site takes you to a copy of the original
SynerVox website.
If you login as democust with password gzambon72, you will be able to play
with the configuration pages of the example applications. As there are no
voice server and no telephone lines, you will not be able to actually try out
the services. For the same reason, I have also had to disable some functions.
You can also see the application examples by clicking on the Showcase tab,
but the configuration pages will be completely disabled. The
product page includes a brief
description of the four examples (VoxAppoint, VoxSurvey, VoxVote, and VoxBox).
Towards the end of 2001 I was living in Zurich (CH) and working for a media
company called Worldzap. I was completely fed up with them. In particular,
I was literally disgusted with my boss, whom for years (before he became my
boss, that is) I considered my friend. I thought I had discovered a market
opportunity in the provision of inexpensive and intelligent (more about that
later) Interactive Voice Response (IVR) services. These are the services that
most of us (including myself) hate: you dial a telephone number and are
confronted with a series of automated questions you have to answer by pressing
telephone keys.
These are traditionally quite expensive bespoke systems. The telco sells you
a full system, with dedicated servers. It will normally cost you a price in
the high five-digit range or higher. And then, obviously, you are stuck with
a costly maintenance contract at the tune of 20% of your initial cost every
year.
My idea was: what if I can sell an IVR service instead of an IVR box? I would
operate as an application service provider (ASP), with the voice services of
several customers sharing my servers. The customers would be able to
configure their system via the Internet by accessing my extranet. The
financial model would be based on initial set up costs (design, development,
and provisioning) plus usage, and initial cost could be recovered as
an additional usage fee, with minimum usage guarantee and minimum contract
duration (what in Australia, I discovered later, is known as a
plan).
One feature I consider key to the success of my enterprise was the possibility
for the customers to [re]configure and collect the results of their services
in real time. Normally, with IVR services, most of the parameters are
hard-coded, and the customer needs to contact the supplier in order to make
changes (and in most cases, unless there are special provisions in the
contract, fork out more money). I wanted to develop a very flexible platform
that would allow the customers to modify key parameters of their services
(e.g., upload a new message) on their own via dedicated web pages. And, most
importantly, then make a telephone call and hear at once the result of their
changes.
I did some market research and realised that nobody was doing anything like
that, not just in Switzerland but, as far as I could see,
in the world!
I decided to give it a shot and, on January 10, 2002, the company SynerVox AG
became operational. "AG" stands in German for Aktiengesellschaft, which
literally means share-based company. In Italy such a company is called
S.p.A (
Società
per
Azioni), in Australia
and the U.K. it is an unlisted Limited company, in French it is an SA
(
Société
Anonyme). With an initial
capital of 100kCHF, an AG it is much more than an Australian Pty Ltd or an
Italian S.r.l., but I wanted to give it a certain credibility from the very
start. AGs require yearly independent audits and more paperwork, but I
thought the additional effort and cost to maintain the company would repay
itself in terms of image.
In 2002, an AG required the majority of directors to be Swiss citizens. It
also required each director to have at least one share. Later, the law was
changed and I could become sole proprietor, but when I opened the company
I had to ask two Swiss former colleagues of mine to become co-directors,
although without any signing capability. Anyhow, I started by setting up
a development space in a room I was renting in the basement of the apartment
block where I lived. I subscribed a separate Internet connection and two ISDN
telephone lines, bought a Dell PC and a Dialogic ISDN interface, and got to
work on my new platform. Two servers were connected via an Ethernet segment.
A Windows server provided the interface to the PSTN
(
Public
Switched
Telephone
Network),
and a Linux server provided
the customer access via the Internet. I would have liked to use a single
Linux server, but I configured the voice interface card via a program that
only ran under Windows. It provided a comfortable GUI to generate the XML
configuration file and significantly shortened my development effort for the
voice part of the platform. Initially, the operational system supported a
dozen parallel telephone connections and resided on the premises of a small
engineering company that provided the Internet connection.
My first customer was the Italian Consulate in Zurich. They were struggling
to keep up with the requests for transit visas, and the Consul was desperate
for a solution that prevented riots before the Consulate doors (literally!)
I proposed a system that would provide automatic appointments. An applicant
would call a premium number (in Switzerland, a 0900 number), choose the
language (Italian, German, French, English, or Serbo-Croatian) and specify
whether he needed transit or tourist visas and for how many family members.
The system would then find the earliest possible time slots and propose an
appointemnt in the morning and one in the afternoon. The applicant then only
needed to press a key to obtain an appointment.
The applicant was identified by his telephone number, but the Consulate also
had the option of requiring the last four digits of a passport number. Only
one appointment could be taken from a particular telephone number, and the
applicant, by calling from the same telephone, could hear again the date and
time of his appointment, cancel the appointment, or listen to a message
describing the visa requirements.
The system was accepted and became operational on December 20th, 2002. Most
applicants were happy because they could get through without delay and
perceived the allocation of appointments to be fair (which it was). They
paid CHF 1.50 per call plus CHF 0.50 for each minute of usage of the system.
The Consulate didn't pay anything, and I got the money paid by the applicants.
More than 99% of the applicants paid less than four Swiss francs to obtain an
appointment. It actually worked so well that a TV consumer programme praised
the Italian Consulate for providing such an efficient and cheap service.
