Your humble servant: UI design, narrative point-of-view and the corporate voice

5 min read

I've been chuntering on about the application of narrative theory to design for long enough that I'm kind of embarassed not to have thought of looking for it in something as everyday as the menu labels in UIs... but better late than never, eh?

This guy is interested in how the labels frame the user's experience:

By using “my” in an interface, it implies that the product is an extension of the user. It’s as if the product is labeling things on behalf of the user. “My” feels personal. It feels like you can customize and control it.

By that logic, “my” might be more appropriate when you want to emphasize privacy, personalization, or ownership.


By using “your” in an interface, it implies that the product is talking with you. It’s almost as if the product is your personal assistant, helping you get something done. “Here’s your music. Here are your orders.”

By that logic, “your” might be more appropriate when you want your product to sound conversational—like it’s walking you through some task. 

As well as personifying the device or app, the second-person POV (where the labels say "your") normalises the presence within the relationship of a narrator who is not the user: it's not just you and your files any more, but you and your files and the implied agency of the personified app. Much has been written already about the way in which the more advanced versions of these personae (Siri, Alexa and friends) have defaults that problematically frame that agency as female, but there's a broader implication as well, in that this personification encourages the conceptualisation of the app not as a tool (which you use to achieve a thing), but as a servant (which you command to achieve a thing on your behalf).

This fits well with the emergent program among tech companies to instrumentalise Clarke's Third Law as a marketing strategy: even a well-made tool lacks the gosh-wow magic of a silicon servant at one's verbal beck and call. And that's a subtly aspirational reframing, a gesture -- largely illusory, but still very powerful -- toward the same distinction to be found between having a well-appointed kitchen and having a chef on retainer, or between having one's own library and having one's own librarian.

By using “we,” “our,” or “us,” they’re actually adding a third participant into the mix — the people behind the product. It suggests that there are real human beings doing the work, not just some mindless machine.


On the other hand, if your product is an automated tool like Google’s search engine, “we” can feel misleading because there aren’t human beings processing your search. In fact, Google’s UI writing guidelines recommend not saying “we” for most things in their interface.

This is where things start getting a bit weird, because outside of hardcore postmodernist work, you don't often get this sort of corporate third-person narrator cropping up in literature. But we're in a weird period regarding corporate identities in general: in some legal and political senses, corporations really are people -- or at least they are acquiring suites of permissible agency that enable them to act and speak on the same level as people. But the corporate voice is inherently problematic: in its implication of unity (or at least consensus), and in its obfuscation of responsibility. The corporate voice isn't quite the passive voice -- y'know, our old friend "mistakes were made" -- but it gets close enough to do useful work of a similar nature.

By way of example, consider the ways in which some religious organisations narrate their culpability (or lack thereof) in abuse scandals: the refusal to name names or deal in specifics, the diffusion of responsibility, the insistence on the organisation's right to manage its internal affairs privately. The corporate voice is not necessarily duplicitous, but through its conflation of an unknown number of voices into a single authoritative narrator, it retains great scope for rhetorical trickery. That said, repeated and high-profile misuses appear to be encouraging a sort of cultural immunity response -- which, I'd argue, is one reason for the ongoing decline of trust in party political organisations, for whom the corporate voice has always been a crucial rhetorical device: who is this "we", exactly? And would that be the same "we" that lied the last time round? The corporate voice relies on a sense of continuity for its authority, but continuity in a networked world means an ever-growing snail-trail of screw-ups and deceits that are harder to hide away or gloss over; the corporate voice may be powerful, but it comes with risks.

As such, I find it noteworthy that Google's style guide seems to want to make a strict delineation between Google-the-org and Google-the-products. To use an industry-appropriate metaphor, that's a narrative firewall designed to prevent bad opinion of the products being reflected directly onto the org, a deniability mechanism: to criticise the algorithm is not to criticise the company.


In the golden era of British railways, the rail companies -- old masters of the corporate voice -- insisted on distinctive pseudo-military uniforms for their employees, who were never referred to as employees, but as servants. This distinction served largely to defray responsibility for accidents away from the organisation and onto the individual or individuals directly involved: one could no more blame the board of directors for an accident caused by one of their shunters, so the argument went, than one could blame the lord of the manor for a murder commited by his groundskeeper.