What can’t Google do? And why does it matter?
The subject came up while I was talking to a journalism class the other day, discussing an upcoming site that would collate and curate information about a particular topic, bringing in context, related information, documents, and background. There wouldn’t be a huge amount of original content, with most of the focus being on collecting and organizing information from elsewhere.
To which one student asked: Can’t I get all of that from Google?
And it’s true, you can. To be sure, there’s already a huge amount of value in curating, verifying and organizing information that’s easily available via a Google search. Just sifting through the flood of links that come back on any search is massive public service. But it does raise the question: How much different do you have to be from a Google search to really add value? Do you have to add a lot more original content – and even if you do, isn’t that content then just available on Google as well?
Or, to put it another way, what is it that Google can’t do that a site – even one that doesn’t create its own content – can do?
And the answer is: A fair amount. (At least so far, until a bunch of PhDs at Google put their mind to it.) Google can’t really add or subtract, for example, so if you store information in a structured form, ala Politifact’s fact checks, you can create a page that summarizes and counts, for example, how many times Donald Trump has uttered falsehoods. Google can’t do that. (Or rather, a Google search can’t return that number).
Google doesn’t do timelines all that well, either, since it generally prioritizes the most recent story on any given topic. But if you want to track an event or an issue over time, that’s not to helpful. So by ensuring that dates and summaries are part of the information structure of a site, it’s relatively easy to generate chronologies and calendars around events, as Homicide Watch DC once did, (and its many successors still do).
And then there’s explainers and glossaries and other pieces of context that work well in a card structure, ala Vox, that allow sites to bring deeper levels of understanding to particular issues without needing to rewrite loads of background every time a story runs. (And, if the late, lamented Circa’s model of data is adopted, it allows for a cleaner, faster read for regular visitors to a site, by cutting back on the background they already know.) (Connected China, too, used a common data structure to power all its different visualizations and content. Had to get that Connected China reference in there.)
In other words, there’s a lot that Google doesn’t do – or doesn’t do well – even with publicly available stories and content, and that good information design (or information architecture) can allow you to bring much more value to users than simply the ability to find stories amid the mass of information out there. What it takes is a focus away from the “now” of finding the latest information, and an examination of the kinds of needs users might have over a longer period of time.
And structure. You gotta have structure. (Or this wouldn’t be a blog about structured journalism.)