will we implement your text and image content (as opposed to overall layout/design elements) into your website or will you?

Accessibility

  • All of our sites are mobile friendly (incorporate responsive design) but do you have any specific mobile requirements?
  • Will your visitors require any special needs (i.e., screen reader ready, larger fonts)?
  • Do you need multi-language support?

e-Commerce

  • Will you need a shopping cart system for e-commerce content (comes with content/inventory/shipping/pricing and content management system)?
  • Do you have an e-commerce system you already use? If so, which one?
  • About how many products are you selling? Do they need to be broken down into categories (eg. dresses, shoes, sweaters)? Do they need to be broken down into sub-categories (such as colors, sizes, styles etc.)?
  • Do you want to suggest related products to your customers
  • Will you be selling/shipping your product internationally or just within Canada?
  • How will your customers pay? With PayPal (etc.) or by using credit cards with a merchant account? Do you need a payment gateway set-up (a secure programmatic link between what your customers want to buy and the credit-card approval procedure of your bank)?

Website Updating and Administration

  • Do you need a content management system (CMS—like WordPress, Joomla or a proprietary system) to manage (edit/change/delete) the content of your website yourself or would you prefer that your developer do that? (If you are selling online with an e-commerce package, a CMS is already included).
  • If you are implementing content yourself through a CMS, would do you need training for making website updates?
  • Will you need multiple levels of access to editable content (e.g. some people can write content but cannot publish it, etc.) and the ability to manage content publishing approval processes?

Search Engine Optimization

  • Do you have a Google Analytics account? If so, can you provide access?
  • What are the basic keywords you want to rank for?
  • What are a few of your main competitors?
  • If you want to do pay-per-click advertising, what is your daily budget for your given search terms?

10 Ways to Tell That Your Web Designer is Awful

sorry your website isnt right, dude

We’ve been around since pretty near the Dawn of the Internet and we’ve seen just about everything that can go horribly wrong in the world of web design. We’ve made our fair share of mistakes too, but with the benefit of hindsight we’ve identified these 10 tell-tale signs that often indicate a web designer is out to lunch.

1. Your Web Designer Promises you a website for less than a $1,000 bucks

Your Web Designer Promises you a website for less than a $1,000 bucksA week-long vacation in the Caribbean island of your choice, a well running used car and a website that you can be proud of are all going to cost you more than a $1,000 bucks. Your website is your 24-hour, 365 day-a-year, tireless salesperson and the face of your company. Go cheap and chances are that your website will be as awful as the designer that created it. In our experience, going “cheap” comes back to bite you in a most unpleasant place because now you’ve got to fork out for a professional web designer plus the cost of your super-cheap designer.

A well-designed, 5-10 page basic site that you can call your own built by a professional designer with experience usually starts at about $1800. About the price of short stay somewhere in the Gulf of Mexico.

2. Your web developer is one of the family

Your web developer is one of the family“My neice is pretty tech savvy and she’s willing to do my website in her spare time for a big discount.” Cringe-worthy. Hiring someone that’s related to you is always a bad idea. First, you can’t fire them. Second, disagreements over how the site should look and when–if ever–it will actually be finished often lead to an awkward family impasse that puts a kibosh on your relationship. Not worth the money you were trying to save, that’s for sure.

3. YOU are your web designer

If you’re not a designer and you built you website yourself, then it’s probably terrible. Sure, there are plenty of cheap web design tools out there and you can learn to use in a few hours. Or days. Or weeks. But it takes a lot of experience to design a site that not only looks like it wasn’t stamped out by a cookie cutter, but also makes users happy, garners conversions and increases your traffic and sales. Plus, don’t you have something better you could be doing, like running the company?

4. Your web designer lives in a country you don’t do business in

It’s not that web designers in other countries can’t do good work. It’s just that lines of communication become culturally, linguistically and bureaucratically long– with lots of middle-men trying to communicate your companies vision, image and goals. That’s often a pretty big disconnect. Also, foreign web designers are typically impossible to vet, unless you don’t mind flying to South Korea and actually checking them out. We’ve heard too many horrible stories about what happens when either you or another party becomes “disgruntled”: you often have very little recourse and you end wasting money and precious time.

