Tips for implementing unified communications?

It depends what you're migrating from, but in general you should deploy this in stages. Unless you're totally comfortable with IP, there are bound to be hiccups, and it's easier to manage these on a small scale rather than have your whole business be out of joint for a bit.

The VoIP piece should actually go fairly smoothly, but UC is still a work in progress. Whether you deploy in stages or not, I would recommend doing just the basics for UC, as there are way more moving parts around this than just VoIP. One of the strengths of UC is flexibility, so once you have some basic UC features working smoothly - unified messaging, visual voicemail, presence, etc. - you can then layer on the more advanced features as you go.

What are you comfortable with and how involved have you been in the design phase are 2 big questions that you should address as a company. The main players in IPT and UC are capable of doing a flash or phased cut but as Jon mentions you will most likely run into some bumps along the way. Are you prepared and capable to deal with this on a flash cut and under pressure? A 250 user cut is not that challenging over a 4 day weekend if all your homework has been done (network ready, user trained, system integration tested, etc). Now when you throw in a Call Center the ability to run this over a 4 day weekend does present some risk. Smaller CC tend to be less complex and easier to flash cut while larger ones have more applications and integration points that MUST be tested. Have you tested all your CC applications? Are your 3rd party vendors prepared to get involved over a 4 day weekend if something goes wrong? Can your business survive if the CC is not operational as planned? These are just a few of the many questions that your vendor should be asking and together the right decision will emerge for your business. I recommend first timers to test and implement in phases.

Hi Marcus -

The way you are describing the migration and move, you are talking about a PBX and voicemail, Call Center Apps, and the desire to add UC functionality to the new system.

If you are moving the office and you will be switching to a hosted VoIP provider or acquiring a new PBX and voicemail system, the new location will be installed, tested and running prior to the move. Your new voice trunks, whether TDM or SIP will also have been tested. The only phone system downtime that you should experience is the time it takes to move your DID's to the new location -- (side note....as part of your PBX decision, if keeping your existing DID's is important to you, make sure your new office address is within the same ratecenter as the old address. If not, you will be put into a position of having to change to new DID's, remote call forward the current DID's to the new location's DID's or be forced to move to SIP trunks in which case, you can move your DID's virtually anywhere. The first two options can be expensive and the third, SIP, will have the biggest impact on your PBX vendor selection. You'll want to make sure the carrier you select and the PBX you select both support the other's implementation of SIP).

How many of the 250 people call center agents? Is your company's primary business dependent upon the call center? If it is, one thing you might consider as part of a longer term business continuity strategy that could also reduce some of the risk involved in your move is migrating to a Hosted Contact Center provider prior to the move. By separating the call center applications from the phone system, your call center would not be down during the move. Rather, agents could login and work from alternate locations using mobile phones, standard analog lines or any other type of phone connection. Another benefit of Hosted contact centers is that if your business is seasonal in any way, most providers will let you increase or decrease the number of agents during the year vs. buying a system where you need to engineer the new system and licensing for peak periods even if those peaks are only a couple of weeks or months out of the year.

Personally, I would consider each component separately and leave the UC features for last -- and as Michael mentioned earlier, only to be implemented after ranking features according to which will have the biggest positive impact to the business.

Good luck with your move. Let us know how it works out.

Hi,
There are lots of very interesting points raised above, but my own view is that the voice component is probably one of the most complex; you can see that from Dennis' reply. As you can also see from above, we've all had to learn the language of telephone systems. I have, didn't want to, wish I could have got away with not doing so. So if you haven't got these skills find someone that does. Mittel are a good example (I am not affiliated with Mittel or, in fact, use any of there services myself - I didn't know about them then)

So first decision is whether to move the voice system first or last. When you do this pretty much depends on how you roll out, who is providing the system and also, just as importantly, how are you proposing to demonstrate value to your management. For most people hard cash can only be saved in the voice part of UC (call costs, line rental etc). If you need to demonstrate savings early doors, then voice first.

Having answered that question then decide what other features you need. There are so many variations on a theme here that using a technique such as MoSCoW (Must have, Should have, Could have, Would like to have) is a must. First round only implement the must have's, then look at the should have's.

Next I would look at rolling out the none voice parts to groups of people at a time. You can use their feedback to inform the following stages. Think about viral deployment - make a big song and dance about the features, but don't make anyone use it - works for google, worked for me.

