CSAT by Channel: The More Assistance, The Higher the Score
I’ve had lots of questions about channel traffic since my recent post, “Interaction Volume by Channel: The 2011 Outlook,” some people asking about effectiveness by channel, or customer satisfaction by channel. I’m spent some time playing with the data this morning to arrive at some satisfaction numbers by support channel, and the results are interesting: the more a service is employee is involved, the higher the score. The less the amount of employee touch, the lower the score.
First, a few caveats about the data.
- I’m using data from our brand new benchmark survey, so the number of responses is low compared to the 300+ responses in our old survey. But, the new survey has more channel options, including chat. NOTE TO TSIA MEMBERS: Please enter your latest data in the new benchmark survey as soon as possible!
- The data combines results of multiple scales. I’ve converted scores to a 1-5 scale, my personal favorite. (1 = very unsatisfied, 5 = very satisfied) I don’t want to go down a rat hole here, but 10 and 11 point surveys have lower average scores than 5 point surveys. Any time you factor multiple scales into a single average I think the final result is a bit iffy. But a few hundredths of a point one way or the other won’t change the findings on this chart.
- I’ve counted data only for direct employees, not satisfaction with outsourced employees.
- The data is for “overall satisfaction” with the interaction, not speed to answer, ease of use, customer service skills, etc.
With all that in mind, here is the chart:
- The m0st labor intensive way to solve a customer problem–dispatching a field service employee for an onsite visits, tops the chart with an average satisfaction score of 4.64. I’m happy that customers love field engineers, but with the cost to roll a repair truck over $1,000, I don’t see companies wanting to shift more traffic to field engineers in order to raise satisfaction scores.
- Phone interactions with support technicians rated 2nd highest, with an average score of 4.33. Web chat, which is still assisted support but a less personal experience than phone, received an average score of 4.05.
- Email, which is an assisted support channel, though less personal than a real-time Web chat with an employee, rated a 3.96 average satisfaction score. Note that some frequently asked questions may leverage an email auto-response tool, making email an unassisted channel for common problems.
- Self-service rated dead last, with an average satisfaction score of 3.27. This is hardly a surprise with the evidence of poor self-service success I’ve written about before.
I suppose it is no surprise that the more assistance the customer receives, the better the overall experience and the higher the satisfaction. But it raises a lot more question in my mind about emerging channels and self-service: are we doing enough work to understand customer expectations? I’ve seen many surveys–including some of our own data–showing customers demanding unassisted support channels. Demographic surveys show that younger customers avoid phone interactions, preferring unassisted support or less personally assisted support channels like email. Are these customers tougher graders, or are we really dropping the ball in delivering the experience they expect? I think some significant work needs to go into surveying younger customers about their expectations for support, not just channel preferences, to understand these numbers, and begin delivering the experience customers expect.
What do you think? What will it take for unassisted channels to catch up? If anyone has insight or examples to share, please chime in. And as always, thanks for reading!