to Home posted 19/04/2011 in Projects about 11 minutes read


A couple of months into brainstorming about bringing WiFi internet access to cabs in Ljubljana (thats a separate project), I ran into Icomera, a Swedish/UK company specialized in manufacturing and deploying wireless routers for the transport industry. The then Channel Sales Director introduced us to the magic Icomera manufactures and a couple of weeks later the demo unit of their M200 was on my desk.

Icomera Moovbox M200

Lets make a step back. The idea of surfing the web while traveling is as old as the internet itself. The realization of that idea is a different story: wireless networks had to reach a “useful” bandwidth capacity stage (EDGE – UMTS – WiMax), routers had to sing and dance simultaneously in order to catch the signal of multiple base stations and keep all users seamlessly connected to the web and the cost of all that had to be, well, reasonable. Because the hardware was not there yet, operators introduced USB dongles (I’m ignoring the PCMCIA thing) as an easy way to connect your laptop to the internet without wires. That was a sound business approach. But consumers did not take the bait. Laptops are WiFi capable and whenever we go somewhere we search for WiFi access points. We don’t want to buy USB gadgets in order to occasionally access the internet. Starbucks identified access points as a way to make customers stay longer and in the end order more. And it worked.

WiFi has one extreme advantage: it is widespread. Default. Every gadget is WiFi enabled, no upgrades or software required. Why USB dongles?

Moovbox in buses

With Icomera we immediately saw that the technology was there. They had the experience (Swedish railways & Stagecoach UK, just to name a few) and vast knowledge and understanding in what they are doing. In one box we had a WiFi access point, 2× ExpressCard slots for HSUPA modems, external antennas, GPS and remote management. If somebody asked us what would we want in a box, we could not think of more. The deployed box would be connected to our cloud via VPN, enabling us to filter bandwidth heavy traffic (YouTube for example) in order to keep the service usable for surfing. GPS data would be used to show the operator a fleet map (just as fleet managers do), or connect to existing fleet management software. We would use 2 separate network operators, one known for big coverage and good quality and the other known for very good speeds in urban areas. In our tests, we reached a stable 2/512 and 6/2 throughput. Wow!

The cloud mentioned before was actually a set of servers separated for accounting (who can and who can’t use the service) and firewalls. The entire user traffic would pass through the cloud out to the internet. The service would initially be free, with a target of 1€/24h. Ideally it would be financed from increased ticket sales, but the business model was not set on covering costs by charging internet access.

I have to explain this because the normal approach would be to just transfer all the running costs to the customers. We don’t do normal approaches. Normal approaches are dull, boring, uncreative and non challenging. The whole point of an idea is to make something very usual in an unusually good way. Instead we approached the problem by drawing a “interest chart” (as we did in the previous project), or plainly listing who could benefit from this service. This way we see who is connected to the service, and the whole revenue stream can be shifted in one way or the other.

Our target shifted from the customer to the operator (owner of the vehicle), network provider and location based advertising. The operator had to pay something, but should not take the whole punch. The network provider should at least waive the monthly fee for using his services in exchange for “Powered by” promotion. The clever advertising should bring in the cash. By clever I mean very a focused mix of location based and general advertising.

In the entire plan there is one catch. We had no clue of how the mobile network would work. Our tests showed very good coverage, but providing a steady service is a different thing. This was the only point that we decided not to bother with. You may say it is the most important one, but besides putting a bigger antenna in the vehicle there is little we could do about it. By using the “beta” cover we would work with the network operators in order to solve problems and coverage issues. We are aware of the problem, expect it would jump out, and prepared to deal with it.

The bus business

Stagecoach WiFi bus

The idea was to offer the service to bus transport companies (regional ones) in exchange for a monthly fee and a 2 years contract. To give you perspective: the monthly fee was set to a 200€/month/bus mark. That fee included everything (hardware, connectivity, support, deployment). 200€ per month. Thats about 5 times what I currently pay to my internet service provider for 10/10 internet and IPTV at home. With 200€/month, we as the service provider would break even after 2 years. If we take an average fare price and play with numbers, we could find out what increase in passenger numbers is needed in order to cover that cost.

The location based advertising would be based mainly on messages like “End of the ride? There’s a McDonald’s just around the corner. 10% off if you use the coupon BUSWIFI”. Simple, easy and targeted. The delivery method would be via classic captive portals or even follow up SMS messages. As the user joined the WiFi network he was required to enter his phone number in order to use the service. The phone number was used instead of email because we wanted a simple and fast way to have some kind of registration. Spamming users was out of question, he was “ours” only while using our services.

The plan was set, everything prepared, and we started throwing the idea around.

