I dare say that contact center project failure rates are similar to those of CRM project implementations:
- 2001 Gartner Group: 50%
- 2002 Butler Group: 70%
- 2002 Selling Power, CSO Forum: 69.3%
- 2005 AMR Research: 18%
- 2006 AMR Research: 31%
- 2007 AMR Research: 29%
- 2007 Economist Intelligence Unit: 56%
- 2009 Forrester Research: 47%
These numbers do not look promising. I also believe that contact center projects are often more complex than CRM ones.
Implementing a contact center project often requires the participation of technical expertise from telecom, networking, databases, desktops, and enterprise applications, and operational knowledge from senior management, supervisors, and of course, contact center staff. Not to mention occasional involvement of external vendors like the telephone carriers and any outsource organization.
Based on my project experiences, I believe the following are critical in ensuring a successful contact center project:
Hire a technical project manager. There are people who propose that a PM doesn’t need technical knowledge in order to run a successful project. That may very well be true, but with a contact center project it’s best to find a PM who is not afraid of digging deep into the technology and also having good PM skills. Contact center projects are complex because they touch upon so many technologies, and a technical PM would definitely add tremendous value to better the odds of success.
And picking a PM who isn’t liked by clients but gets things done is probably good. To paraphrase a PM friend and ex-coworker of mine, Don: If the client doesn’t like me then that means I’m doing a good job for the project.
If your own organization doesn’t have somebody suitable, then by all means hire from an outside source. I have seen too many cases of clients trying to save money by installing one of their own to run a contact center project. Even if this person is the start IT guru within the company, he’s almost useless if he doesn’t have the telecom knowledge required.
Vet the team members. Assemble the project team as if you are running a political campaign. Interview each team member, call out their skills and experiences in their resumes, ask for multiple references (from past clients and peers), and scour the Web for information about them. Why? The contact center industry is quite a niche, so that presents two problems: 1) Good people are hard to find; and 2) Bad people get away with pretending to be good.
And because of #1 many clients often settle with whoever they can afford to pay without vetting the contractor. The good news? Because it’s such a niche group of people, you can probably get plenty of reliable references easily.
Question the technical promises from the vendor. How do you know when a salesperson is lying? Well, you won’t know unless you know who and what to ask. Again, this is why it’s important to have a knowledgeable resource in contact center technologies readily available, especially during the sales demos and presentations. The contact center business is highly competitive, and with dwindling margins to be made in professional services, more often than not the profitability of a project depends greatly on making the sale — unfortunately, sometimes with empty promises.
Understand that there are technical limitations to CTI because the software depends on the telecom equipment, primarily the PBX. To make matters worse, there are half a dozen major PBX vendors each with their own implementation of CTI messaging. Therefore, if your enterprise consists of multiple PBX platforms, it’d be unrealistic to expect that the CTI or softphone works the same everywhere. If a salesperson makes a pitch that the desktop experience will be the same throughout the enterprise in a mixed PBX environment, then he or she is not being completely truthful.
The other frequently made sales pitch is for additional software licenses and hardware in order to implement a redundant system. Sure, redundancy is wonderful and makes the bosses think you’re doing the right thing for the company, but think carefully about the necessity and various architectures to support it.
In a typical contact center, having the CTI system go down isn’t the end of the world. So your agents don’t get screen-pops and can’t use the fancy softphone — big friggin’ deal. If the PBX is still operational (and most of the time it will be), you have no excuse to run around like a headless chicken. As long as the PBX is up the contact center should not lose calls and can continue to provide service. Obviously the talk time and wait time may suffer a bit, but your agents should have been trained on dealing with a no-CTI situation; in other words, they should still remember how to do their work prior to the CTI implementation — with pencil and paper if required.
If you must have redundancy (for example, your boss needs to burn the remaining IT budget), then determine the failure points first. Concentrate on how to remedy single points of failure in the system. Sometimes you won’t need more hardware but just a well-documented and well-executed procedure.
Avoid being an excessively demanding client. That’s right. Sometimes a project fails because the client is way too difficult. The vendor gives you the best-in-class hardware and software, the contractors work their butts off to design and implement an awesome solution, and the PM is pushing everyone to make deadlines and stay within the budget. All that won’t help a bit if you, the almighty client, is acting like an ass.
I had worked on a project once which the client team never had anything good to say about the product and the project. Constantly we’re hearing complaints and gripes and accusations from the client during the implementation. The client site was a black hole of negative energy that’ll suck your soul dry. They were arrogant, demanding, and near habitual liars.
Nobody likes dealing with assh**es because guess what? Everybody already has one.
Do you have any other tips or insights to share regarding contact center project implementation?