Think back to the last time you visited a museum or gallery. Think carefully – consider each step you took and every decision you made.  Which way did you turn? What did you see and what did you miss? Did you look at any labels? Did you move through some spaces quicker than others?

Answering these questions – what visitors do, why they do it, and how museum design can influence such behaviour – is fundamental to visitor research, and what better way to find out than by watching them?

Early research: drift in; drift right; drift out

The first systematic studies of museum visitor behaviour were conducted in the 1920s and 1930s. These publications[i] continue to be influential and are generally credited as the foundational works in museum visitor observation.

Armed with a stopwatch and notebook, early researchers discreetly followed lone visitors to (mostly) art museums, taking care to observe without themselves being noticed (and thus disrupting the behaviour they were trying to document).

These early studies built up some general patterns of visitor behaviour, and coined terms such as ‘attracting power’ – a measure of how many visitors stop at an exhibit – and ‘holding power’ – a measure of how long they stay – which are still studied today.

A commonly-observed pattern of movement was visitors entering a gallery, turning right, and closely following the wall until they reached the nearest exit. Thus art on the left or centre of the gallery went relatively unobserved. This ‘right turn’ bias is still studied, and while it is by no means universal it has been observed elsewhere (Paco Underhill, in his book Why we buy: the science of shopping, describes a similar rightwards drift in retail settings).

Observation also revealed that as a visit wore on, the pace picked up: visitors took their time to look at art early in their visit, but successively sped up until they were moving through galleries quite quickly, barely pausing to look at any works. This was one of the earliest observations of the ‘museum fatigue’ phenomenon, where visitors gradually run out of physical and / or cognitive ‘steam’, or simply had seen enough for one day.

Late 20th century: diligent or dilettante?

After the second world war, museological priorities shifted and one apparent casualty was visitor observation. The field was virtually neglected until the 1970s and 1980s, when visitor studies came to be recognised as a distinct discipline.

This renewed interest arose with a shift in the perceived role of public museums. Collections, in and of themselves, were no longer seen to be enough. Museums increasingly had to justify their presence (and their funding) as sites of public education and enrichment. The emergence of hands-on museums, with an explicit educational role and a funding-driven need to evaluate their exhibits and programs, also catalysed a shift to a more visitor-focused outlook across the museums sector as a whole.

Again, most of these later studies were done using a stopwatch, pencil and paper, marking where people went, where they stopped, what they looked at and how long they stayed. But the question remained – what can be generalised from these observations? Are they anecdotes or data?

In the late 1990s, through a meta-analysis of over 100 of these observations, some patterns did emerge, and the results might have been disappointing for the curators who had carefully selected objects, designed interactive exhibits and crafted interpretive labels. Most visitors breezed through their lovingly-produced creations in just a few minutes, passing most displays with barely a second glance. One metric of the study was the proportion of ‘diligent’ visitors, defined as those who stopped at over half of all exhibit elements. On average, only a quarter of visitors fell into this category, meaning the vast majority took in less than half what was on offer.

But was this necessarily a bad thing? Recent commentators have pointed out that low ‘diligence’ is an inevitable consequence of visitors following their own agendas. Visitors come to satisfy their own curiosity, which may be gratified long before the exhibition has finished ‘talking’ about a particular topic. From the point of view of the visitor, skimming an exhibition can be just as successful as carefully studying it.

Caught on tape: audio and video recording of visitors

Obviously, researchers cannot faithfully record everything visitors do, say and notice by pen-and-paper recording methods. Real life has no replay button to catch those things you missed the first time around. Thus in-depth study of specific exhibits calls for audio and video recording.

Video recording works best either for individual exhibits, or in small spaces which can be captured in a single camera’s field of view. Using this method, researchers have produced detailed and sometimes profound vignettes of visitor-exhibit exchanges. Audio recordings have caught visitors repeating snippets of label text in their conversations, proving that they are reading more text than first thought.