You don’t have to meet your web designer face to face, but our advice is to always hire one from within an area that you could reasonably drop in on if need be.

Your web designer lives in a country you don't do business in

5. Your web designer uses super-cheesey stock images

head set hottie wedge of power Your web designer uses super-cheesey stock images

The Head-set Hottie”, “The Power Wedge” and “The Disconnected Handshake” are all sure signs that your web developer thought about how your website should look for about twelve seconds. Using unimaginative stock photos not only cheapens your site, but it makes it look like so many other cardboard cut-out, non-performing, cheese-ridden sites that are still out there either scamming people or doing nothing at all. Images don’t have to be as radical creative as some of ours, but they should enliven your visitors to your product–not put them to sleep with pictures of people that don’t even work at your company.

6. Your web designer doesn’t call, doesn’t write

Your web designer doesn't call, doesn't writeIf you give your web designer a down payment and they scurry off to work but you don’t hear from them until you call wondering what’s going on, it’s probably time to hoist-up a big red flag. Even if you did hammer out what you thought was every detail at the outset of your project, we find that there’s always a thousand questions, comments, improvements and suggestions that come up during the design and implementation of any website. So not hearing from your web designer is often because they have way too many clients and haven’t even started your project yet or simply don’t care enough to ask the important questions. Both are bad: a good web designer always stays in contact and keeps the momentum going.

7. Your web designer is all artist, no businessman

Your website should look beautiful. But if users can’t find what they need quickly, they’ll flee now matter how pretty your site is. A website that’s too “arty” often loses focus on “conversions”–that is, what you’d like the user to do–buy a product, fill out an email sign-up or contact form, click a button or make a phone call. A good web developer will help you clearly define your conversions and make your site work for you. A bad web designer will often over-indulge himself with what he thinks is pretty, without giving your users a clear and persistent idea of what they’re supposed to do on the site. And making your site function as a successful business is the whole reason you started a website in the first place, right?

Your web designer is all artist, no businessman

Are we going boating or getting our hair cut? How do I make an appointment?

8. Your web designer makes you run back to her for every little change

Your web designer makes you run back to her for every little changeIf your web designer makes you come back to them for every little change after you website is live, you’ve got an awful web designer. These days, we have content management systems that let you modify most of your content without running back to your designer and (gulp!) paying them more money. And, what happens if your web designer “goes away?”

Most content management systems (with names like WordPress, Joomla or Drupal) are open source (public and pretty much free, built by thousands of programmers throughout the world) and your designer can build your website right inside these platforms from the very beginning. Then, once your site is live, you can make content changes to it yourself. Opinion alert: Beware of some web companies that want to sell your their own propitiatory content management systems. If you go that route it almost always will cost you more money and you’ll always be beholden to them to support that system. Unless your site has some incredibly unique content demands, why reinvent the wheel?

9. Your web designer can’t, won’t or didn’t make your site responsive

Your web designer can't, won't or didn't make your site responsiveWe’ve heard this one a few times: “I got this web designer to build my site and it looked great. But then one of my clients said he couldn’t view it properly on his phone. I went back to the designer and they said it would be even more money to make it mobile friendly.” Disreputable web designers use this “don’t-ask-dont’-tell” policy to put the squeeze on you. All web sites designed today should be mobile friendly–why alienate half your audience? Anyone who says “that’ll cost you extra” is a terrible web designer.

10. Your web designer doesn’t help you promote your website

Optimizing your website so that people can find you begins with good design, usability and correct programming and if your designer doesn’t know how to do this this you’ll be off to a very bad start. If the search engines can’t index your site when people type in the relevant search terms or if people can’t find what they’re looking for quickly and easily when they do get to your site then any strategic search engine optimization strategies, social media marketing and even paid advertising are going to fall flat. Any good web designer will be able to tell you all about the complex world of web site promotion, all you need do is ask!

