Skip to main content
Search form
  • 00:04

    [Research Methods, case study][Researching Mobile Consumer Behavior Using Eye Tracking& Mixed Methods]

  • 00:17

    ZOFIJA TUPIKOVSKAJA-OMOVIE: Hello, my nameis Zofija Tupikovskaja-Omovie.[Zofija Tupikovskaja Omovie, Creative Economy EngagementResearch Fellow, Manchester Fashion Institute,Mancehster Metropolitan University]I'm CEEF Postdoctoral Research Fellow,at Manchester Fashion Institute with Manchester MetropolitanUniversity, UK.And I'm researching how digital consumer experiences can beimproved on shopping platforms.

  • 00:39

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: [How did you become interested in this research?]I think it comes back to the time when I startedmy PhD, which was back in 2013.I started researching mobile fashion consumer behavior.

  • 00:59

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: And after researching current literature and availableresources, it became clear that there were no actual frameworksor models applicable to mobile fashion consumer behaviorresearch.Therefore, I was looking for alternative research methodsto use in my research project, and here a mixed methods

  • 01:20

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: approach seemed to be the most appropriate.I was thinking of mobile consumer behavior as quitea new research area and we need to research and understandbehaviors in more detail of one particular such methodwould not answer the question.Therefore, I've chosen a mixed methods approach incorporating

  • 01:44

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: eye tracking technology, as well as user observation,interviews, retrospective think aloud,and statistical analysis.It's a very complex set, which I developed back during my PhD.And since then, I've been applying and reusingit in my other projects, with certain changes required

  • 02:07

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: as we come along.[What is eye tracking technology?]There are several types of eye tracking technology.Eye tracking technology used with desktop computers,which is set on a desktop screen whereby the user will

  • 02:31

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: be exploring something shown on the desktop,like a website or images or videos.And there is mobile eye tracking technology,which is eye tracking glasses operatedas spectacle type, which has internal cameras tracking eye

  • 02:52

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: pupil position, as well as in front cameras recordingwhat the person is looking at.And this technology overlays video recordingof where the person is looking, with the actual eye positionon that particular video, which makes it very exciting.

  • 03:13

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: [How did you design and plan your research?]Here, the research design as itself is very important.And planning which research method to use first and follow

  • 03:34

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: is very important, depending on what research questionand research aims at.And seeing how one or the other research methodcan complement each other, and understandingwhere to start with.For example, setting up eye tracking experiments.It was not only to track the actual eye tracking

  • 03:56

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: experience itself, but also to make observation notes as userswere exploring websites on mobile apps,as well as interviewing them before eye trackingand after eye tracking, which givesan overview about their self-reported behaviors,as well as a reflection after the actual experience.

  • 04:19

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: And there is another point which I used within eye trackingwith retrospective think aloud.We are showing users what they've done on the website,so they can actually see what they were searching forand where their eyes were wandering around,which makes them very interested in the beginning to focuson an actual reflection.

  • 04:40

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: And allowing them and asking themto reflect on that particular activity they were doing.[How did you recruit your participants?]It also depends on the purpose of this particular research

  • 05:02

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: project.And, therefore, the current researchwhich I'm presenting at this conferenceis based on cooperation with actual fashion retailer.Therefore, we analyzed the retailer's problem.We looked at their database of visitors to the website.

  • 05:25

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: Who are those visitors, their age,and which devices they use, what operating system they use.And based on that, I defined the proposed, the best, targetsample in terms of gender, age, and all those requirementswere sent back to the retailer who then recruited participants

  • 05:45

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: based on the requirements we needed.[Were there any ethical considerations?]Yes, of course, as with any other such project,ethical considerations are very important.We need to make sure that all data is anonymized.

  • 06:07

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: Therefore, I saw no participants enterinto the testing lab or any other interviewing setting.They are given a specific coded numberwhich is used all the time afterwards.So there is no single name mentioned or contact detailsor personal information at all.It's just numbers.

  • 06:27

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: So the only information we connect to that particular codeis probably more specific like gender, age,and some other little details to make surethat we can analyze later on.And also, working with industrial partners,it's also very important, if presenting findings

  • 06:49

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: at the conferences or publishing,to make sure that they themselves are also anonymized.So they're not revealed as the actual retailer's name.[How did you collect your data?]

  • 07:10

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: Before starting any eye tracking experiments,I always run a couple of demonstrations.I invite some random person to take partin a kind of preset research session, whichallows me to test the flow of the experiment, as well

  • 07:31

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: as the timing, and all other required procedures as datatransfer from one device to the other one,in order to make sure that everything goes smoothly,to know how much time is required.Because working with mobile eye tracking technologymeans data files are video based,which means they are really large files.