While audiovisual recording is a source of rich and detailed data, it does not scale well to studies of whole galleries. Trying to follow individual visitors through footage of multiple cameras presents a monumental data management task (and that’s assuming the camera angles catch what you’re looking for in the first place). Audio recordings can be hard to follow after the event, particularly if you don’t know where visitors were standing at any given point in time, and you can’t always distinguish who is speaking keeping the cheap, practical and flexible pencil-and-stopwatch method the standard, at least for the time being.

Where next?

Trackers, smartphones and the ‘O’

To borrow from Niels Bohr: prediction is difficult – especially about the future.

As electronic technology becomes a more ubiquitous presence in our lives, a reliable and cost-effective tracking tool may eventually supersede paper-based methods.

Like it or not, we already share a lot of information about our day-to-day lives through GPS-enabled smart phones, social media and internet use. For market researchers, this is raising the bar of expectation – if we can follow website visits click-by-click, what’s stopping us getting similarly rich data about real-life visits to malls and museums?

Radio Frequency Identifiers (RFIDs) are already widely used in the retail sector for supply chain management. In theory, RFID-tagged tickets or lanyards could be used to track people too – although such an approach can be expensive (not to mention feeling a tad intrusive).

Smartphones offer more promise[ii] as we are already used to carrying them. Data is based on the signals the phone is emitting anyway, and can be collected while maintaining visitor anonymity. Using visitors’ own phones also makes it feel like less of an imposition. This means that the tracking process is less likely to influence visitor behaviour.

Onsite receivers can be installed to make use of the TMSI – the electronic handshake that every mobile phone periodically makes with its nearest base station. Piggybacking on the TMSI signal can allow time-stamped locations to be taken with an accuracy of about 1-2 metres. Weaving together these snapshots can then give an overall visitor path. Drawing on phones’ inbuilt Bluetooth and Wi Fi capacity are also possibilities, but no solution on its own is a silver bullet.

So far, it appears that phone tracking has yet to cross over from retail to museums, although in Australia the Powerhouse Museum is currently trialling a phone-based system in their Lace exhibition, supported by the NSW Trade and Investment’s Collaborative Solutions Program.

Another possibility is to make tracking an intrinsic part of the visitor experience. This is what Hobart’s Museum of Old and New Art (MONA) has done with its “O”, an interpretive device that doubles as a visitor tracker.

MONA has taken the bold step of having no printed labels at all, just the O (an iPod Touch loaded with specialised software) which visitors are given on arrival as their tool for navigating the exhibition spaces. At the same time as providing interpretive material, the O is recording a visitor’s every move: where they stop; what they look at; how long they spend looking at it. “A dedicated sensor network in gallery ceilings and walls monitors the position of an RFID tracking tag attached to the mobile device. It is similar to how GPS works, but for indoors,” says Tony Holzner of Art Processors, a Mona venture created to commercialise the O. The O is thus quite different from website analytics tools, which are based on IP addresses.

Visitors to MONA can get the O to save a copy of their track through the museum and information about what they looked at to their own email addresses. Meanwhile the data accumulates, giving MONA a good sense of what visitors are doing on the gallery floor.

While MONA already has an enormous wealth of information to draw upon, this is still not the whole picture. Where are visitors coming from? What’s being said on the gallery floor? What else are visitors doing? MONA hopes to find out, but so far at least, they are beyond the capabilities of the O. As curator Nicole Durling acknowledges, ‘there are some things you still need to stand there with a clipboard for’.

[i] A more detailed review and bibliography can be found in Timing and Tracking: Unlocking Visitor Behavior by Steven Yalowitz and Kerry Bronnenkant (2009) Visitor Studies, Vol 12 no 1 pp47-64

[ii] For a more detailed overview of mobile phones as visitor tracking devices, see

NB: This article was commissioned by Artlink magazine and was first published in the Art and Surveillance issue (Vol 31 No. 3) in September 2011.

Posted on March 5, 2012