The feeling when the responses came in was a mix of disappointment and WTF moments. The #1 reason was: “We don’t need this”. After a couple of cold showers, the answer was actually very clear. Bus companies in Slovenia operate under concessions by the state, not by free-market competition. Basically they don’t care what service they offer as they are the only ones offering it. The buses are not upgraded because passenger feel better in them, but because the running costs were lower. Reasoning that bus usage was in decline was not an unfavorable argument as people are expected to be more inclined in using public transport if they don’t have to stare at the backhead of the person in front of them for the entire trip. We do admit that the idea landed just as “cost savings” became the buzzword managers used as an excuse to squeeze something more out of the companies without increasing expenses.

Wrong timing, wrong country, wrong mindset. The idea was abandoned.

The rail business

Siemens Desiro

After a couple of months, a door appeared on the horizon. A mobile operator, client of a marketing agency we work with, rented an entire train for advertising. The rail operator in Slovenia is a wreck. The company is losing money everyday, it is at the center of public criticism and perceived as dinosaur on his last journey. We had only one shot at getting in and we had to avoid a public tender at all costs. Public tenders have nothing to do with sanity or quality. They are a big table of poker where the big shots have so much money they just play “all in” on blinds until you are left without everything and out of the game. We are way too small to play at that table. We had to turn the business model upside down so that the rail company paid nothing. Zero. Nada. Niente. Only that model would work and avoid a tender.

The idea was to expand the WiFi service with digital signage. The added service would then cover the entire cost of the operation. At least thats what we hoped for :)

The initial meetings went extremely well. The railway company gave us full support in developing the solution, and the mobile operator was excited to be part of such project. I’ll be honest, we were almost 100% sure that the engineers that take care of the trains would say “no freeking way you’re going to put anything in our trains”. Boy were we wrong. We received so much support from them that they were actively working with us on solving problems. And we did have problems that were showstoppers. The hardware we picked (including the rail-certified Icomera router) was high quality stuff. Huber+Suhner antennas and cables, custom built fanless PCs for LCDs… We knew we were investing in our showcase.

After 3 months of researching, testing, observing – we were ready. The deployment was drawn, green light from the engineering team, everything was ready for installation. The entire cost of deployment (research, programming, deployment) was estimated at a 30k€ mark. That amount included both labor and hardware, pushing the cost up for the first train. Now we just had to agree on the business part with the rail company. And that proved to be the death trap.

Apart from WiFi access via 2 access points in 1 train, we would also install 3 LCDs (1 double sided, 1 single sided). The LCDs would primarily replace the buggy displays showing passengers time/date/next station and provide many more options for increasing travel quality. From the route map, to expected arrival times, weather at destination, map with current location and general info to location based advertising on stations and general advertising during travel.

The deal with the railway company was simple: they allow us to deploy the service in 1 train (a single joint Siemens Desiro EMU, as in the picture above) for a duration of 2 years. We would buy all the hardware and their engineering teams would install it. We would cover all costs from running the project and we sell advertising space on the LCDs and introduce paid WiFi internet access (1€/24h) to cover the costs. After 2 years the on-train hardware becomes ownership of the rail company, and they are free to continue to use our services or pick a different service provider. The 1 train deal was presented as a test to prove such service can be deployed without additional costs to the operator. In short: if we could prove the business model works, we would multiply it and put more to the table.

The idea was torpedoed by the marketing department on the last meeting in late October 2010, as they expected some form of payment from us in exchange to use the train. The official explanation was that they have to use all possible marketing opportunities in order to increase revenues. We just couldn’t afford paying any fees as this would lower our chances of breaking even, in the end making all attempts to expand the service unrealistic.

3 months were left dead on the rail track. We abandoned the project (well, there aren’t any other rail operators in Slovenia to turn to) and decided to move on. Wrong timing, wrong country, wrong mindset.

The cold reality

InterCity Slovenija

The picture above shows the ICS fast train that travels between Ljubljana and Maribor, two biggest cities in Slovenia, some 120km apart. The car ride takes about 1h 10min, while this train travels just short of 2h. The track is simply unfavorable for faster service. There is no train that would shorten that travel, the only way this train will win over car is to offer more than a car. The strategy of Slovenske zeleznice (Slovenian railways) clearly states that onboard internet access is a key point.

I’m betting that in the next couple of years, SZ will pay millions of euros to one of the big companies (preferably Siemens) to install and maintain onboard WiFi internet access services.

I presented this story in a couple of chats with various CEOs and entrepreneurs and the vast majority told me the main reason for failure was not offering any form of payment to the ones that were deciding in the rail company. Bribery in short. I believe them.

Maybe we did not work enough on making sure the companies we approached understood the importance of improving their services. The vision we had, of a small country that marked her own spot on the map as being “one of the first” was indeed a far fetch. The vision of showing everyone that giant leaps are possible even without throwing millions at big telecom companies was indeed unrealistic. I don’t mind my idea being turned down, but I feel disappointed and even angry when the mindset of people that temporary hold key positions is wide enough only to fit their own horizon instead of being as wide as possible to benefit everyone.

~ the end ~

to Home posted 19/04/2011 in Projects


blog comments powered by Disqus

More reading