Recommendation for a new IP phone system needed

My company is looking to get a new phone system.  We have around 40 employees.  All employees have desk phones and we have several other common area phones (Conference rooms, lunch room, etc.).  We need around 50 extensions total.  We also would need several analog extensions (we have 2 analog polycom conference phones and a couple of fax machines).  Currently we use an old Vodavi PBX.  We have 1 PRI line and several POTS lines coming into the building.  We are also only in one location.

I have looked into some asterisk based systems, and some online hosted systems.  I also hear good things about Cisco's UC500 series.  I do not have telephony experience, so we would either need to go with the online system, or get a vendor to set it up.  

Our needs:  Auto-attendant, call routing, multiply Toll Free numbers directed to the different auto-attendants, multiple blast(hunt) groups, detailed call/extension reporting, voice mail to email

Our wants:  Presence, call recording, voice conferencing, mobile/smart phone integration

Obviously, these days, cost is a major factor, but we need something robust and stable and would pay a little more for that, rather than having something on the cheap that we have to fix continually.

Does anyone have any good recommendations that they are using?  What are the pros/cons of using these systems?  What range can I expect to pay for that solution?

Who is Participating?
lrmooreConnect With a Mentor Commented:
Sounds like your desires fit Cisco's Call Manager Business Edition. Full blown Call Manager, Mobility services, Unity Connection for call handlers and voicemail, voicemail to email, conferencing, all of the features in your list with the exception of call recording. There is some inherent capability, but not much. You would still need a voice gateway router, something like 2801 to handle the PRI. You would have to have a vendor come in and set it up for you. Requires 2 servers, one for CMBE and one for Presence. The Presence is far and above what you might expect.
I would not recommend the UC500 for anyone. They are limited in functionality and difficult to deal with.
The step in between is Call Manager Express on a 2800/2900 ISR router platform. Basically the same as the UC500 without the hardware restrictions and with full Cisco TAC support.
Depending on your existing network switching infrastructure, you may have to look at upgrading that also to something that supports VLAN's and QoS for the voice.
You can expect to pay in the neighborhood of 50-60 Grand, depending on all the features that you want, the types of phones you get (from basic to high-res color touchscreen executive level), and the implementation vendor.
I've been going around behind other vendors that have sold and installed things like NEC, Avaya Office, Panasonic, and even UC500 and ripping them out to install a "real" phone system like the Cisco UCMBE.
Anything by Cisco or Nortel should be good :)
Reliable easily scaleable, Works with dect and sip phones too.
Fitted and supported by these guys.
Improve Your Query Performance Tuning

In this FREE six-day email course, you'll learn from Janis Griffin, Database Performance Evangelist. She'll teach 12 steps that you can use to optimize your queries as much as possible and see measurable results in your work. Get started today!

We have been using the Epygi Quadro solution with Snom phones for about 4 years, the quadro can intergrate with Snom, Aastra and Yealink phones. it can integrate with Cisco handsets as well.

I believe the Quadro and the handsets will cater for most of your needs, and it is really easy to configure and maintain. Alloy's support for the products is really good. go to the following link,
 the products are worldwide so if you let me know your country (if your outside of Australia) i can get you details of who to speak to for advice on what you would need to implement. I am sure it will be a lot cheaper than Cisco :)
jfaubiontxConnect With a Mentor Commented:
If support down the road is a consideration, don't go with a Nortel product as they circled the bowl are on the way down the toilet. Product support for them may come from other providers but I would not hang my reputation in that. NEC and Avaya both have good systems if your willing to be locked to only their phones and support. With them and Cisco expect to pay a licensing fee for everything you want to do.

There are many different VoIP systems on the market now. Rolling your own is even an option if you have the time. The best recommendation I can offer is to call a consultant, let them analyze your business communications needs and then recommend a system. Try to find one that will consult with you even if you buy a system from someone else. There are many factors involved. A good consultant can ask some key questions and you will save a bunch of money in the long run.

ccomleyConnect With a Mentor Commented:
Nothing you say gives me cause to jump to a particular solution, so I'll mention the systems we have found successful to maybe narrow down your shortlist:-

Hosted systems are always worth considering these days!! We use a Mitel based service called "Communicator" in the UK - I believe they are available across the pond. But there are plenty to choose from. It takes away a lot of the maintenance hassle, you just own the phones! Communicator ticks all your "want/need" boxes.

I wouldn't go Asterisk, simply because of the learning curve. If you want a "soft" PBX, consider 3CX. This runs on Windows (very small footprint - you could run it on an XP machine or an existing server!!) and if you want a fully VOIP system, you need nothing else. If you wish to support PRI, POTS, etc., trunks, you need to use gateways such as Mediatrix. The 3CX system will *program these for you* so you just need to choose a gateway from 3CX's supported list and the hassle level is minimal. 3CX also directy supports a wide range of popular phones.  It incorporates all your needs and wants except possible the mobile integration bit - depends what you're actually expecting. You can forward calls to mobiles.  3CX user interface is *very*well thougth out, the learning curve is minimal.

