Measuring the ROI of Online Support: It’s hard but rewarding
by Francoise Tourniaire

This week, I got two questions that looked very different at first blush. One said:

“I’m trying to measure deflection, i.e. customer self-service (web) resulting in not having to involve a support agent. Is the exit criterion when s/he leaves our domain, closes the browser, and/or logs-off our support site?”

The other one asked:

“I'm looking for ROI-like metrics to help prove the benefit of our online support site. Not a complete ROI, mind you... Just something to help the execs understand why continued investment in our support site is important.”

This must be a good week for talking about ROI for online support.

The good news is that online support typically produces great financial returns. The bad news is that creating a realistic ROI analysis can be a little tricky. As is often the case with tricky projects, ROI analyses can be done badly, and they can be made to tell outrageous stories in the blind pursuit of the creator’s goal – usually resulting in the purchase of a new self-service tool. Let’s look at how to create an honest ROI analysis you can be proud of.

What are you spending?
Some of the costs of online support are easy to figure out. The cost of the tool, in particular, is usually easy to determine, but don’t forget all the other pieces that go with it including:

• The license fee for the online tool(s), appropriately amortized.
  
• Maintenance and support fees.
  
• License and maintenance costs for ancillary items, whether hardware or software.
  
• Implementation costs, including training, if it’s a new system. Implementation can run much higher than the license cost, depending on complexity.

Besides the tool, there’s the cost of the human resources needed to make the online system work. It’s often difficult to compute that cost, especially if staffers only spend part of their time on online support. Do your best to capture it, even if you have to estimate. If the online system includes a knowledge base, people costs typically outweigh the cost of the tool.

What are you saving?
Most of the savings from online support fall into two categories: so-called “deflected” cases (cases customers resolve on their own, without human intervention), and time saved by support reps when resolving cases because they can leverage the knowledge base or other online tools.

Deflected cases
To measure deflected cases, you have to estimate how often customers find their own answers online rather than calling or emailing support. If you require customers to log into the online system (or have another way to identify them when they enter), you can cross-reference online tool usage with the cases logged that day to get an excellent estimate of how many cases were deflected. Of course, there will be some customers who use the online system to accomplish a particular task, then contact support with an unrelated request, but that should be a very small percentage.

If you don’t have a way to track who uses your online tools you’ll have to make do with the online usage measurements. For the best accuracy, track sessions rather than mere hits, and spot-check with the support staff on whether customers report having tried to help themselves before contacting support. Blindly counting each hit as a deflected case would lead to overestimates.

Once you have the volume of deflected cases, multiply it by an average case cost to get the savings. This is where the boldest amongst us cheat big time, sometimes without realizing it. Do not automatically use the average cost of a support case: Deflected cases are typically much less complex than the ones that get reported. Use a lower, more suitable figure, for example, the average cost of the bottom quartile of the simplest cases resolved by the support team.

Increased support efficiency
Estimating savings related to increased efficiency is even more difficult. I recommend using a simple task analysis, breaking up case resolution into a set of discrete steps and estimating the savings for each step. For example, if customers log their own cases, the 30 seconds it takes to check entitlement are no longer required. If it used to take an average of 10 minutes for level 1 reps to troubleshoot cases, but now, thanks to the knowledge base it only takes nine minutes, you’re saving 10% of the time required for simple cases. Using a task analysis gives you the discipline needed to make a realistic estimate.

Other savings
In some cases, you may enjoy more esoteric savings. For example, your online tool may let you do away with night-time support. Or you may be able to sell proactive monitoring and alerts as a support option. Or you may be able to increase employee satisfaction by freeing them from the more mundane cases.

Tread lightly here: if you can’t put a solid dollar amount on a particular item, it may be better to mention it as an extra without including it in the analysis.

Compare what you’re now saving to what you were spending and enjoy. Online support is typically very worthwhile, without any creative accounting required.

About the author
Francoise Tourniaire is the founder and principal of FT Works, a consulting firm that helps technology companies create and grow their support operations. She is the author of “Just Enough CRM”, a practical guide to selecting and implementing CRM systems targeted for business owners, including support executives. For more information, visit www.ftworks.com or call 650 559 9826.

How do you handle price increases to your support maintenance?
› View Answer