by Justin Anderson, Content Editor at GoodPractice
It was us, about a year ago.
With our new Dynamic version of the toolkit then on the horizon, which has a much more visual style than our Classic site, our director, Owen, felt it was the perfect time to think about adding more images.
We decided every article should have at least one unique image, which would be a nice visual metaphor for the content. It would appear near the top of the article, and alongside it in search results.
Great idea, right? Then we thought about it.
There are over 3000 pieces of content in our toolkits.
Hey, you have to start somewhere, right? So we began looking for ways to source large numbers of images. We quickly discovered just how much good quality is available online under Creative Commons licenses. It’s a lot. It turns out photographers can be very generous with their work, as long as you credit them for it – which seems reasonable!
So as a test run, we allocated a selection of articles throughout the content team and set about finding pictures for them. Google Advanced Image Search was our ‘go to’ source to begin with. It had a nice function that allowed us to filter for licence, size, colour etc.
Often, we found this led us to two particular sites that were good sources of content: Flickr and Wikimedia. So we also started searching those sites directly. And another site that proved helpful for images of specific locations or buildings etc. was Geograph.
After a process of deciding what we collectively liked and didn’t like, under Owen’s guidance, there came an interesting discussion amongst the group: how much flesh is too much? You see, we occasionally found a great image that we just weren’t sure wouldn’t cause offence to someone, due to the general state of undress of one of the subjects. One image of a girl writing caused some debate, as we couldn’t decide whether her bare legs were, well, too risqué. To be fair, it’s a pretty short skirt.
In the end, we decided it was too good an image and too perfect for the article not to use, so bare-legged writer girl got to stay. And nobody complained. So were we being overly cautious? Maybe. Judge for yourself.
The next image that brought the debate back around was a pair of runners. A great shot of a man and woman jogging on a road. The only problem? The man was shirtless. Again, we discussed it and this time we decided to leave the image out. This led to the establishment of our general rule on where the flesh line was to be drawn:
No nipples. (Don’t tell Rumer Willis!)
It’s a general rule and we’ve broken it a few times, notably for a great image of a diver wearing pretty much exactly what you’d expect a diver to be wearing. Context and all that. But we do still have to consider the wide-ranging sensibilities of over a million registered users…!
Once we had sourced, edited and uploaded a selection of articles ourselves, it became clear that the four of us couldn’t do the whole job and still continue to do any other work at all. Instead, we opted to outsource the job to a selected group of freelancers, managed by my colleague Lee and quality controlled by her and me.
I spent several days creating a brief, explaining what we wanted, giving examples of good and bad image choices, and creating a wiki for how to go about searching, resizing etc. (I’ll talk more about this in the next Inside GoodPractice piece.)
Then Flickr immediately changed its site and I had to update the wiki. Sigh. (It’s changed again since then but, thankfully, our freelancers were clued up enough by then to work it out for themselves.)
Lee organised the recruitment of three candidates and set about allocating selections of work out to them in databases. The potential for the same article to be sent out to more than one person was huge, as was mixing up the unique ID numbers we needed to have for every article. I’m frankly convinced it was a work of sorcery that she managed to avoid any significant mix ups over the course of the year!
One of the biggest spanners in the works, however, was when Google Image Search did something utterly odd. For lack of a better description, it broke. We suspected for a little while that it was pointing us to images that weren’t really available under Creative Commons licenses, but one search finally confirmed it. I found an ideal picture for an article, only to discover it was on the Daily Mail website. We decided that, on balance, it was unlikely that the Mail were offering their images for free, and Google got binned.
Much of our searching now is done directly on Flickr, or via a useful site called CC Search. But even that comes with dangers! If you’ve ever used Flickr, you may be aware that there are some images in there that make our ‘no nipples’ rule seem, frankly, Victorian.
Accidentally go hunting without their ‘safe search’ filter and you will be astounded at the images seemingly innocent search terms will return. (Stef and I regularly wave each other over with barely concealed mirth/horror to see what’s come up in one of our searches! Who knew there was such a thing as a naked bike ride?)
It took our freelancers a little while to get up to speed, but once they did, they were producing a regular flow for us to review, and Lee and I were perusing several hundred images a week. What was interesting was how often we independently had exactly the same problem with an image; we’d really come to a common understanding of what was a good image for us and what just didn’t work. We also had one strict rule from Owen: “no cheap-looking wooden table backgrounds!”
Another issue we had to get over, though, was the overuse of particular image themes for topics. (“No more butterflies for Change Management!” “No more chess pieces for Strategy!”) But then, as I said, with over 3000 images to source, and multiple articles on similar topics, the whole job was a big ask to begin with.
Once we’d got the bulk of the images identified, there was another logistical nightmare to negotiate. Each one had to be edited and then cropped to two different sizes. Again, Lee recruited some freelancers and briefed them on our needs. They rattled through the work with a minimum of fuss, which was another testament to her organisational skills.
And now, the end is in sight. We’ve nearly selected images for every article across all our toolkits. I’m personally really delighted with the results and very proud of what we’ve all achieved together. Plus, I’ve discovered how much I enjoy sourcing and working with images, which is a relatively new skill for me.
And I couldn’t pass round the kudos without mentioning our tech team – Jonny, Alistair and Kirsten – who worked out a script to insert all of the images and credits en masse automatically, so that the four of us didn’t have to go through them one at a time! (Thanks, guys!)
The toolkits are looking better and better. On top of the image project and our ongoing infographics development, we have plans to add even more visual and multimedia content in the coming year. I’d say I can’t wait to see it finished, but I know we’ll constantly be developing our products to keep them fresh and engaging.
If you’re interested in seeing just how gorgeous our new Dynamic toolkit looks, one of our lovely account managers can show you around in as little as 15 minutes. Why not book a site demo today, and see just how engaging and useful our content is? I’m sure you’ll agree our effort was worth it!
As a Content Editor at GoodPractice I get involved in creating all of the various learning and performance solutions we offer. That includes: researching, writing and editing articles for our toolkits; sourcing and editing images; developing infographics; scripting e-learning modules; producing bespoke client products; filming and editing video interviews; some (extremely basic!) graphic design and even the very odd spot of acting! (Or perhaps a spot of very odd acting…) My background is over 15 years of writing and editing for a wide variety of media and publications. Out of the office, my first novel, Carpet Diem, is due to be published in April.