I recently spoke to somebody who wanted copy for their website. And the good news, they told me, was that the site design was already done and dusted. On the home page, there were three square boxes in a row, spaced evenly. Underneath, there was a box that took up the entire width of the page, and under that were another two boxes. The proposed design looked attractive, making good use of white space and complementary colours. It was when I asked what went in the boxes that we ran into trouble. I’d made the assumption that three boxes meant three distinct offerings. Or three target audiences. Or three offers. But they didn’t, any more than the one box underneath was destined for anything specific. Or the two boxes below that. In fact, the whole design was chosen on the basis that it looked pleasing, its boxes filled with the ubiquitous mock Latin (Lorem ipsum etc.). But when we actually looked at the copy that was needed, it didn’t fit neatly into the boxes. Or neatly on the page, for that matter. So we did the only thing possible: turned the approach on its head, and started with the copy. For at the end of the day, you have a story to tell, and an audience to engage. And the design should support, not dictate, the way that story is told. I’m not saying copy trumps design. The two have to work hand in hand, so there’s not a disconnect between what you’re saying and the way it’s presented. I was reminded of this balancing act as I watched a TED video last week. Margaret Gould Stewart, Facebook’s director of product design (and ex-YouTuber) talks about the huge impact that little design changes can make. Like changing the Like button on Facebook. That tiny graphic took the lead designer 280 hours or work (that’s seven weeks at 40 hours a week) to redesign. She also talks about the Facebook photo take-down request that failed to engage users. Until, that is, the designers tweaked it to include the reason for the request, and how the photo made the requester feel (sad, angry, embarrassed and so on). From a copy point of view, the take-down story is fascinating, proving that context is everything. If people understand why you’re asking for something, and what a difference it will make, they’re much more likely to comply. In Facebook’s case, usage of the feature jumped from 20% to 60% of those wanting photos (usually embarrassing ones) taken down. And research showed that 90% of people who’d posted photos wanted to know if and how they’d upset people. Gould Stewart also talks about knowing who you’re designing for, which in Facebook’s case means a huge number of users who don’t have access to cutting-edge hardware or fast internet. The exact same approach applies to copy: if you don’t know who you’re writing for, you’ll never come up with copy that connects with your target audience. How giant websites design for you (and a billion others, too) has lots of insights into how little things can make a big difference, and may just get you thinking – as it did me – about the importance of getting them right. And of really thinking about who’s out there, and what matters to them. [If you’re reading this in an email, click here to view the talk on TED.com]