Making Your Web Site Mobile Friendly: Responsive Website Design and More

According to Pew Internet, 63% of cell phone owners browse the web on their phone and of the people who use a mobile device (which includes tablets) to go online, 34% of them usually do so using their phone. These numbers have been on the upswing for the last few years and show no sign of slowing. Yet, says an Adobe study, only 45% of “marketers” have a mobile friendly site. We’re not quite sure who “marketers” are, but—in our estimation—the vast majority of websites out there are not mobile friendly, much less mobile responsive. If a potential customer goes to your website and finds a page that’s difficult to read or navigate and doesn’t contain what she’s looking for, how long do you think she’ll stick around before going to your competitor’s mobile friendly site?

If you’re lucky, the website you had built a few years ago for desktops will at least display properly on a mobile device. Which is to say it “mini-fies”—just gets smaller to fit the smaller screen real estate without missing or overlapping chunks. Being mobility responsive, however, means that the content itself–or at least the way the content displays– actually changes when viewed on a smaller screen. Being responsive is only one way a website becomes “mobile friendly”, a term of larger scope that includes a responsive site as just one of its components. We’ll get to that in a moment.

A responsive website includes styling code that allows it to detect when a device with a smaller screen size is viewing your site and to adjust the content and layout accordingly. You (and your web programmer) will have to tell it what “accordingly” means. But let’s say your “big” site has a horizontal menu that spreads all the way across the page. You can probably guess what happens when the menu display on a small screen: it may “break” in the middle or the menu items will be so small and hard to read and click on that people won’t even bother. With a responsive site, that broad horizontal menu might be told to display vertically and be toggled on and off with an icon (for maximum view-ability of the other aspects of the site). Similar adjustments might have to be made for galleries, news items, forms, font sizes and other things that have lots of horizontal real estate on your “big site”. Unless you have a lot of applications in your site, you probably don’t need a completely separate site for mobile devices, but do get yourself a responsive site that can handle both desk tops and mobile devices.

Another reason to avail yourself of a responsive site goes beyond mere style and readability and centers on the fact that mobile users are often looking for a different type of content on their cell phones than they might do on their desktops. Not only are mobile users even less prone to reading than their desktop cousins, they are more likely to be out and about, needing your location quickly or wanting a quick comparison between some products. This means that you’ll want things like maps and directions, succinct answers to the most often asked questions, and—especially for e-commerce—prices, pictures and descriptions of products that are readable and at the forefront of site. So, on your “big” widget-selling website you might have tons of testimonials about how good your widgets are. You may want to fold these up and put them someone else when the site is opened in a mobile device. The less scrolling to get to the most important information, the better. And for gosh sakes, if you’re main site uses a login to view content or pop-ups, can them for mobile users. Finally, the one thing that mobile users really hate is a slow-loading site—so concentrate on the essentials and get rid of that big transitioning JavaScript photo gallery of your store aquarium. Video content is a great kind of content to have on your “big” site, but getting it to work for both desktops and mobile devices is hard. Whatever you do, don’t use Flash—it’ll just show up a s big gray box on all Apple device—and even embedded YouTube videos can take forever to load on mobiles. Ask us about other things you can do to speed the mobile page up.

Overall, for a mobile friendly site, you’ll need to tailor the site’s user experience to what context it’s being used in. Are your mobile browsers actually in your store comparing prices with other stores (offer price comparison) or are they trying to find a specific item (offer a way to guide them around the store)?

The time is passed where you can concentrate solely on a desktop site and ignore the different needs of people with mobile smaller screens. But you can still have one site and do both.

Google and the Semantic Web: A Revolution Toward a Star Trek WorldPart IV: Going Where No Search Engine Has Gone Before: the Internet of Things

This article is part of a series:
Part I: The Hummingbird Platform: Google Gets “Context”
Part II: Google as an Answer Engine That Can Know All There is to Know
Part III: Web Pages That Mirror the Structure of Human Knowledge
Part IV: Going Where No Search Engine Has Gone Before: the Internet of Things

In the last article we saw the beginning of a Web that is full of structured data, which allows for a far more efficient, comprehensive and relevant “search” engine that mimics the actual structure of all possible human knowledge. But, up to now, computers–and so the Internet– are wholly reliant on human beings for the input of data, whether via typing on a keyboard, uploading digital media or even by scanning a bar code on a product. But humans are notoriously inaccurate and have both limited time and attention.

