Interesting Failures - Group Calling
Another in an occasional series of blog posts where I discuss products I've worked on which failed.
It boiled down to this. Was there a way we could make normal people want to be on a conference call with their family and friends?
Let me explain. ARPU - Average Revenue Per User - that's all mobile network operators care about. When it's up, the money flows like wine. When it's down... the drought makes people do crazy things.
Voice revenue had been falling for years. Skype and WhatsApp mean you can call the world for free. The only people you call on the phone line are your grandparents, the pizza company, and your employer's conference call system.
If social-conference-calling became a thing, ARPU would go up.
Here's how the system we built worked.
- Install an app (duh!)
- Select phone numbers from your address book. Say your mum, dad, and brother.
- Hit "Go!"
- The app dials the number of our conference call system.
- As soon as you connect, the system dials your mum, dad, and brother. But - and here's the magic - it uses your Caller ID, so they think you're calling them directly.
- When your mum picks up, she's put into a conference with you. If your dad doesn't pick up, the system cuts off rather than playing you his voicemail message.
- You brother misses the call, but dials you right back. Here's the second bit of the magic - the network redirects his call into the conference system.
You get it, right? We came up with a dozen user stories (5-aside football team, curry night, wanting a lift home, granny's birthday), and built demos of the technology.
It worked! The network integration was tricky, but possible. There were a few rough edges, but it was good enough to try out.
Customers... were split. Some people got it straight away. It was perfect for their needs - kids calling separated parents, managers wanting an ad-hoc team call, that sort of thing.
Other people were ambivalent to it. No one like conference calls at work, why would friends and families enjoy them?
But the thing that killed it - what always kills products in the mobile industry - was how to charge for it.
Many years ago, when SMS cost 12p a go, the mobile networks launched location based services. For 12p you could have your location sent to your phone. Then Google came along and fucked it up for everyone by making location services free.
That's OK, because networks could still charge £2.50 for a premium rate SMS to deliver a game - and take 50% of the revenue. Then Apple came along and fucked it up for everyone by making games free.
Still, that SMS revenue, eh? Oh, look, WhatsApp has fucked everything up by making messaging free.
You see the pattern, right? People like free stuff.
How would you charge for this? Here's a quick run-down of every charging scheme that was rejected.
- FREE! Not quite as dumb as it sounds - most people already have unlimited calling. This gives them a reason to love their mobile network.
- Just as a normal call.
- Pro-rata. You call three people, you get charged for three calls.
- In-app billing for a subscription to the service - perhaps with a set bundle of minutes.
- Ever increasing levels of complexity which would have been a nightmare to bill for and impossible to understand.
Unsurprisingly, the last one was the most popular with the network. Complex billing leads to misunderstandings, which leads to increased ARPU. Until people get fed up and churn to a different provider.
The project spluttered out, and never took off. Maybe it would have flopped anyway, we'll never know.
The central challenge to the mobile phone industry is "Why do I need a mobile network?"
Skype and WhatsApp provide calls and messaging at a higher quality and lower price. WiFi is free and plentiful. Everything a network can do, someone else can do better, faster, and cheaper.
At the moment, a mobile network is just a company which gives you a payment plan on a £1,000 smartphone.
Mobile networks used to be afraid of being a "dumb pipe" - that moment has long since passed. They're now embracing being a loan company which sponsors a sports team.
Alex Gibson says:
If only the MNOs even cared about the quality of that 'dumb pipe'! It would be amazing if OFCOM could arrange for the MNOs to feel the slightest impetus to complete the job of achieving comprehensive mobile coverage. At home in suburban Reading, on EE (but also Three), I have extremely sketchy mobile coverage - it's not even reliably bad - I get 3 bars depending on atmospheric conditions, but often during a call (almost certainly during a long call) the signal will drop to 1, then 0 and I lose the call. At work in Green Park, forget mobile signal from anyoneunless your specific network has installed a booster - which they won't sell you as a home customer. I get 5 bars there from the EE femtocell, but if I dare to wander out of the building, even if the outdoor coverage is OK, I lose the call in the handoff to the cell tower. On the Paddington to Reading line, one of the busiest (and most expensive) commuter lines, there are 2 distinct dead spots that have remained present for at least the last decade. I've experienced these on EE and Vodafone. Can we not get this gap plugged? Mobile phone companies should be rated by calls lost to poor connection - not notional '98% coverage of populated areas' based on outdated census information and an overly optimistic view of cell tower strength and reliability... Definitely no more signal bandwidth should be allocated to MNOs until they can prove they actually use what they have!