The builtin recording and monitoring features are manually triggered* by a supervisor in CSD on a call-by-call basis. Recording is done through CAD or the server directly if you can get all agent phones into a single SPAN/RSPAN session. The server will only retain them for seven days, though a supervisor can extend that to 30 and the server has a quota of 2.6Gb total storage for them which cannot be increased. You must manually export anything you wish to keep.
MediaSense on the other hand is the opposite to nearly all of the above points. It uses the phone's Built-in Bridge or a Cisco voice gateway directly to have audio forwarded to it (no SPAN required). It is automatic and can record 100% of calls. It can retain recordings essentially as long as you have storage. The unique thing about MS is that Cisco hasn't been willing to compete directly with any of the established market players. As such, MS is positioned more as a middleware that is for another product to integrate with. The idea is that MS offloads all of the capture effort so the user-facing product can focus on features. Cisco has bent the rule slightly by included a *very* basic Search and Play widget.
At the end of the day if the built-in feature doesn't meet your requirements you should be looking for a recording vendor. During the solution design you need to ask if MS completes the solution by doing the audio capture.
* It is possible to build a CAD workflow that manually triggers the recording automatically for each call. This does not scale well because the server will not delete old recordings to make room for new ones if the quota is reached. It will just stop recording. Also, CAD's days are numbered.
Please remember to rate helpful responses and identify helpful or correct answers.