Right now, things on the Internet are represented by our ideas of them–that is, they are represented in sentences in our native language. But imagine if their was a website whose content–by the use of the micro-data with a similar structure to that of Schema.org–could represent the qualities or states of a machine instead of an idea or concept. As more machines become connected to the Internet and as more bar codes, Radio Frequency signals or QR codes become embedded in objects, your computer could receive a message from you dishwasher’s “website”–where all of its qualities and states are represented as structured data (like our example of a movie site in Part III). By monitoring the dishwasher’s current state and comparing that to its ideal or brand new state (via the qualities hierarchy of Schema.org’s hierarchy), your computer could tell you when the dishwasher’s motor is about to fail and order a new one that is delivered before your stuck washing dishes by hand!

Dishwashers, of course, are a drop in the bucket. If all things were outfitted with identifiers, like bar codes or RF signals, they could all be managed an inventoried by computers. By tracking and counting everything, businesses would no longer run out of stock or create waste products. The precise state of everything from airplanes to sewing needles could be immediately known and reacted to. We would know when things needed replacing, repairing or recalling, and whether they were fresh or past their best. If we had computers that knew everything there was to know about things—using data they gathered without any help from us—we would be able to track and count everything, and greatly reduce waste, loss and cost.

We know this stuff sounds pretty futuristic and “Star Trek”. But, the contextual or “semantic” understanding within the Google Hummingbird platform, to the use of structured data within web pages themselves brings these ideas that much closer to reality. In closing, consider these facts:

  • in 2008, the number of things connected to the Internet exceeded the number of people on earth. By 2020, 50 billion devices will be connected to the Internet.
  • and those devices aren’t just cell phones or tablets, they’re….everything. For example…A Dutch startup is using wireless sensors on cattle, so that if one is sick or pregnant, it sends a message to the farmer. Each cow transmits 200 mega bytes of data each year. Human health is beginning to be monitored in the same way, especially with pace makers.
  • at the end of 2011, 20 typical households generate more Internet traffic than the entire Internet in 2008.
  • we already have cameras that are one cubic millimeter: you could fit 150 of them in one square inch.
  • the new IPv6 Internet addressing protocol has 340, 282, 366, 920,938,463, 463,374,607,431,768,211,456 possible Internet addresses. That’s 100 addresses for every atom on Earth.

Google and the Semantic Web: A Revolution Toward a Star Trek World Part III: Web Pages That Mirror the Structure of Human Knowledge

This article is part of a series:
Part I: The Hummingbird Platform: Google Gets “Context”
Part II: Google as an Answer Engine That Can Know All There is to Know
Part III: Web Pages That Mirror the Structure of Human Knowledge
Part IV: Going Where No Search Engine Has Gone Before: the Internet of Things

In the last section we realized that, currently, web pages contain no structured data. In this article, we’re going to think a little more philosophically and ask “what is knowledge and what is its structure?”

While leaving lots of room for argument, the basic structure of our knowledge–that is, how we know something–can be found in the relation of subject-to-predicate, substance-to-quality, or, more generally a thing and its relation to a description. For example, a raven is an animal, a raven is a bird, a raven is a black thing. If we knew every quality that could be ascribed to every subject in the world and how all those things are related, we could arguably claim to “know everything”.

Enter Schema.org, a relatively new Web standards authority that has as its lofty goal to organize all possible human knowledge into a complex and inter-related structure (called an “ontology”–the study of everything that “is”). Webpages can use this structure to mark-up their text-content with tags that emulate the overall structure of knowledge for any particular topic. Since text content will now be “structured” or “named” by a subject-description relationship, Google can parse the data to better suit the search queries of its users.

