We have recently visited the UPRISE festival in Amsterdam - a smaller version of Paddy’s Websummit, but possibly better as far as networking opportunities go. Or so we hear.

If you’ve been following our blog, you know that we are startupers ourselves. Sure, our core business is based around the agency model, but we do more than that. We build products, and Industrial Access, YArooms, GoDev & ialoc are fine examples of what we stand for.

We stumbled upon UPRISE completely by accident and considering our rebranding and repositioning last spring, we decided to go there to meet as many startup founders as possible. And see if we can help them out.

Highside is one of the few companies that piqued our interest - they’re nice people and, judging by their work philosophy, their business acumen is quite impressive.

Highside began as a VoIP service provider over 15 years ago and have since evolved their productts to the point tha they offer a Voice API. You know those smart bank-bots that take your support requests and direct you to the right representative? Those could be built with their API.

( While we’re on the subject, check out the English Language API. It’s fun. )

But enough chit-chat, let’s run through our interview with the Highside’s management team - Pieter Koenis and Martin Savelsberg.


“If you work with smaller companies you get better feedback and you find new ways of improving and developing your product.”


How did the company begin?

Martin : When I started the company 15 years ago, I was importing VoIP devices. B2b equipment.


What about the structure of the company / team ?

Highside: We are a small company - 4 people in our NL offices and 2 people working out of office as developers. You need to trust people if you manage a distributed team. You can’t control people. We’ve been working with our external developers for 10 years. We tried to speed up development by adding more people, but that didn’t pan out… We identified 1-2 people who seemed enthusiastic in working with us, but 2 days before starting the project, they bailed on us - they had lost interest.


A Voice API? How did you get to that crazy idea?

H: The Voice API was born 3 years ago, at the Mobile World Congress in Barcelona. I was there as a visitor with one of the engineers. And it dawned on us that It is now easier for small companies to connect directly with mobile carriers. VoIP telephony margins are small, despite the business volume. Price is rarely the most important thing - bringing new stuff to the table is what matters most, which is why we started the voice API.


How much time did it take you to bring your idea to fruition ?

H: It took us about 1, 1 ½ years.


Have you validated the idea before building it ?

H: We got a request for such a thing from a large company and since then we’ve been working on making the platform more user-friendly. The API without a GUI is better suited for big companies, but we wanted to cater to SMEs’ needs as well.


How much time did it take to break even?

H: We’re still investing in it. We have some clients on that specific service, but it is not profitable yet. We need to drive more customers to the GUI platform, but we don’t have a sales target.


Are you constantly developing it ?

H: Yes. We hope to get hold of a couple more developers quickly. The platform is stable, but there are a lot of things we’re looking to build. We’re rolling updates every 2-3 weeks.


What are your plans for this year ?

H: We have a master plan, but we take our users’ feedback into consideration while building the product.

We know what the end-product should look like, but what gets developed first is constantly under debate. At the moment, we’re looking to integrate our API with platforms like FB Messenger & IFTT - doing things like sending notifications via SMS or Facebook ID (the latter being cheaper, so the notifications would first be sent to FB; when that fails, SMS; when that fails, call). Our end-goal is to have a message-API, a centralised place for sending and receiving messages (like Postmark for emails).

We don’t want to be the next Mailchimp, but we are aiming to deliver the same degree of usability and to integrate with them.


Have you thought of integrating with services like Zendesk?

H: I’ve heard of them, but never thought of including them in our integrations list. We see ourselves like a Twilio with a GUI and no coding required. You shouldn’t be a programmer to use our API.


Have you had any hacking attempts on your infrastructure?

H: People are regularly trying to hack our system, looking for open ports and such.


If you could give your 2010-self a piece of advice, what would that be?

H: Focus! It’s really difficult to do, but you need to dream and clarify things at the same time. We were doing too many things at the same time - you know how it is: one client asks for something custom-made, another one as well and you end up losing focus and getting sidetracked by overly specific client requests.


Closing words: If you were to think about a creative way of using your voice api, what would that be? What unusual service/ app would you build using your api?

H: It can be used as a one-time authentication in a platform, via voice commands. I’d like to see it being used the following way: I want to buy something online, I place an order and the shop sends me an sms asking me when I’d like to receive the package and I can converse in natural language with the system (send it tomorrow, send it after 12:00 and so on). At the moment, we can do scripted conversations (the app understands keywords).