Considering Hosted IP-PBX Offerings
Zeeek posed an interesting question earlier today : “What hosted service would you choose for a non-technical friend?”
Here’s my take on the subject:
Configuration Flexibility
The whole idea of hosted PBX is that you retain the flexibility of having your own PBX, but avoid the headache of its physical installation, maintenance and cost. There will be some configuration, maybe more than a little. There can still be a role for a consultant if the end users are truly non-technical.
Ultimately the ability to craft a dial plan that will adequately serve the end user will revolve around the configuration flexibility of the provider. Here are some features to look for:
- Ring multiple extensions at once
- Ring multiple extensions in sequence
- Time based routing logic
- Route calls to external numbers via PSTN (cells phones)
- Route calls to SIP URIs
- Restrict/allow international dialing (per user)
- Dial by name directory
- Custom voice prompt recording
- Time zone aware controls, for dealing with offices across the continent
- 911 support, per location
- Fax support
- 24/7 telephone support
To anyone familiar with Asterisk the management portals of hosted PBXs are similar to the GUI frameworks for some Asterisk distributions. There’s been much and very heated debate about GUIs for Asterisk. The reality is that any GUI going to limit what you can do compared to manually editing the dial plan. Similarly, with a hosted PBX there is no way to manually edit the dial plan, so you’d better be happy with what the GUI offers.
Of course, most hosted PBX offerings are not wholly Asterisk based anyway. They tend to be something more scalable, like OpenSER with Asterisk as a media gateway.
Consider the Technology Model
Does the end user have a hardware preference? Is the host happy to support the kind of hardware that the end user will appreciate? Many providers will only support a short list of hardware since that limits their requirement for configuration support every device in the SIP universe.
I like Polycom and Aastra phones. I’ve used them for years. We already owned them so there was no sense in migrating backwards to a provider that only supported analog phones at ATAs.
Consider the Company
The largest and only publicly traded player in the space is 8×8, owner of Packet8. Nuvio is substancial, and been around a long time. You want to pick a company that has an established track record and truly has a business focus.
Here’s some sage advice from the CEO of Vocalocity:
My Personal Experience
Last summer I was investigating hosted IP-PBX providers for my own purposes, intent upon transitioning my home and home office to a more rational combination of solutions. In my searching I came upon OnSIP just at about the time that my employer saw the logic in migrating to a hosted PBX for our US operation. FWIW, OnSIP doesn’t support Aastra phones but we chose them anyway since they were otherwise the best match to our needs.
For two years prior to this my numbers were hosted through CarolinaNet, which a is a reseller for Nuvio.
The experience with Nuvio was really very good. They have solid tech support, handled number porting well and have a 911 solution. The trouble is that their billing model is the common “unlimited minutes/month for $x/phone.” That has cost implications that we wanted to avoid.
Metcalfe’s law states that value of the network increases in proportion with the square of the connected users. This holds true for hosted IP-PBX situations. The more extensions you have the greater the utility of the system. With a fixed per end-point (user) cost the cost of the solution can add up quickly as you add extensions.
I’m just one user on our hosted PBX, but I have five extensions in my office, and two more in the house. That could be $150-300/mo just for me. Add a dozen more users and the monthly cost can skyrocket.
Also, do your people travel? A little or a lot? If they are frequently out of office then lots of minutes available to their desk phones will be less than ideal.
The point here is that generous or “unlimited” minutes per device is not necessarily a good deal. That model makes big presumption about your usage patterns. It works for call center, but not for all businesses. In our case it made more sense to pay for actual usage. That’s literally no free minutes per month.
Were it not for OnSIPs unique billing model I think we’d have tried Vocalocity for Pixel Power US.
Look For Simple Solutions Where Possible
For some very simple situations hosted PBX may be overkill. Some phones provide considerable flexibility internally. My old Panasonic KSU (KX-TG4000) was like this. It could map POTS lines to extensions, provided multiple VM boxes, hold, conference, transfer, DND, etc. This combined with some dialing logic in a 4 port ATA and you could have a solution for some people.
Best Of Both Worlds?
If you accept that some consultancy is necessary at least periodically then it broadens your options. For some businesses having hardware on-site makes sense as long as remote support is readily available.
I find Arretta an interesting option. They offer an Asterisk based hosted PBX where you can get as much or as little support as you need. I think that an Asterisk consultant could do a good business using Arretta as the hardware/software and selling the configuration & support as a service.
If hardware on-site makes more sense then Star2Star might be worth a look. I know that Kristian Kielhofner, the project lead on Astlinux, is one of the group behind Star2Star . To paraphrase, “Astlinux has been very, very good to me.”
There are a plethora of Asterisk appliances cropping up. Given suitable connectivity any of these could serve as the basis of a hybrid hosted/on-site solution. Especially when coupled with a good consultant.
Summary
It’s not easy to recommend a hosted PBX to a complete beginner. There are many options out there. If they examine their needs thoroughly and ask the right questions then a good fit can be found.