The highest level in Schema.org’s (check it out!) hierarchy is equivalent to asking what qualities a “thing” must have to be a thing. All things must have a name, for example, or they must be either tangible or intangible, to guess at just a few of a thing’s qualities. On the next level down in the hierarchy you could ask what kind of qualities an intangible thing might have: and just one answer might be that an intangible thing could have the quality of being a Creative Work, such as a novel, play or movie. Continuing down the hierarchy, we could ask what qualities a movie has. The answer might include, its name, its director, its actors, its running time and the number of reviews it has. Now, how can we represent this structure in a webpage about movies? By using “micro-data”.

Here’s what an “old,” unstructured HTML page about “Pirates of the Caribbean” might look like (We don’t usually include code, but this time this extremely simple code is necessary to gain an understanding of what’s happening.  The <h1></h1> simply delineates the start and stop of a heading and the <p></p> the start and stop of paragraphs. This is simple styling tags only.)

<h1>Pirates of the Caribbean: On Stranger Tides (2011)</h1>

<p>Jack Sparrow and Barbossa embark on a quest to find the elusive fountain

of youth, only to discover that Blackbeard and his daughter are after it too.</p>

<p>Director: Rob Marshall</p>

<p>Writers: Ted Elliott, Terry Rossio, and 7 more credits</p>

<p>Stars: Johnny Depp, Penelope Cruz, Ian McShane</p>

<p>8/10 stars from 200 users.</p>

<p>Reviews: 50.</p>

Here’s the new way, using the Schema.org micro-data markup:

<p itemscope itemtype=”http://schema.org/Movie”> translation: this is a thing (“itemscope”), and the type of thing (“itemtype”) is a Movie, with a link to the schema for movie. You can see that schema by copying that link into your browser, if you want.

<h1 itemprop=”name”>Pirates of the Caribbean: On Stranger Tides (2011)</h1>translation: this is the name of the movie, which is only one of the item’s properties

<p itemprop=”description”>Jack Sparrow and Barbossa embark on a quest to
find the elusive fountain of youth, only to discover that Blackbeard and his daughter are after it too.</p> translation: this is the description of the movie, which is another of the item’s properties.

<p itemprop=”director”itemtype=”http://schema.org/Person”>Director:
<span itemprop=”name”>Rob Marshall</span>

</p>translation: this is the director of the movie, which is another of the item’s properties. Note the link to the schema for persons. And so on and so on….

<p itemprop=”author” itemscope itemtype=”http://schema.org/Person”>Writers:

<span itemprop=”name”>Ted Elliott</span>

</p>

<p itemprop=”author” itemscope itemtype=”http://schema.org/Person”>

<span itemprop=”name”>Terry Rossio</span>

</p>

<p itemprop=”actor” itemscope itemtype=”http://schema.org/Person”>Stars

<span itemprop=”name”>Johnny Depp</span>,

</p>

<p itemprop=”actor” itemscope itemtype=”http://schema.org/Person”>

<span itemprop=”name”>Penelope Cruz</span>,

</p>

<p itemprop=”actor” itemscope itemtype=”http://schema.org/Person”>

<span itemprop=”name”>Ian McShane</span>

</p>

<p itemprop=”aggregateRating” itemscope itemtype=”http://schema.org/AggregateRating”>

<span itemprop=”ratingValue”>8</span>/<span itemprop=”bestRating”>10</span> stars from

<span itemprop=”ratingCount”>200</span> users.

Reviews: <span itemprop=”reviewCount”>50</span>.

</p>

</p>

Now, if you searched on Google for “who is the director of the Pirates of the Caribbean” how much more likely would it be that Google could return exactly the answer you’re looking for, as opposed to thousand of webpages about the Caribbean or about pirates and so on. Also, consider, also the example below in terms of e-commerce (Note: tags have been simplified from actual Schema.org protocol for simplicity):

<p itemprop=”name”>Mousetrap</p>

<p itemprop=”price”>$19.95</p>

<p itemprop=”availability”>Available today!</p>

<p itemprop=”brand”>My Own</p>

<p itemprop=”condition”>New</p>

Now, when you search Google for “the lowest price on a mousetrap,” Google could look for every site that contained “Mousetrap” as the name of a thing (product) and which had a price as one of that thing’s properties. Then, it could compare all the prices it found for the mousetrap and show you the website that had the lowest price for the mousetrap. THAT will revolutionize e-commerce.

