There was a time, early in my career, where I exhibited Tom Cruise levels of enthusiasm for the use of mobile websites in development projects. Sitting in meetings trying to convince clients that they could use a mobile site to deliver vital information and support to thousands of people at a time would trigger sofa-bouncing displays as I tried to convince them that it was possible.
And it worked - funders took the plunge, access to the mobile web has continued to grow, and I went on to lead the design and launch of over 20 mobile sites aimed at connecting audiences across Africa for fantastic projects ranging from sexual and reproductive health and rights (SRHR), to gender equality, to advocacy.
As time went by, however, I started doing less sofa bouncing and more soul searching. Showing off a website I’d designed to some friends, I once dissolved into tears because I knew that the site itself had very little chance of making a real impact in an area I cared deeply about. Not because it wasn’t well designed, but because of a number of fail-factors which I had by then become savvy to.
1. Where’s your research?
This one might seem like a no-brainer, but so many projects seem not to recognise the need for research as part of the ICT component of their programme. I’ve often been asked to design a website whose success is directly tied into project indicators, only to discover that there is no evidence that the target audience has the regularity of access to a web-enabled phone required to make a lasting impact. There are a few things that should be answered in any proposal:
What is the web penetration in the country or region targeted?
What proportion of your audience actually owns or has regular access to an internet enabled phone?
Can people afford the data to use that phone frequently enough? In Rwanda, for example, there is 98% 2G penetration thanks to government efforts to develop the country’s ICT infrastructure, however, data and smart devices remain prohibitively expensive for most. 
If you’re targeting women or young people, what barriers to access and usage are present? For example, do your female audience have time in their day to sit and browse content? Do their fathers or husbands restrict their access to the device in the first place?
What level of digital literacy do your target audience have? Are they comfortable browsing multiple web pages and using social media?
Finally, do they already use the web/apps as a source for information and support? If not - then how are you going to fund an on-boarding process into the project?
Making sure these questions are answered is vital before making any sort of financial commitment to building something - and if they’re not, then finance the research first as part of an iterative approach.
2. Where’s your content?
So you’re confident that your target audience is online, digitally literate, and that the Patriarchy leaves them be once in awhile? Great! Now you need to factor in the capacity and skills to develop content that will have a chance of shifting attitudes and behaviours in a way that is suited to the mobile medium.
The best social change communication projects involve adapting expert information in a way that makes it accessible, relevant, and interesting to the target audience. For health behaviour change projects, for example, that might mean working with script writers to produce compelling storylines that weave in information, nudges, and relatable scenarios and behaviours. Needless to say, this task is not easy even with a captive, literate audience.
On the web however, particularly when targeting youth, you need to have the capacity to create content that is eye-catching enough to compete with all the content noise out there, quickly and easily digestible on a small screen, and that won’t eat up your user’s data. What’s more, if you do a good job and your users like what you have to offer, they will want to come back to you again and again.
This is a golden opportunity for ongoing, two-way communication with your audience, and between members of your audience, but you need to make sure you can deliver long-term. If you don’t think you can, you might want to ask yourself…
3. What’s your digital Theory of Change?
I’m not an expert on Theory of Change, but I like to illustrate its usefulness using this great cartoon:
Failing to unpick exactly how consuming content and doing activities on a digital channel contribute to realistic change is an interesting example of the way we see technology in itself as a silver bullet. Having some information on a phone and pointing people towards it does not social change make. We need to be thinking about exactly how people use their technology, and where that matches up with the learning pathways that could lead to changes in attitude or behaviour.
For example, on a project aiming to shift gender-based social norms in Nigeria, we experimented with fun, informative, interactive magazine style content. Were thousands of users reading this content? Yes! Did that mean success? A definite no - because we struggled to maintain the consistency of learning opportunities needed to shift deep seated attitudes. In other words, we had some users seeing multiple pieces of valuable content in a fairly sequential manner, and lots more seeing only one before disappearing.
That’s absolutely fine if our M&E framework defines success in terms of ‘reach’, but less so if we really want to prove our digital content is making a measurable difference. It would be great to see some projects designed to scientifically test the ideal mobile user behaviour patterns that lead to measurable impact in the same way that D.M.I have done with radio.
A good web design process involves making (informed) assumptions about archetypal users and how they will hear about, reach, and interact with the site - and how often. If we make sure our projects include an exercise taking these users and journeys, and seeing if and how they match up with type of impact we are trying to achieve, then we’ll avoid a lot of disappointed stakeholders down the line.
Ultimately, anticipating some of these challenges comes down to investing in intelligent and adaptive project design, not putting solutions first, and working with partners who have the guts to say ‘actually, mobile might not be right here’ even if it costs them business. If we start doing so, we won’t have so many users let down when a valuable online community is canned because it didn’t meet unrealistic expectations.
And for what it's worth, we’ll have fewer tears from programme designers, too.
 Rural coverage: strategies for sustainability, GSMA Country case studies, July 2015