Support Communities and Case Deflection

challenges

(Nick Emmett) #1

Do you manage a support style community? Do you measure case deflection? (If not why not?!)

If so, which platform are you using? Is the case deflection part of the platform or do you have a third party tool? How easy or hard is it to get your deflection metrics? What action are you taking based upon those metrics?

Our Community is built on the Salesforce Community Cloud platform but our case Deflection comes from a third party tool we use to power our search and case creation - Coveo, as Salesforce doesn’t yet index external web sources, which is where our online help documentation is stored.

I’m interested in how others are approaching the whole concept of case deflection as it’s something we’ve just started analysing “properly”!


(Richard Millington) #2

Lithium tends to rule the roost on this one.

But it does tend to vary a lot.

The Lithium model is largely…

a) Divide your customer support costs by the number of tickets to get a cost per call/ticket.
b) Count the number of accepted solutions.
c) Measure traffic to questions with an accepted solution. Use a survey to discover what % of them had their question resolved and what % of those whose question was resolved would otherwise have called customer support. Now multiply these metrics together.

Now you add answers b) and c) then multiply that number by a) to get the total value of calls deflected. That gives you the return.

Assuming you can setup a pop-up survey for visitors, you should be able to get these metrics pretty easily from a combination of Google Analytics, internal data (ticket numbers/costs), and survey results.

I’m sure there are some 3rd-party platforms that do well. But I’m struggling to think of them at the moment.


(Kristen Gastaldo) #3

Just to throw a wrench in the works - I used to look at our top case creators vs our top community members, thinking that active community members would create LESS cases.

Yeah, I was WRONG.

We found that customer who were active in one channel, were often active in ALL them. They post more in the community, create more cases, and some even used the chat channel often. Now, this doesn’t necessarily mean bad news. Sure, they create more work for us, but they are more engaged users, using more areas and functionality of the product and answering others’ questions in the community.

We DID see an overall drop in case creation (per organization) after creating our community (we think due to better educated users, less of those “how do you” type questions, and proactive support with known bugs, etc).


(Richard Millington) #4

This is a really important point and I suspect happens quite widely. The more active people are in a community the more the posts questions they are likely to ask (and answer). So the community might be creating cases it’s resolving. Though I think this should be covered in the pop-up survey (as they shouldn’t, in theory, select they would’ve called customer support next).