Broken Knight: Consequences of Rushing into Production

Tuesday June 23rd was a huge day for people like me in the video game world. Batman: Arkham Knight was releasing and all of the reviews had been stellar. Reviews were glowing about how flawlessly the game ran and performed on the PS4, which made me incredibly happy as the PS4 is my platform of choice. However, the PC port of the game was a complete disaster. The game was so broken on PC that the publisher has stopped selling the PC version until it is fixed. That is quite a blow to all the PC players out there and puts a really ugly stamp on the game itself. The game itself had been delayed twice, and it appears that even with these delays, the PC version was still rushed into production.

Rushing your contact center ACD into live situations can also have consequences.

  • Using untested IVRs: While using the Q-Suite’s visual IVR builder can be intuitive, the builder itself cannot apply automatic logic to the flow of the call. This logic must be accounted for by the administrator who is building the call flow for the contact center ACD. If the IVR is poorly constructed, odd or unintended situations may arise when a customer calls in. Test, tweak, and test some more to avoid this.
  • Under/Overestimating call volumes: If you speed along the turn up of your contact center without any prior data or analysis in regards to call volumes, you may end up in a scenario where you have a massive call volume and a grossly understaffed contact center, which is likely going to result in a larger than expected number of missed or abandoned calls. Customers can get very unhappy if they are unable to reach an agent. On the flip side, call volume is much less than predicted which results in having numerous agents sitting around doing nothing while getting paid to do so. Certainly not an ideal situation for the contact center owners.
  • Hardware: We touched on a few hardware issues in a previous post, and even talked about using some common sense when constructing your contact center ACD’s infrastructure. I cannot really stress this enough though. If you are expecting 200 concurrent calls, do not make due with hardware that can only support 100, just so you can be in production. It’s simply not a good idea.
I know that in the hectic world of software and contact centers, time is money, so deadlines get pushed and can get pushed hard. Just try to remember that implementing a poorly designed and poorly resourced contact center ACD just so you can be in production as fast as possible, could have serious longer term consequences that may result in not being in production at all.