All other Consulates (e.g., from Austria, the U.S.A., and France) used a call
centre to give appointments. The applicants had to wait for minutes before
getting through, at a cost even ten times higher than that of the SynerVox
system. On January 1st 2003, also the Italian Consulate in Saint Gall
adopted the system, and the Consulates in Lausanne and Basle did the same
later during the course of 2003.
In March 2002, with only Zurich and Saint Gall, SynerVox grossed CHF 57,715.29
and the following month CHF 71,252.70. By then, the system had 30 telephone
lines. I tell you, it was a thrill to see them all busy. The total revenue
in 2003, the first year of operational services, was CHF 339,154.89.
Unfortunately,there were also some
problems. There was a sporadic bug that caused the Windows system to crash.
I was literally thrown out of bed by an angry Consulate employee asking why
the system was offline. I found the problem and solved it, but it took me a
while (weeks, because the problem was sporadic), and it was very stressful.
To make my life a bit easier, I developed a check that would restart the
system if an error on the voice interface was detected. It also sent an SMS
to my mobile phone to notify of the occurrence. It happened less than once
a day, but it was a vary bad feeling to hear my phone tweet the reception
of an SMS...
There was also another problem of a different nature. The Consulate in
Zurich employed six people in the visa section, and they were very unappy
when the system was introduced. Till then, they could answer the phone when
they pleased and write the appointments on a copy book. That gave them total
control of the situation. I don't want to speculate on the possibility that
some of them might have taken a bribe to provide appointments, but it would
have been very easy for them to do so. Suddenly, when the system became
operational, they had no control on the situation. The number of visa per
employee they made was half that of the other consulates. I suspect that
it was a mixture of lazyness and obstructionism. Anyhow, the Consul didn't
manage to whip them to a higher productivity, and the schedule filled up.
The system was designed to work if there was enough capacity from the part
of the Consulate. When the schedule filled up, and with no possibility to
increased the capacity, some stupid (for lack of a better word) applicants
started calling incessantly, in the hope that a slot would open. They
accumulated very high telephone bills and complained with the Consulate. It
happened perhaps once or twice a month, but it was enough for the Consul to
be unhappy.
In September 2002, somebody (I am pretty sure I know who that was) convinced
the Consul to switch to an alternative solution based on a call centre.
And this despite the fact that the call centres had already proven to be
a poorer alternative to my services. The Consul cancelled the SynerVox
service. He did it via a lawyer claiming that the service had security
problems (which was clearly untrue). The company that provided the new
service was brand new, with close to no experience in providing such services,
and had four shareholders, two of whom where people I knew to be friends of
an employee of the Consulate. The other two were unknown because they owned
unnamed shares. It's clear: the Consulate employee set up the company with
his friends and convinced the Consul to switch. The telephone bills of the
applicants jumped higher by a factor of ten or more, while the bottleneck at
the Consulate remained unchanged. But the employee and his friends had
managed to steal my largest customer. Meanwhile, the other three Consulates
kept the services running. No security problems there!
Meanwhile, I had rented premises for SynerVox. It was a small office with a
separate sound-proofed room for housing the servers. I had a cabinet with
a handful of servers and sixty telephone lines. Besides the Italian
Consulates, I managed to get as customers a radio station and a television
channel. I also built up a partnership with a separate company to offer
an automatic daily horoscope service. I absolutely do not believe in
horoscopes, but a lot of people do, and I worked for one and half year to
convince Alois Treindl to joined forces with me. Alois is the founder of
Astrodienst (
www.astro.com), a company
that provides astrological services online and in print. I then managed
to sell the service to Blick Zeitung, the largest Swiss newspaper in German.
Blick agreed to publish a premium telephone number below their printed
horoscopes. I provided the automatic telephone service, and Astrodienst
provided the horoscope texts and logic. I was very hopeful that a lot of
people would call to hear their personal daily horoscopes, but the service
was a complete failure.
SynerVox never really took off. Perhaps I should have sold out to an investor
in exchange for contacts and funding, but I wanted to keep my independence.
I had a couple of mailing actions and also tried to link up with some VoIP
(
Voice
over
IP, Internet-based telephonie)
companies, but
without success. Perhaps I was a bad marketeer, but I believe that the lack
of growth had a more basic reason: companies large enough to be interested
in IVR services simply went with Swisscom, the incumbent telecom operator; the
smaller companies, which could have benefited from my services, didn't really
think that such services would be useful.
When the European Union relaxed the requirements for transit visas, the
business with the Italian Consulates was significantly reduced. The other
services didn't bring much income either, and I decided to look for a job
before SynerVox started becoming an expense rather than a source of income.
I started working for World Television in July 2005, but kept SynerVox alive
till the last quarter of 2006. I gave notice to all existing SynerVox
customers and ceased operations in October 2006. To avoid keeping the
financial books for a decade, on 2006-10-15, I sold SynerVox to Leotrade AG
for 4 kCHF.
SynerVox sustained Monika and myself for a few years. I did everything,
including bookkeeping, setting up the premises, developing the platform and
the services, selling and maintaining the services, invoicing, calculation of
value added tax, ...
It was fun as long as it lasted!
.
