4. (Perhaps) Don’t build it

This standard is linked to the following Principle for Digital Development: Reuse and Improve.

Most of the time, building something new is a mistake. Think: there is no need to ‘re-code’ the wheel.

The three options for any project:

  1. Use an off-the-shelf platform
  2. Build on top of or use existing open-source platforms and Digital Public Goods (DPGs) 
  3. Build something new

The quickest and most effective option is to buy an ‘off-the-shelf’ platform using the SaaS (Software as a Service) model. You will not be able to change the code, but you will be able to change settings to customize the platform. There will be no need to hire a technology team, and you can focus on roll-out and implementation.  For example, Microsoft Teams you use every day is an off-the-shelf platform. 

Beware: it is not an ‘off-the-shelf product’ if it takes longer than 2-3 days to set up and you have to speak to a human being before buying. 

Building on top of existing platforms will generally mean taking open-source components and editing the code to meet your requirements. The benefits here are that you’ll be leveraging years of other people’s experience and effort.  You can have many small platforms working together instead of one giant system. You can launch products 5-10 times faster this way than building something completely new. Many Digital Public Goods (DPGs) are built in a modular way and will allow you to do this. 

The third option is to build something new.  Only consider this for smaller products or extensive opportunities where the solutions can be reused globally. The reason is that successfully making something new is incredibly difficult, time-consuming and expensive. To create a mature, scalable product or service, you’ll need to ensure a team and budget for a minimum of three years. 

The critical question to ask yourself is:

Can you get 70-90 percent of what you need for 10-20 percent of the effort? If so, don’t build anything but try options #1 and #2 above. 

There are exceptions to this, especially when building on/for government partners, then Open Source and DPGs should be favoured over off-the-shelf SaaS solutions, because of the nature of government data. But in general, the rule applies.
UNDP can work with government partners on identifying suitable existing DPGs for their problems (see the link to the Registry), and identifying if DPGs are a good fit. While they can be quicker to implement and give governments more control over their solutions, they might face limitations with regard to maintenance and customer support, depending on the maturity. A key advantage of using open source/DPGs is that they prevent vendor lock-in.  Chief Digital Office can support in conducting such assessments with country offices.

 

Do:

  • Check if other people in UNDP have developed or procured a similar tool. Ask the UNDP Digital Advocates Network if you are unsure
  • Check the Digital X Solutions CatalogueDIAL catalogue and the Digital Public Goods Registry for similar solutions 
  • Check if ready-made commercial solutions are available
  • Review all three options and report on your findings
  • Create a Lean Digital Impact Canvas (see here)
  • Consult with technical experts

Don't:

  • Build it, without first considering other options
  • Get distracted by shiny technology
  • Default to building from scratch. This should be the last consideration

Tools

 

UN Resources:

 

To Watch

 

To Read

 

Case Studies

  • TrackAI: Stopping blindness in its tracks Read about how Dr. Pueyo and the team behind the TrackAI phone and tablet app have effectively removed barriers to early diagnosis of eye disease in children, even in the world's most remote communities that don't have access to high-end medical resources.