Inside: Network for Learning (N4L)

It may have had $211 million to work with, but N4L was very much a start-up when it set out to connect 2500 schools to fast internet. It worked – where sister education tech project Novopay failed dismally. Here's what went right.

IN 2013, THE Government announced it was spending $211 million connecting New Zealand’s 2500 or so schools to safe, fast broadband internet. It was one of the biggest Ministry of Education technology projects ever and it was coming off the back of one of the biggest education technology flops ever – Novopay.

N4L’s remit was to build a managed network for 800,000 users and make it work. At the same time, it was to figure out a way to make online content more accessible. Government was – not surprisingly – nervous. Two years later, the Novopay teacher pay system is back in the Government fold after chewing through an additional $45 million of taxpayer money – on top of the original budget.

At the same time, N4L has connected half a million users in almost 1500 schools. It is a year ahead of schedule, and within budget.

It has also developed a world-leading education portal, Pond, with 6000 teachers signed up less than a year after launch. Its partners range from some of the biggest telcos in the land, to one-man-band tech experts working with tiny rural schools.

So how did N4L succeed, when Novopay failed so spectacularly?

Idealog editor Nikki Mandow talked to three of N4L’s senior team – CEO John Hanna ( pictured above, right), marketing expert Andy Schick (left) and Chris South, the brain behind Pond. She asked them: what the hell went right? If you could take only one key lesson out of this whole experience, what would it be? You have to think about the experience for the end user – from the beginning, and constantly. Other projects might begin with a brief to build this technology stack or design that piece of software. But all our conversations began with what was the experience going to be for the school – the teachers, the students and the administrators. We consulted and collaborated with users, and we are still doing it.

Early on, the executive team spent hours in schools understanding as much as possible the type of experience they needed. Each of us sat in 30 or 40 classrooms watching the teacher and the class happening.

Even when it got busy and stressful, when we were negotiating large contracts with Spark, for example, the leadership team was encouraged to spend as much time in schools as we could. What did the schools tell you? They asked us why we couldn’t deliver commodity internet. Like electricity – you flick a switch and it works, period. In the classroom, we saw if the technology didn’t work, teachers might give it 30 seconds, and then they moved to plan B. The internet had to be consistent and reliable, otherwise they wouldn’t use it. But aren’t you all techies? Surely geeks and nerds don’t do all that fuzzy collaboration, end-user stuff ? We realised early on we were a technology company, not an education company. But we also realised the education sector was cynical about Government-run tech projects. We had to prove ourselves.

We needed to employ highly competent individuals who never had to say they were competent. We didn’t want hard, knock-the door- down technical people, because that wouldn’t work for schools. The sales people You never rush. And you never compromise when you are looking for the person you need. In the development space, it can take three to four months of constant searching to find the right person with experience, technical capability and a good head space. You have to keep the faith with the organisation’s value system. We often find we employ people who aren’t looking for jobs, so sometimes it can take a long time to convince them, and extract them from where they are working. Were you ever scared that everything was going horribly wrong? Yes, but we called it being risk aware – and note, that’s different from being risk averse. Being risk aware means understanding what could prevent you being successful, so you can mitigate against it. We always had a Plan B, C and D and we made sure Plan A didn’t preclude the other plans. Did things go horribly wrong? Not horribly, but several times we had to admit we were wrong, and choose another course. For example, we changed the way we designed Pond several times before we got it right, and its development is ongoing. It’s about failing fast and then pivoting. Plans B, C and D help. Novopay was roundly hated. Did you cop any flak? Not really. Rather the opposite. We learnt the key thing was to take our stakeholders on the journey. They didn’t always get it completely (and that didn’t stop us going on our journey) but we tried to make sure our messaging was clear, and we didn’t stop talking. We communicated; we thought we were overcommunicating, but then we realised it is incredibly difficult to annoy a stakeholder with over- communication. As a rule of thumb, you need to tell stakeholder about what you are doing and what you are thinking four times as many times as you think is necessary.