Last changed 8 Apr 2000 ............... Length about 1,000 words (7,000 bytes).
This is a WWW document maintained by Steve Draper, installed at http://www.psy.gla.ac.uk/~steve/mkt.html.

Web site logical path: [www.psy.gla.ac.uk] [~steve] [this page]

Notes on a topic, to be discussed at: the workshop "Effective Teaching and Training in HCI". Many thanks to Jackie Moyes, Matthew Chalmers, and others who contributed to a brief GIST discussion on this.

HCI and marketing

by Steve Draper.

The question is: Should HCI include a study of marketing issues? and hence, should this be on the agenda for research, and on the curriculum in HCI courses?

There are 2 general aspects to this: is there an important overlap of subject matter in any way? and should we be seeking to recruit marketeers to the interdisciplinary teams for HCI and UI design? Here are some points.

Supporting the purchase decision

I'm interested in addressing what I think of as "the purchase decision task". I.e. most things, and certainly software, gets selected for use (and often purchase); this is a user task; designers should support it (even apart from commercial imperatives); this selection will have a huge effect on subsequent user satisfaction; and it may be that the limited spread of good usability is due to the fact that, like safety features in cars, purchase decisions are not (currently) driven by usability.

This may be the most sensible response to the recurrent moan that HCI has so little apparent effect in improving our average cultural standard of usability design: people buy bad interfaces because of their "features", not because of the wasted user hours and effort. If we aren't studying the whole user task set, it is not surprising our favoured designs aren't winning.

If we accept this, it is another case of most HCI failing to do more than a very blinkered task analysis. Most TA only looks at the actions a perfect operator would do to accomplish (with ideal efficiency) the most obvious tasks. Cursory observation of real users show that other tasks are important: error recovery being the most vital. But purchasing (or selecting for use) is another.

The website case

Websites are an acute inverse example of the issue. Many of both clients and designers feel they have to organise their website around serving the first glimpse — the moment when visitors decide whether to stay or go — rather than around any other user function such as being able to get information out fast, use it for quick reference frequently, etc. It is quite hard to avoid getting polarised on this; but presumably a rational view must take full and explicit account of both user "functions", and find designs that address both, not one at the expense of the other. (This will include beginning by analysing how important attention grabbing is for this particular web page, rather than some context-free guideline on the matter: after all, the main job of some pages is fast repeated information reference, but of other pages, to catch new visitors of all kinds....)

Once stated like this, we can start to see how to attack it. For instance, we could use as a metric, the number of repeat visits to a site: obviously, sites that don't attract anyone to stop in the first place will fail, as will those which arrest them but seem trivial and don't yield a fast information service.

Marketing people (and their goals)

Jackie complained about how marketing people don't care about the users' interests, but about grabbing their attention and manipulating them into actions that are profitable for marketers. On the other hand, they don't see the point of HCI people as they seem to be doing much the same thing (talking to customers).

I guess my provisional view would be:

The difference is, not in the methods, but in the dominant measure used: for HCI people, increased usability (e.g. better learnability, etc.); but for marketers, increased sales.
What's right about the marketing perspective is that things like communicating brand essence are in fact direct cognitive supports to user (purchase) decisions. These decisions are costly to users in similar ways that a poor UI is: waste time, lead to suboptimal actions. Supporting them is a real public service. After all, as I like to say in my lectures, supermarkets are the biggest familiar example of a DM UI: they spend huge amounts of "unnecessary" money renting floor space so that the goods can be laid out for customers, who are too stupid or lazy to think what they want, to react to the physical goods. Before supermarkets (I can just remember this) you had to know what you wanted and to recall (not recognise) it so you could ask the human interface for it. Now you just go and react. It is an enormous cognitive service: saves user effort, allows them to learn new possibilities at no cost when they visit the shop, etc.
Of course, going back to the first point above, the interests of the customer and vendor are not identical. But far from ignoring this issue, surely HCI should be addressing how to support the purchase decision task, and devising metrics for measuring how well or badly the users are served e.g. compare actual decisions made to choices made in artifical conditions of perfect information.

Dual vision

An effect of this is to generate a set of issues which have quite different aspects — an HCI and a marketing aspect — of the same technical issue.

The first is the one above: brand identity ...; catching the eye: attracting the customer vs. supporting purchase decisions.
The pressure to buy each new version of Word. Users experience not having the latest version as not being able to read attachments that are sent them. The vendor experiences this as the profits from monopoly and planned obsolescence. Of course other solutions are possible e.g. proper and convenient translation procedures.

The theory of value