If you want a hardware PBX, consider Shoretel or Epygi. These are modular and have optional-extras so you may need to include several extras to tick all your boxes. But they both give you a highly reliable system with supported hardware. The Epygi GUI could do with a re-design, I wish they'd look at how 3CX works then re-think some of their menus!

Maybe it is different in the UK but we find that hosted solutions here in the states don't scale well beyond 3-4 phones. Many high speed providers here are 6-10 mb down but typically only 384-512k upload which greatly hinders hosted solutions. In the case of customers with more than 3-4 phones we typically recommend an onsite VoIP PBX. Which one depends on the customer needs and experience level.

We like systems based on Asterisk but we do agree that if you plan on rolling your own there is a steep learning curve. As for running a phone system on Windows, there is no way I could possibly recommend that. If I want a server to be able to run 24x7 with very little to no hands on contact, I'm choosing a Linux based system every time. We kid about the Microsoft method of troubleshooting by doing a reboot. Yes, it may work but do you want to have to continually reboot your phone system?

We offer two different hosted sytems.

One is based on individual phones and as you say scales not well above four or five phones on one site - though it works well for companies who ahve lots of sites with few or single people at each (we have more than a few "work from home" clients).

The other is designed for larger systems and scales nicely up to fifty or even a hundred seats per site, and also works nicely on multiple sites. It doesn't require a Voip channel per phone, and "internal" calls don't get carried out and back but just go phone-to-phone on the LAN. This way the backbone connection only needs to scale to the requirement for concurrent *external* calls just as it would with an VOIP PBX based system.  And the management sytem is excellent.  It's a great solution for clients who want to avoid hardware capex and maintenance costs, upgrade aggros, etc.

I find it hard to believe this service and others like it are not available on the other side of the lake, indeed, I happen to know of a VERY large operation on Mahatten using it, though I'm not at liberty to divulge who. If anyone *is* interested I'll attempt to find out who the local suppliers are over there.

As for a soft-PBX, I would still recommend 3CX (and it has a free trial edition!) and there's no problem running it on a modern windows system. It will run on XP and up but I run ours on a Server 2008 platform on a server grade PC costing only £350 (about $400). The main problem with windows systems (and, indeed, mac and linux systems) arises when you run lots of different apps on the same system. In fact I have clients running 3CX on virtual PCs running nothing else just to avoid this issue. 3CX has thousands of installed users, no problem - anti-windows paranoia isn't helpful. With any VOIP system you have to *know* your onions, both in terms of the PBX paltforms you choose to run, and how to make sure your LAN is suitably configured. If you treat it with respect, there's no issue.

Final point - 3CX can be backed up in about two minutes, and restoring a backed up config to a newly installed system takes no more than five, so you can if necessary totally re-create the entire PBX (including voicemailbox contents if you included them in the backup) in less time than it takes the users to notice the box is offline!!
Most of my customers use their systems for calls coming in and out not station to station. So using VoIP with the limited upload speeds is still a problem for your larger system. I do agree that using it for multiple sites and telecommuters does work well. Additionally large corporations, in large cities are much more likely to have fiber or at least metro Ethernet which make the hosted system work better.  
However the second part of why I do not believe that hosted solutions scale well is the cost. As an example consider a customer of mine that recently switched from an old key system with analog lines to a hosted solution. His IT person suggested using the hosted solution because his company used a hosted solution and they were happy with it. Unfortunately he only has a single line coming into his office and over DSL it works fine. So he suggested using the DSL they already had but failed to consider bandwidth required for 22 phones. After the bad experience we were contacted to make it better. We proposed a system using a PRI. This allowed them to keep their DID numbers, removed the bandwidth issues and saved them money over the hosted solution.  
Their hosted solution wasn’t working properly because they did not have enough bandwidth but even with the DSL line their annual cost was nearly $18,500. Our first year cost with keeping the DSL line for Internet, the PRI, the system and installation was $14,208. The annual cost after that is only $5,160. Another way to look at it is the 5 year cost. The hosted solution had a cost of $92,321 over 5 years while our 5 year cost was $34,845. Even if we add the maintenance contract to handle everything for them, we save them more than $50,000 over 5 years. Maybe I just don’t charge enough I guess.  
As for the Windows paranoia, I love Windows and I love Bill Gates. If it were not for Bill and Windows, we would not make the money that we make. I think that goes for most of us on here. Also if I were to run 3CX it would definitely be on a Windows Server rather than on XP. Consider that the hardware is the same for both systems. The box is the same; the interface cards the same, so no difference in price there. However trixbox CE and AsteriskNOW are both available for free. Yes, you can download a trial version of 3CX for free but it is limited to 4 concurrent calls. If you were only talking 4 concurrent calls I’d probably recommend the hosted solution. You also have the cost of what ever flavor of Windows you decide to run it on.  
As for your final point, I really have not seen that much difference in speed in backing up either system. However if your going to bring up the time to restore the system you might want to also add in the time to install Windows, 3CX and then restore the system because if you are having to go back to the backup you have to consider why. Could it be one of the millions of virus, Trojan or other malware that targets Windows? Linux only has a few hundred targeting it and all of the require file permissions or simple passwords that should not normally be set. Do some research on mission critical systems and you will find that most run some version of Linux.      