It doesn’t take much imagination to realize the impact this kind of HTML-based structured data could have on searches and information relating to medicine, law, all branches of science, literature and indeed, every topic, concept or thing that can be known in the world. If you can see that, you’ll begin to understand how much Google and your computer are becoming more and more like that know-it-all Star Trek computer. Why on Earth didn’t we think of doing this before?

Note: Any HTML 5 site will accept all any any of Schema.org’s markup tags. Google says it isn’t indexing sites on the basis of the existence of those tags yet, but many Search Engine Optimization experts think that it can still improve your rankings somewhat. We’re betting that Google will begin to index based on these tags very soon in a gradual roll-out.

Up Next: Part IV: Going Where No Search Engine Has Gone Before: the Internet of Things

Google and the Semantic Web: A Revolution Toward a Star Trek World Part II: Google as an Answer Engine That Can Know All There is to Know

This article is part of a series:
Part I: The Hummingbird Platform: Google Gets “Context”
Part II: Google as an Answer Engine That Can Know All There is to Know
Part III: Web Pages That Mirror the Structure of Human Knowledge
Part IV: Going Where No Search Engine Has Gone Before: the Internet of Things

Google made another quiet change a few months ago that may not seem such a big deal at first, until you think about what’s going on under the hood. If you search for a person, place or thing, as in “who is Barack Obama” you will now be presented with a “Knowledge Panel” at the top right of the search results. This panel is not a listing of web sites where you might be able to find out about Mr. Obama. Instead, it contains actual information about the President, like his date and place of birth, his relatives and so on. Before you dismiss the importance of the Knowledge Panel as simply a handy little info box, you should consider that it represents a complete change with respect to how Google does “search”. In the past, Google only delivered links to websites to satisfy a searcher’s quest for knowledge, but now it has begun delivering the information itself. It is becoming–not just a search engine– but an Answer Engine. When Captain Kirk queried his on board computer about a person, he didn’t get a list of websites to peruse: he got (usually way too much) information.

Not only has Google  quietly amassed an encyclopedia of 570 million unique concepts (and 3.5 billion facts), but–by using basic logic and structured data– it can also determine  the relations among those concepts and facts. Just SOME of those concepts include Actors, Directors, Movies, Art Works and Museums, Cities and Countries, Islands, Lakes, Lighthouses, Music Albums and Groups, Planets and Spacecraft, Roller Coasters, Skyscrapers and Sports Teams.

Right now, Google is accomplishing this feat by gaining access to the databases of sites like Wikipedia and Freebase. A database, unlike most websites, contain what is called “structured data,” a concept that is crucial for understanding the future of the Web. “Abraham Lincoln Barack Obama Jimmy Carter might seem like structured data to us humans reading those words, but to a computer they are just strings of letters….until those strings of letters are contained under the heading of American Presidents. Google can go to, say, the Wikipedia database, look under the Table called “People”, then into a sub table called “American Presidents” and within another sub-table find “Barack Obama” and into yet another sub-table to find Barack’s date of birth, for example. Because the data is structured–i.e. one named piece of data is contained within another named piece of data, the data can be displayed to you with a meaningful structure. Anyone who has worked on the simplest of databases might know this.

There are two important limitations to consider at this point. The first is that although Google might be able to negotiate gaining access to (and be able to understand the structure of) some very large and established databases like those of Wikipedia, it is unlikely that it could do so for the database that might support your website, or anyone else’s for that matter. Secondly, the HTML (Internet code) that your website is currently built with contains no structured data in the sense we’ve been talking about. HTML is simply a “mark-up” language, meaning that some group of letters can be marked off as a “paragraph,” let’s say, so that your browser knows to put extra space before and after it. There are many other things that HTML can “markup”, such as headings, forms and so on. Google can see this stylistic structure, but other than that, your site is just one long string of letters. For example, if you had these two paragraphs in your site:

