[VoiceOps] What does you use for Hosted PBX and why?

[Fred Posner]

Another +1 here for open source.

I came to the open source side after working with Broadsoft and Acme
Packet... finding out that not only could we reduce unnecessary costs
with an open source model, we could also improve the customer experience.

Projects such as Asterisk, Kamailio, and FreeSWITCH are mature, tested,
and trusted by large institutions, carriers, governments, and more.

If you don't want to build your own, there's great products out there
building on top of open source, such as:

CSRP
http://www.csrpswitch.com/
high-performance SIP "Class 4" call routing platform for VoIP service
providers of all kinds

Kazoo/2600Hz
http://www.2600hz.com/

and... of course... you can always build your own (or hire a consultant
to help you).

--fred
http://palner.com


On 3/15/17 10:51 PM, Alex Balashov wrote:
> Since we work predominantly with service providers that take open-source
> approaches, we would be remiss not to plug that as an approach here.
> It's taken by a significant percentage of the industry.
>
> No, open source is definitely not free. But to those accustomed to a
> Broadsoft-type experience, I think there are a lot of misconceptions
> about it that stem from a lack of familiarity with how to approach the
> open product ecosystem. There's a perception that it's "highly
> technical" or, "Okay, I got FreeSWITCH to compile. Now what?"
>
> In reality, there are _lots_ of "packaged" answers to this problem, if
> you're just willing to look. If you don't want to build anything
> yourself, there are many good commercial systems built on top of an
> open-source technology stack. They cost some money, but nothing on the
> order of the big brands, and are competitive and growing in the larger
> operator and enterprise markets. Example:
>
> https://integrics.com/enswitch/
>
> It's built on top of Asterisk, Kamailio and MySQL, but very turn-key.
>
> If you have the engineering core competency to build something yourself,
> the ROI is excellent. Yes, there's a cost and a GTM lag, but once you
> sink the cost, nobody's going to hit you up for $MEGABUCKS for another
> 100 seat licences ever again. And again, it doesn't mean you have to
> write a million LOC multitenant software product yourself.
>
> There are lots of people on this list who provide hosted PBX with
> Asterisk or FreeSWITCH and have not had to do this. There are many
> approaches; you can certainly roll a home-spun multi-tenant platform if
> you really want to, but you can also sell individual instances of
> ready-to-go PBX distributions to customers inside a virtualised or
> containerised environment. In the latter, the port density / unit
> economics relative to hardware are excellent. You can put 100+ such PBXs
> on a single commodity box. Automating deployment with all the tooling
> out there these days isn't that hard. A lot of the tools already exist
> in FOSS land, if you just look around. And you can use feature-rich (and
> easy white-labelled) PBX distributions such as FreePBX or FusionPBX out
> of the box. Just Google around.
>
> The companies who have put a little bit of work into this have
> definitely forked out some CAPEX (mainly engineering time or
> consultants), but their ongoing OPEX commitments are comically low, and
> accordingly, their gross margins and cash flow are that much the better
> for it. What's more, the sunk cost tends to be largely fixed, so your
> ROI gets better and better with subscriber growth.
>
> It's not for everyone. Some organisations are sales-heavy cultures best
> suited to selling cookie-cutter products and don't want to do anything
> nerdy, or can't. But there are lots of people on this list making
> excellent money with open source, and I count some of them among our
> customer base (though we are not in the C5/hosted PBX platform
> business).
>
> -- Alex
>