I don't think this is really the place for this discussion as it doen't really add anything to aid the guy who asked the original question. Suffice it to say that I can see how in YOUR case all of the solutions I suggested won't apply. But subject to finding out MORE about the requirements of the original questioner, all the solutions I outlined WOULD potentially be suitable for him!!

And I agree about thre restore time EXCEPT you can install 3CX and restore a backup to it in abotu five minutes and if you didn't *have* replacement hardware ready you could TEMPORARILY run it on just about any available existing windows machine as an emergency measure. The mere fact that you COULD consider doing this is a plus point for the system IMHO. Any other PBX based service, you;d have to wait for spare parts or spare units to arrive, possibly along with an engineer.  But if *we* were setting it up, a 3CX system would not be prone to trojans etc., cos it would be properly protected and not used for anything else...

... it stands to reason that ANY POSSIBLE solution to any question offered on this system is always subject to *sensible* proper security, management, sane setup, etc... any any solution taken without due care could end up being mis-installed, abused, broken...

I agree with your last statement about the security of systems. The point I was trying to make is that Windows systems are targeted for attack far more often that Linux systems. We have a hard time justifying an operating system that is 10-20 times more likely to be targeted and costs more to install. But that is just us.

I do disagree that this is the place for this discussion or that it doesn't aid the asker. The asker specifically states that he requires 50 phones. For the reasons that I specified a hosted solution does not scale to 50, yes five-zero phones. Is the elimination of a possible solution through it's limitations not just as important as the mentioning of said solution? As I also mentioned we offer hosted solutions as well but based on the asker's requirements we do not believe a hosted solution is in his best interest. At least not in the interest of his bank book.

And I say that a hosted system DOES scale to 50 and more - if it's the right system and properly implemented (which includes providing it with sufficient bandwidth.)  I will admit that perhaps the system you supply doesn't scale well enough! The system we supply was just implemented for a major event with about 270 phones in total around a single campus sized site. The customer was very happy with its performance! It was set up only for three months for the period running up to and during a particular event. The actual users of the system were not just the main customer but several hundred of THEIR customers present on the site for the duration and they were all happy with it too! It was, of course, properly designed and implemented, not just parachuted in on an existing lan. :-)

Again a change in the scope. Your temporary event makes perfect sense for a hosted solution because no one wants to have to resell the equipment after the event. They are more than willing to trade cost for convenience at that point. But just because it works for a large scale temporary event does not mean it works for a medium scale long term customer.

Just be sure you calculate the costs over the expected life time of the system and not just the initial setup. There is far more to the costs than the up front out of pocket.
You're going to keep digging for excuses arn't you? Perhaps instead you should go out and have a look at a few of the more up-scale hosted systems on the market!! :-) I don't disagree you should consider the TCO. But most "trad" systems are worked out to a cost-per-desk-per month by most customers, amortising the major costs and rolling in maintenance fees, et al. That makes it quite easy to compare with the month-on-month costs for a hosted system. Just quoting five-figure TCOs for ten year ownership might make for a fightening headline, but you can do the same for an "owned" system.  Esp if you factor in taht a hosted system will always have the latest features, and an owned system may hit an upgrade limit and you find you have to junk and re-purchase to get onto some new feature if you decide you can't live without it. Or if you bought a system that only scales up to 80 users when you had 20 users, and now you realise you'll hit 120 desks by the end of the year (I have a customer who's just done exactly this and is now cursing the day some flash salesman sold them an Avaya IP Office, the cost of new kit to get them enough phones for their current staff level is HUGE.)

I take it that by "digging for excuses" you mean pointing out that your keep changing the scope of the system. We were beginning to call what you were doing grasping at straws. Like I have mentioned before, we sell hosted solutions as well as current open source VoIP systems. The numbers just do not support using a hosted solution unless that is all you sell and then everything fits a hosted solution I suppose.

Many times "up-scale" just means a fancy over priced fad. However when it is all said and done, the features lined up apples to apples, the disadvantages lined up lemons to lemons, form most businesses it does come down to the cost per seat per month. It also really isn't about the five figure TCO but the 3-4 figure difference between the TCO of each system.

Looking into the crystal ball to determine just how many phones you will need can sometimes be tough. Which is why we like to point them to the VoIP systems that don't have the proprietary licensing issues of the Avaya, NEC, Cisco, NOtel, ESI and other systems that make these future upgrades so costly. Thankfully the speed of today's computers make those limits more of a thing of the past. Even then adding a second server in a high availability mode isn't that expensive. As you pointed out the hardware is only a few hundred bucks. Using standard SIP phones means that we have several manufacturers to choose from and no licensing fees keeps costs lower.

Just out of curiosity, what happens to that hosted phone system if/when the provider goes out of business? Does it still keep working like a traditional system?
We use the Grandstream GXE IP-PBX and it (for the last 12 months since they got their firmware right) does everything we want for a small business - in fact it does a lot more and does it well with no ongoing costs. Combined with HT503 at staff homes we have a nice integrated system without incurring the costs other systems would incur. We have international phone numbers, staff home extensions, multiple call routing etc.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.