<p>The Best Mousetrap Ever</p>

<p>Price: $6.00</p>

Google has no way of knowing that those two paragraphs are related by the first paragraph representing the name of a product and the second paragraph representing the price of that product. Google also does not even know what a mousetrap might be or even if a mousetrap is a separate thing from the word “Ever”. Because of this fact, when you search in Google for “the lowest price on the best mousetrap ever” it tries to find the best matching words from websites all over the world, which might include pages about the play called “The Mousetrap”or pages on how to build a mousetrap etc. It can have no idea what a lowest price means or where it might be located on the page. This is all due to the fact that HTML pages contain no structured data.

But all that is about to change.

Up next: Part III: Web Pages That Mirror the Structure of Human Knowledge

Google and the Semantic Web: A Revolution Toward a Star Trek World Part I: The Google Hummingbird Platform: Google Gets “Context”

This article is part of a series:
Part I: The Google Hummingbird Platform: Google Gets “Context”
Part II: Google as an Answer Engine That Can Know All There is to Know
Part III: Web Pages That Mirror the Structure of Human Knowledge
Part IV: Going Where No Search Engine Has Gone Before: the Internet of Things


Series Introduction

Most revolutions start with small things and grow incrementally before they’re even recognized as being part of a revolution. We believe that there are a lot of things going on in “search” these days that will soon add up to a revolution which will bring us closer to what was once regarded as the “far-fetched” world of Star Trek. Because this blog is designed for the non-technocratic, we’re going to approach our topic like a PBS special and do things in parts that all fit together in an ever-increasing level of sophistication that will culminate in a mind-blowing understanding of the future of the Internet, search and even man’s relationship to the computer. We’ll start in this first section by exploring Google’s new Hummingbird platform, then move on to “search” engines becoming Answer Machines, continue with the new wave of structured data coming to websites soon and then end up with the far-out by real concept of an “Internet of Things”


Last month, Google quietly rolled out “Hummingbird”, which is a completely new search algorithm said to effect 90% of all searches. Although there have been many updates to the Google algorithm in the last few years, most of these simply made Google better at gathering and filtering website content. Hummingbird, however aims to better understand the the searcher.

Before Hummingbird, Google would look at each word within a search query and return a collection of websites that had those words in their text. Hence a search query like “barber shops near my house” might get you a Wikipedia article on barber shops, some map results based on your address and some home-renovation sites that have the page title “My House”. With Hummingbird, Google tries to understand the meaning of the user’s search query as a whole by attempting to understand its context and intent by analyzing the meaning of language and the way people actually communicate. Users who are signed in to their Google accounts will of course benefit the most from these improvements, as Google ties in all the information it has about you–your location, previous searches, time of day and social connections on Google Plus. Because of this, searching for “barber shops near my house” is far more likely to actually bring up barber shops that are near your house.

Part of the prompting for such changes is our increasingly “Star Trekian” ability to search the web using voice queries, with tend to be more complex, longer and more conversational in structure. Users of voice search–which will include Google Glass users–can now ask Google “when was the Empire State Building constructed?”, receive their answer, and then ask “how tall is it?” and receive the height of the building. And “show me a picture of the Empire State building” does just that. Very impressive.

Google is preparing for a future when users interact with their search engine often, quickly, verbally and in a way that puts much less reliance of keywords and more on a site’s semantic relevance, authority, and user experience. To do that, they must better understand real people and the way they talk. That way, the very non-Star Trek keyboard can go by the wayside.

For site owners, your focus should now be on understanding your target users, producing awesome unique content, establishing your visibility and authority and giving your users a great experience on your site.

Up next: Part II: Google as an Answer Engine That Can Know All There is to Know

The New Way to Do SEO

In just the last two years, the way SEO is done has almost completely changed. Not so long ago, optimizing for the search engines meant figuring out ways to get as many other websites to link to your site as possible, stuffing the page with your keywords and nabbing content from almost any source. After two huge updates to the mechanics of how Google ranks your site, called Penguin and Panda respectively, linking schemes, keyword stuffing and simply copying someone else’s awesome text are all definite no-nos.

