Improving systems

It's often better to improve a system than develop one brilliant thing:

Making systems better is not particularly sexy work. It tends to be incremental, slow and messy, taking knotty problems and carefully unknitting them. In the time it takes to make a widely-used system very slightly better, you could probably make half a dozen gorgeous one-off pieces of journalism that the world would love.

But if you make the system better, you potentially make lots of people's jobs easier, or you save dozens of person-hours in a month, or you make hundreds of pieces of journalism work slightly more effectively. It's not flashy, and probably most people won't even be aware of what you’ve done. Most organisations need people doing both, because without the brilliant beautiful one-off pieces, how would you know what the system needs to be able to do in the long run? But people who do the flashy things are plentiful, and people willing and able to graft on the stuff that just incrementally makes things better are in sadly short supply.

Mary Hamilton, 13 things I learned from six years at the Guardian