Hi Marcus,

If I were in your shoes I would certainly take my time and do this in phases. Most PBX providers will let you take a 60 or 90 day test drive with software apps. Talk to your vendor and see if they can set-up a lab environment for you to test. UC is such a broad topic and everyone's understanding of UC is different. So, I would suggest trying to figure out what efficiencies you want to gain from the deployment and what applications are the employees looking for. Avaya and Cisco have some pretty cool apps that can reside on cell phones and turn any cell phone into a roaming desk phone. However, some of the employees might be happy if they could simply get their voicemails into their email.

So, my advice would be to define what everyone wants and test before a staged deployment.

Good Luck,

Derek Roush

I would implement any technology upgrade in stages (especially VoIP and UC) using a migration plan for the following reasons:
1) Ensure that you set up a pilot trial to determine the cost benefits and technology impacts when a select few employees are used in the trial and determine what you expect from this new technology.
2) If a technology upgrade is implemented in flash-cut mode and serious problems occur with the introduction of this new technology then you can migrate back faster and with less disruption to your network and business.
3) A migration plan will not overtax your employees with many changes at a time and the training will be faster and less disruptive to employees.
4) There will definitely be problems when migrating and it is easier to troubleshoot the problems with vendors with minimal technology changes in stages versus full technology flash-cuts.

Typically any migration especially for deployments has managed in stages, in order to avoid network failures and out of services, therefore avoiding any impact to your business.

I'm with the camp recommending a phased approach. Hopefully the vendors with whom you're working have done their homework, starting with an in-depth sight assessment to determine viability of the cable plant and network, and a full inventory of all the voice and call center applications currently in use, along with a business assessment to determine which VoIP/UC applications could be delivered to improve productivity and customer service that aren't being delivered today. Typically, the LAN will need work, i.e. layer 3 switching, and virtual LAN implementation to segregate and prioritize voice traffic on the network. VoIP/UC has a powerful feature set to enable business productivity - so a successful implementation of VoIP/UC should be about focusing on which features will drive business functionality - often that's vetted in a pilot setting with a controlled group of users. As the call center is incorporated in the plan, I would pay special attention to this class of user and study the applications being used vs. the customer needs. There will likely be opportunity to improve the user and customer experience by tuning existing applications.

A phased approach is my recommendation specifically if it involves a call center. Even if the call center is a basic "next agent in the queue" call center, it would still be beneficial to leave that for last and deal with the users that are non-call center users first. This will demand an integration between your existing PBX and the UC solution, to at least ensure continuity between the call center staff and non call center staff. LAN work will be needed to ensure treatment for voice being preferred over data, which should ensure a improved quality of user experience(QOE). To handle this on a 4-day weekend might be a stretch, and not leave you enough room to troubleshoot call flows, especially if you have multiple attributes being used for call treatements and routing to appropriate skillsets.

I think the consensus is a phased roll-out. I would have to agree. There are typically two areas of opportunity. The first is the installation and cunfiguration of your solution. People tend to discount the configuration. Yet, this can be a dificult step, especially in working our minor issues in your environment. The second area of opportunity is user training. A full deploymenmt of UC/UM/VoIP will significantly change the range of functionality available to your end users.

You can't assume that all of your user population will openly embrace the change you are about to throw at them. You probably want to over-communicate. Let them know your game plan up front. Provide demos of how the product set will work. Expect a series of short classes for each employee (don't try to show them all the functionality at nce). And, make sure to supplement with user tips/tricks, reference materials, and even some CBT classes (typically available fro your vendor).

Make sure to utiize a wide variety of resources available from your vendor and ask your vendor about best practices that they would recommend for deploying their solution. Professional services from your vendor will add some cost to the project but can be well worth the cost in reduced after-implementation support costs by your team.

Which ever platform you choose, don't forget about the headsets. It's important that if you spend the money on a new system that you don't scrimp on the device (headset, phone, audio conference device) that's your voice to the outside world. Check to see if your existing headsets will work. If you currently have commercial grade headsets from Plantronics, Jabra, Sennheiser or VXi, typically all you will need is the appropriate interface. i.e. QD to modular cable, QD to USB, amplifier, etc. For your wireless headsets, you may need a new EHS cable or handset lifter. Some companies wait until the last minute to worry about heasdets and end up going cheap due to budget restraints. I strongly suggest that you ask for some budgetary quotes so no one is "shocked" by the price tag for good quality headsets.