Today, Google looks for websites that have fresh, constantly updated, quality content that really informs the visitor and is written in “natural language”. This means no “duplicating” someone else’s content and no keyword “stuffing”. Google has the ability to check if even one sentence has been copied from another website. It also now regards overly repeated words as a lame attempt to trick it into thinking your website is relevant for a given search term. Links must be “natural” too, in the sense that the link text should contain your keywords and should link from logically related web sites. Bye-bye link farms. Finally, since Google regards new and unique text as an indicator of a vital and popular website, today’s SEO must include a method for writing and disseminating new content each and every week.

Social networking has also become a rapidly rising requirement for optimizing your site, because it gets your content in front of a lot of people very quickly. Google can now count how many mentions you have on the growing number of social networking sites and will rank your site for a given search term accordingly.

Google now also looks at the ease of access your website has–is the code valid, is the navigation clear, is the page easy to read, containing bullet points and headings, and is it truly relevant to its keywords.

Overall, today’s SEO has become less of a technical exercise and more a part of an overall marketing plan that includes constantly updating content, maintaining a user friendly site, interacting with social media and linking to relevant sites.

Google Partners Pilot Program Launched in Canada to Aid Web Professionals

Google Partners Pilot Program

Google  announced a new pilot program that has  launched in Canada called Google Partners which will aid web professionals and agencies succeed on-line. Businesses  can connect with a certified web professional or agencies that have been verified by Google as using the best web practices and who “have met high quality standards in customer care and on-line expertise.”   Google has amalgamated its existing programs under the Partners platform so that web professionals can access Google resources, training, and support all in one place. Google has also created a new Partners community to “foster knowledge sharing within the industry” and “provide more opportunities for dialogue with Google specialists.”

If you become a partner, you’ll have to earn the Google Partner badge (below) by signing up with Google Partners and taking some exams.  Certification means that you become a trusted Google Partner. You can also “hangout with Google” by joining an exclusive Google + community where you’l get support from Google specialists and talk with leading industry experts. You will also have access a library of training manuals, case studies and pitch materials to learn the latest and greatest about Google.  Finally, and most interestingly, after you earn the Google Partner’s badge, Google promises to “introduce you to potential clients through Google Partner Search”.

Google estimates that the Partner program will target 6,000 web companies and professionals across Canada, and, if successful it will role out in other countries.

One Reason Your Google Page Rank Goes Down When it Should Go Up

Google Page Rank Goes Down When it Should Go Up

If you’ve recently optimized a web page, adding your targeted keywords in all the right places, you’d expect that your Google rank for that page should go up.  But, in some circumstances, the page might be down-ranked, at least for a short while.  The reason is that Google is trying to add another layer to its sophisticated attempt to catch people who “spam” or “over-optimize” their web pages.

Google seems to figure it this way:  if a web master puts new keywords up on a page and that page’s rank actually drops but those new keywords are not removed in reaction to the drop, Google considers (after a few weeks) you to be legitimately optimizing your site and that the changes you made are real.  After a few weeks you’ll get your old (or better) rankings back.  On the other hand, if a web master puts new keywords on a page and that page’s rank declines and the web master than removes or revises that content, Google considers him as trying to manipulate the search results.  Hence, his lower page rank becomes permanent.  In other words: if a web page reacts too quickly to changes in the search results, it’s more likely that the page is using spam tactics. This line of attack, however, is usually only instigated if Google already suspects the site or page to be “spamming”.  Spamming here means using the “Black Hat” techniques of keyword or META tag stuffing, invisible or tiny text, link manipulation or page redirects.
Remember that Google is not against search engine optimization if you create real, helpful and relevant content.  But Google does not like any attempt to “trick” its search results.  If you optimize your web pages for new keywords that are relevant to your website, don’t worry if the rankings of your pages temporarily drop. This is “normal” Google behaviour and your pages will receive higher rankings after a while (often between one and four weeks). One Reason Your Google Page Rank Goes Down When it Should Go Up