Productivity Portfolio

There’s no question the brain is a remarkable tool. It can take the following paragraph and make sense out of it.

Aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it deosn't mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer be at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe.

It would be easy to dismiss spelling errors, but we shouldn’t. Although the brain may process the words, it subconsciously notes the mistakes. If a reader sees enough of these mistakes, they start to question your credibility. Avoiding typographical errors is one of the 10 items cited in the Stanford Guidelines for Web Credibility.

Checking for these errors is not as easy as you think. One problem is that not all spell checking tools work for the web. Certainly, spell checkers like those in Microsoft Word help, but many mistakes are introduced later. Most of my mistakes have come from editing HTML code for links or other modules such as our “Sites We Like” section.

There are spell checkers that work with HTML editors, but I found them lacking. They either flagged all words with an apostrophe or had a dictionary that was too small. Some don’t even allow you to enter your own words. The result was it took too much work to find too few mistakes. You reverted to the old eyeball method and hoped you got them all.

TextTrust’s Two Tiered System

What’s unique about TextTrust is they use a crawler and professional editors. Each week, I see in my server log where their user agent crawled my site. By default, the crawler works on the weekend since that’s when most sites have lower traffic. I didn’t see or hear of any performance issues from readers so I suspect its impact is minimal.

The emphasis is on what your reader sees when they see the page. The Meta information is not included. Any errors the crawler spots are reviewed by the editors. This two stage process minimizes the false positives you see with other services.

At first, I thought the service was having a problem with apostrophes. The service flagged the word “don’t” where I used the entity name. I sent an email to support explaining the issue thinking I found a bug in their system. The word looked fine in Firefox which was why I thought the error was with TextTrust.

Click to enlarge

Click to enlarge

If there is a bug, it probably resides with Internet Explorer which failed to display the apostrophe. TextTrust was correct in flagging the word. I opened the story in a HTML editor and substituted the entity number. It was an odd situation as IE correctly displayed other examples using that same entity name. All the same, readers using Internet Explorer would have seen this mistake.

Getting the Results

Each week, I get a summary email showing the number of mistakes. In the email is a link to an Excel spreadsheet that provides the details. The worksheet lists each error, context, URL and page title. It also includes hyperlinks to the story and Google’s define: feature. This format makes it easy to review and correct my mistakes. The only missing item I’d like to see is an easy way to add a word to my custom dictionary from this sheet.

Click to enlarge

You’ll notice in the spreadsheet above, I have a number of errors where the apostrophe is missing. My first reaction was “oh not again!” with some extra non-printable words. What perplexed me was these stories weren’t new, but ones written earlier. Surely, these mistakes would’ve shown on earlier reports. What changed to make these mistakes now appear?

I’m sure some of you know the answer. It was related to our changing content management systems that week. For most pages, the conversion went fine. But, for 34 words, the apostrophes were missing. I can only figure something happened with the SQL database export.

Different Plans for Different Sites

TextTrust offers two types of service. The first is a one-time check called TextTrust Express. This is good plan for people that put up a site, but don’t plan on updating the content. The costs vary based on site size and range from $29.95 for a 250 page site up to $449.95 for 20,000 pages.

The company also offers a recurring weekly scan called TextTrust Pro. This service also has a sliding scale cost starting as low as $49.95 to $749.95. This plan also allows you to request on-demand scans. You also have the ability to add words to a custom dictionary. For example, we were able to add “blimperskins” to our dictionary so it didn’t show on subsequent reports.

We’ve been using the service since May and have been pleased with the results. There’s no question in our mind it would’ve taken months to find and clean up the 34 mistakes from our CMS conversion. That’s in addition to the other mistakes we made on our own. When we did have a problem with the service, the company was quick to respond. One example was when we switched CMS systems; the crawler started hitting our “print friendly” pages.

I suspect that I’m like many webmasters in that I know spelling is an issue, but I don’t make it a priority. You think you can handle the task on your own or expect contributors to have checked their work. You make corrections as people point them out. But how many people take the time to inform you of mistakes? And as we found out, there are other ways in which spelling errors can creep in to your site. Regardless of the size and quantity of the mistakes, they all impact your reputation.

For anyone convinced that they don’t need this service, I’d suggest you try their test. The company has a free trial offer where it will check 10 pages and email you the results. This is the hook that got us started. Although that test was only on 10 pages, it found enough mistakes that we knew we should act.

Additional Information
Site: http://www.texttrust.com/
Cost: $29.95 to $749.95
Trial: www.texttrust.com/public/free-trial.aspx (Free 10 page trial)