Using feedback loops to improve IT department service

As I’ve written here before, I strongly advocate thinking of IT in general as a service organization to the rest of the business.

Any service organization needs one or more forms of “feedback loop” to be able to gauge whether it is successfully accomplishing its mission.  However, I’ve observed relatively few IT organizations that actively seek to implement such feedback loops on a regular basis.  At best, the IT executive does it informally by consulting with his peers at the executive table.  But with any such anecdotal feedback, the information gathered that way tends to be fleeting and unreliable, and it is especially influenced by strong personalities and emotions during crisis situations.

Here’s a better, and simple, suggestion, one that I’ve implemented to varying degrees at several firms with a good amount of success: Survey your constituents regularly and then publish the results.

Sounds daunting?  I promise it really isn’t, not in this day and age of easy-to-use web-based surveys.  With less than an hour of work, you can design and initiate a survey using a free service like Zoomerang or SurveyMonkey, and easily gather high-quality results (reports and statistics) in just a few days that can help you gauge (and present) how you’re doing.  Here’s how.

You actually need to think about three broad groups of constituents whose pulse you want to take regularly (at least once a year, preferably more):

• internal customers (your help desk customers)
• peer executives and key business stakeholders
• your IT staff

I’ll focus here primarily on the surveying of key business stakeholders.  For the first category, you need to get a trouble ticketing system that incorporates user surveys after each ticket closure.  For the third category, well, that deserves a whole other post to discuss, since in some ways, it’s the most complicated and pitfall-riddled area of all.

For stakeholder feedback, most importantly, you need to start off (within the first few weeks of engagement in your role) by getting a general “baseline” understanding of current satisfaction, on the part of internal key business stakeholders, with IT in general.  Of course, don’t let a mere survey ever substitute for actually getting out and talking to as many people as possible about how good the service has been from IT.  When you send the survey out, emphasize to everyone that their input is anonymous, unless they choose to identify themselves in their comments. All you will know is whether or not any given individual has completed the survey.

Keep the survey as simple as possible, and you’ll get more responses.  You’ll want to balance free form input (usually just one broad question) with questions with quantitative answers, ones that will let you accumulate metrics over time.  And you’ll need to be prepared for some harshly stated criticisms, so get used to that idea.

Here are some sample questions I’ve asked in such surveys:

1. Compared to other companies you’ve worked at, how would you characterize the overall level of service of IT at this company?
Choices:

2. Of the following areas of responsibility of the IT department, which ones would you say most need improvement? (pick all that apply)
Choices:

3. Of the following areas of responsibility of the IT department, which ones are currently being handled especially well, in your view? (pick all that apply)

4. Compared to six months ago (assuming you were here at the company then), would you say that the IT department’s overall service to the company is…

5. Please give the IT department an overall rating on a scale of 1-10 in each of the following areas, with 10 being perfection.
Choices:

6. What other general comments do you have regarding IT, interaction with your team and department, etc.?

Send out the link to the survey in an email, explaining your purpose and goals in doing so.  Give your stakeholders (hopefully at least 20 people across the business; don’t arbitrarily limit it!) at least a week to respond, and nudge them every few days with reports on how many have filled out the survey so far.

When the survey finally closes, thicken your skin a bit (OK, maybe a lot) and take a good amount of time to look carefully at the results, and to compose a report to the people who responded.  By doing so, you’ll be showing that you take the feedback seriously, and have focused on specific action items as a result.  Then, you’ll want to send out the survey again in no more than six months; your subsequent report there can talk about the specific actions that you took in response to the last survey.

The good will generated by this service-oriented approach should not be underestimated.  Face it: it’s the right thing to do, and people will instantly recognize that.  As Mark Twain famously said, “Always do right. This will gratify some people and astonish the rest.”