  • 07:53

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: And eye tracking glasses are connectedto a mobile-based device which records the sessionAnd, therefore, if I want to use it for the retrospective thinkaloud afterwards, it has to be transferred to PC desktopdevice, which takes time.So taking all those considerations in mind,and making sure that participants will not

  • 08:15

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: be sitting bored, I plan my sessions in the waythat we have pre eye tracking session of interviews.We have actual eye tracking.Then post eye tracking interview.Then we have retrospective think aloud,and then just final bits and parts.

  • 08:35

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: [How did you analyze your data?]Analysis takes way much more timethan actual data collection.And it's very important to transcribe all the interviews.

  • 08:57

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: Also, making observation notes as eye tracking sessionsare recorded, which is also helpful for interviews post eyetracking, which gives some ideas of certain questionsto ask to clarify why that user spent longerlooking at a certain area.You might think that this person was excited about the garment,

  • 09:19

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: but it appeared that the page was not loading.Which, if you don't really ask those questions,you wouldn't really find out.And so transcribing interviews is the first thing to do,as well as analyzing transcripts,which is another quite time-consuming area.I normally use Nvivo software for that.

  • 09:42

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: In terms of eye tracking data itself,we use BeGaze 3.7 software, which is specificallyattached to the eye tracking kit we have,which allows us to generate certain datasets from actual eye tracking experiments.So depending on what this particular project

  • 10:04

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: is interested in, they could be scan paths files,could be focus maps, could be heat maps, statistical datafiles generated, and certain other things done with it aswell, just to mention a few because it would take longto explain all of them.And, based on my research projects I've been involved in,

  • 10:27

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: I've found scan paths data files the most informativein this case.Because scan paths is a video.Well, in this case, we get video because this ismobile eye tracking technology.Scan paths is a file which allowsus to see how eye move from one pointto the other point of fixation, as well as

  • 10:47

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: showing how long that particular user has fixatedon that particular point.So it displays circles, which are bigger or smaller in sizedepending on the time spent fixatingon that particular moment.And these particular data files areused to create shopping journeys, whichis another thing I developed during my PhD,

  • 11:09

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: and I've been using since then, as the most valuable outputfrom eye tracking in this case.It's not actual-- it's not generated by the software.You have to manually mark it on different software,looking at the video back and forth, which is another verytime-consuming element.

  • 11:31

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: [What are the benefits of eye tracking?]Take into account that eye tracking recordingis more or less a quiet session by participants.I'm saying quiet because I don't tend to ask anything

  • 11:55

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: during the actually eye tracking to make sure that users areexploring the website or mobile app in the most natural wayas they would normally do.And that's what I normally ask them to do, and to makesure they are kind of absorbed in an actual shoppingexperience, as they would do it at home, or on the bus,or anywhere they are.

  • 12:15

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: In the beginning they feel a bit watched, which is normal.And then after a couple of minutes,they kind of forget about it and carryon browsing the way they would normally do.And that's very, very interestingbecause allowing the user to actually explorethe shopping platform the way they would normally doallows us to learn about actual shopping experiences

  • 12:39

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: and shopping behavior, which we're interested in for marketresearch.[How did you combine the interview and the eye trackingdata?]So having this mixed methods approach and having eyetracking interviews, observations,

  • 13:00

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: allows us to triangulate our research findings.And, therefore, eye tracking is capturing actual user behaviorand experiences, in the way it tells us what users do,how long is the time they spend on certain things, whatthey look at.Whereby interview data allows us to understand

  • 13:22

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: why they do certain things, how didthey find those experiences, what featuresor functions they might have found usefulon the digital platform which they didn't come across.Certain things that you cannot unpick just from one researchmethod.Therefore, having an ability to triangulate different data sets

  • 13:45

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: and developing this robust understandingabout digital user behavior is very important.It's like building a pyramid with actual three solid walls.So knowing the participant and the userfrom all sides of their experiences.[How did you visualize your data?]

  • 14:13

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: For home page activities, then refining activities as well.Third one is product page, and the last one is check out.So when the users are exploring the website,they're not seeing just one particular page.They are actually seeing live digital platform,which is huge and has loads and loads of pages on it.

  • 14:36

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: And, therefore, each user is free to explore any pagesthey're interested in.And, in this case, to make it more standardized,it was easier to use this four-level shopping journeymap, which allows us to map the actual experienceas it goes along.And then, as presenting for final reports

  • 15:00

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: and analyzing further up, the shoppingjourney itself is not enough.We need to look at more numerical data.So what I'm interested in--I develop tables of average steps usershave done on home pages, on product pages,at the check out, browsing, how many product pages

  • 15:21

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: they add to the basket, how many product pagesthey decide to purchase, and things like that.And then, based on that information,we can do comparisons.I've done that to compare mobile apps and mobile websitesof the same fashion retailer.And, to be honest, the differenceswere very, very big.

  • 15:44

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: On the mobile app, users visited three times more product pagesthan on the website, and this was the same users explodingboth shopping platforms.Whereby, in terms of time spent, it was more equal.So, therefore, the question is--the retailer might think that users enjoy the mobile appmore, because they're actually viewing more pages.

  • 16:04

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: But there was a really big problemon the platform, which the retailer was notaware of unless we had done this comparison, which revealsthe need to know why certain activities are morecommon than others, for example.And then, statistical tables comparisons are good but,

  • 16:27

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: in terms of filing a report, it'svery good to have qualitative datawhere interviews come along.And having that journey, and then they find certain usersgoing in circles, and then talking to them,and interviewing them, and they'retelling you that they actually struggledto find certain things.Therefore, having an eye tracking video recording, which

  • 16:50

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: can be visualized as screenshots of certain pages,as well as having that qualitative datato add and describe certain activitiesmakes it very visually impactful to understand,and much easier to understand than actual complicatedshopping journeys.

  • 17:12

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: [Did you face any challenges and how did you overcome them?]The last project we'd done, we asked a retailerto recruit participants who own an iPhone,because that was the main customerbase accessing their website.

  • 17:34

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: And we provided all participants with the same iPhoneto use to make sure that we have all exactly the same researchconditions for all of them to avoid any bias.And then participants come along,and they actually accessed the websiteon three different browsers.And that's something that you wouldn't really

  • 17:55

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: expect that much.So then, as a researcher, you carry on collecting the data.And then checking whether those websites differ a lot or not.So luckily, in this case, the differenceonly was in terms of where the back button was.So, for example, if you go on a particular pageand then you decide to return back the previous one,

  • 18:17

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: that's the only difference.Otherwise, the website layout was the same.So, in this case, there was no problemto carry on collecting data and analyzing,because the actual website layout and pages were exactlyidentical.I think when you work with human beings,it's very important to be able to be flexible and adapt

  • 18:38

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: to different sorts of situations.Because you cannot really have a very hard,specific selecting criteria which will make much moredifficult to recruit participants.And eye tracking sessions are time consumingfor participants, as well.They might need to spend between one and oneand a half hours, which requires quite an effort as well.

  • 19:01

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: So it makes it harder to recruit.This is not the best.I think being a researcher, it is very importantto be creative and adapt as things come along.There was another research projectI've done in the past whereby we discussed with the retailer,in terms of gender, and somehow--

  • 19:26

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: well, I suggested we would recruitfemales of a certain age.But somehow the retailer managed to recruit more malesthan females, which was an absolute shock.So, in that situation, having kind of balanced samplingto a different direction, I suggested

  • 19:47

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: recruiting more females to add in,and then making a comparison between females and males,which is additional resource to use as well.So having a creative approach I think is important.[If you were to redo the research,is there anything you would do differently?]

  • 20:09

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: This particular research design I'vedeveloped, applying mixed methods,including eye tracking, interviews, and observations,worked quite well for projects that I've been using.However, if I would have to use it in a different setting,let's say, in media design or manufacturing domains,

  • 20:34

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: potentially there would be certain differences in termsof setting eye tracking.And setting times and tasks would be different.In that case, we would need to testwhat things can be done, what the client needs, and adaptin those settings accordingly.

  • 20:54

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: It's very important.So far, in fashion retailing domain,this framework works really well,and I'm very happy with it, and I would carry on applying it.[What tools and resources do you recommend for this research?]

  • 21:16

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: There are different eye tracking kits available.And it all depends on the institution,what they have access to.So a very important thing is to explorethe kit yourself in your free time,to make sure you know what you can gain from it,and to know exactly what you're trying to achieve, and design

  • 21:40

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: it accordingly as well.So, in terms of tools, developing shoppingjourneys I've found very useful, Bizagi, a process modelsoftware, which allows us to add--I don't know really whether thereis a limit, kind of limitless number of elements into it,so the actual shopping journey can stretch

  • 22:02

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: more than 100 steps ahead.But I've tried all different softwarethat didn't work because they have limitedkind of framework, which doesn't allowyou to add many things in.Nvivo is a great tool to analyze qualitative data.In terms of statistical analysis, SPSS is quite good.

  • 22:23

    ZOFIJA TUPIKOVSKAJA-OMOVIE [continued]: But it also depends what type of analysis is required.Sometimes some simpler tools are enough.


Zofija Tupikovskaja-Omovie, PhD, Postdoctoral Research Fellow at the Manchester Fashion Institute, Manchester Metropolitan University, discusses her mixed-methods consumer-behavior research incorporating eye tracking technology.

Looks like you do not have access to this content.

Researching Mobile Consumer Behavior Using Eye Tracking & Mixed Methods

Zofija Tupikovskaja-Omovie, PhD, Postdoctoral Research Fellow at the Manchester Fashion Institute, Manchester Metropolitan University, discusses her mixed-methods consumer-behavior research incorporating eye tracking technology.

Copy and paste the following HTML into your website