The Twitter Business Experiment

POV
By Sherry Holdridge on 10.14.2008

Download The Twitter Business Experiment

Does Twitter, a wildly popular social network that allows members to post short updates — up to 140 characters — offer any benefit to businesses? We set out to find the answer by posting updates — popularly known as "tweets" — to our Twitter account, twitter.com/tocquigny, during a three-week trial, and tracking the results. Here are our findings.

First, we needed to determine what we’d post about. Across all of Twitter, most tweets can be divided into two categories: "current activities" — such as "I’m eating lunch" — and "sharing information," which often include a link to an article. For this experiment, we took both approaches. We wanted Tocquigny to be both approachable (with our current activity tweets) and professional and knowledgeable (with our informational tweets).

We also found that major brands were using Twitter a wide variety of ways including: customer service, twitter.com/tysonfoods, breaking news, twitter.com/nytimes, research, twitter.com/forrester, and emergency support for Ike supplies, twitter.com/thehomedepot.

Next, we faced logistical obstacles. How often do we need to post? Who should post within the company? How do we get people to read our tweets and become loyal followers? Do we need to read all the tweets from the people we are following?

Time also played a role. Writing a 140-character tweet may not seem daunting, but finding something of value to add to the Twitter community required additional thought — and time. One solution was to follow emails that were floating throughout the company, extracting comments on new gadgets or tools we were experimenting with. Again, time was a major factor: During the experiment several new client projects cropped up that demanded people’s attention, and those emails dried up.

To try and increase my posting frequency, I added the Twitter40 widget, which pops up a "What are you doing?" reminder window on your computer desktop every 40 minutes. Though the widget does not require you to post to Twitter, after closing the popup four times in a row, I did feel the pressure to find something to share. Since the company was receiving accolades during the experiment, I posted a few announcements along those lines. But I did wonder if those tweets seemed too much like PR grabs rather than an effort to provide solid value to our followers.

Speaking of followers, how did we get them? It was easier than I imagined. After happening upon a posting that listed the Twitter addresses of hundreds of search engine optimization professionals, I followed them — and soon after most of them started following us back. The second way we gained followers was by replying to tweets. Our reply was not only visible on our main Twitter page, it was also visible to all the followers of the original poster. If their followers liked our reply, they would start to follow us too. The downside to responding to tweets and actively looking for new individuals was, again, finding the time to do it.

Now, onto the results. To gather our data we used BudURL to track who clicked on our URLs within the tweets and Google Analytics to watch our website traffic. Here is a selection of our tweets and the number clicks they generated:

  • Tocquigny was just selected as a finalist for Global Six Sigma & Business Improvement Award. http://budurl.com/gjtw
    13 clicks
  • Just found a great social community site which overlays a forum over articles posted on your site. http://budurl.com/b2ac
    22 clicks
  • We are celebrating Talk Like A Pirate Day. Everyone has tattoos. The tech industry loves the day too - Arrrg - http://budurl.com/nca6
    11 clicks
  • Search engine that allows you to search all Craig's List locations at once http://budurl.com/f2e7
    9 clicks
  • Want to improve your strategy...ask better questions http://budurl.com/yeys
    17 clicks
  • Technorati stopped by the office. Impressive new launches scheduled plus cool RSS blog banners. State of Blogosphere http://budurl.com/gjvc
    11 clicks
  • Great idea from a job seeker wanting to break into the ad world-- Make some buzz http://budurl.com/6csn
    16 clicks
  • Jason spoke earlier today at Webmaster Jam Session in Atlanta http://budurl.com/6fg3
    15 clicks
  • We tried out the new SEOmoz tool Linkscape. Its main value is giving a ranking to inbound links. http://budurl.com/6b7p
    8 clicks

At the end of the three-week experiment, we received 106 clicks from nine postings. It took us 2.25 hours of our time to manage the tweets (an average of one minute, 27 seconds per click). Based on one tweet about a job opening at Tocquigny, One Day, One Job wrote an article about us that generated the greatest amount of referring traffic to our website all month. As for whether Twitter offers any benefits to businesses, though we cannot directly connect a new client lead to our three-week experiment, we did gain visibility and additional traffic to our website. In addition, we learned of how we can offer those clients interested in using Twitter a measurable way to test results. Therefore, if a simple test reveals whether Twitter is a great place to convert customers using social media, then yes, it does have a business purpose.

Follow us on Twitter at twitter.com/tocquigny and we will follow you back.

Portfolio banner

Reader's Comments

  • Excellent article! I found great value in reading this article about value smile My company is currently in the same conundrum; to tweet or not to tweet? So, helpful stuff here, I’ll be referencing quite a bit I’m sure.

    By Aaron Hughling on 10.15.2008

  • Thanks for the comment Aaron. We also just found a great article that discusses how the CEO from Zappos shoes started using Twitter because it was fun instead of having a ROI goal. He created a page which listed all the Zappos employees which had Twitter accounts and the page no has 2647 has links pointing to it--thus increasing their search rankings.

    Full article at:
    http://searchenginewatch.com/showPage.html?page=3631269

    By Sherry Holdridge on 10.23.2008

  • Thanks for conducting and sharing this lab experiment...and yes I did tweet this!

    By Chris Herbert on 11.12.2008

Name:
Email:
Please enter the word you see in the image below:


Your comment: