What’s the colour of money?

Tim:
Thanks for the oppor­tu­ni­ty to tell you a bit more about CX Lab and how we use behav­ioral sci­ence in both the vir­tu­al and the real world to inves­ti­gate human behav­ior, specif­i­cal­ly in both cus­tomers and employ­ees of busi­ness­es. So, we’re very much focused on the busi­ness side of things and emu­lat­ing some of the things that Dan does, but not as excit­ing, I’m sad to say. We’re inter­est­ed in what real peo­ple do in real-time, so that we can both improve their expe­ri­ences and the com­mer­cial returns for our clients and busi­ness­es. So, we’re focused on that busi­ness side of things. So, today I want to illus­trate that, espe­cial­ly with our online research, by talk­ing about a par­tic­u­lar study that we did, which will hope­ful­ly throw some light on a very hard prob­lem. And that is, how does col­or influ­ence human behav­ior in the real world, in the real com­mer­cial world, and how might a bet­ter under­stand­ing of this help busi­ness? In oth­er words, what is the col­or of money?

Tim:
Now col­or, as we know, plays a huge role in busi­ness. In terms of busi­ness­es, com­mu­ni­cat­ing with their cus­tomers, spend­ing bil­lions and bil­lions of dol­lars a year to get the right col­or on logos, com­mu­ni­ca­tions, store design, all those kinds of things, to make cus­tomers more like­ly to buy prod­ucts and ser­vices. And in fact, here’s a wheel that I found from a design com­pa­ny that tells you exact­ly what all col­ors actu­al­ly mean. Very inter­est­ing. How­ev­er, we know from exper­i­ments like this one here, and I’m sure you’ve seen this before, that human per­cep­tion’s actu­al­ly flawed. If you look at the two squares, A and B, you’re like­ly to see square A as being dark­er than the square B, when in fact they’re the same col­or, because the con­text that you view it in, things are dif­fer­ent. And sim­i­lar­ly, with this col­or­blind­ness test here, if you see the num­ber 21 in this bunch of dots, then you’re like­ly to be one of the 8% of men, although only a half per­cent of women, inter­est­ing­ly, who are red-green col­or­blind. If you see 74, then you’re not.

Tim:
So, I think what this shows is that our reac­tion to col­or is not pre­dictable because it’s indi­vid­ual and it’s dif­fer­ent. So, it’s not just the mechan­ics of how we per­ceive col­or in our brains, how our visu­al neu­rons respond to light, but also the con­text with which we view the col­or can make a huge dif­fer­ence to both our per­cep­tion and then indeed our response. So, on our col­or wheel here, you can see that red is asso­ci­at­ed with bold­ness and pas­sion, but also in many cul­tures, as we know, it sig­nals dan­ger and some­thing to be avoid­ed. So, clear­ly for a prod­uct or a ser­vice that might not nec­es­sar­i­ly be a good thing. So, col­or per­cep­tion is an issue. So, there’s the phys­i­o­log­i­cal and indeed the psy­cho­log­i­cal aspect of what col­or means to peo­ple. But there’s an even big­ger prob­lem too, and this is one of the rea­sons that CX Lab exists, and it’s not just con­fined to col­or research but gen­er­al­ly in busi­ness, how busi­ness­es find out stuff is they ask peo­ple questions.

Tim:
And as we’re increas­ing­ly real­iz­ing from behav­ioral sci­ence, ask­ing peo­ple ques­tions about things is a poor way of get­ting to the truth. David Ogilvy, known as the father of adver­tis­ing, summed that up very neat­ly in the 1970s, a quote I’m sure some of you have seen before is that, “Peo­ple don’t think how they feel, they don’t say what they think and they don’t do what they say.” And this is not nec­es­sar­i­ly because peo­ple are being delib­er­ate­ly dif­fi­cult. Although, we know that this can hap­pen, but rather that we’re unaware of these uncon­scious dri­vers of our behav­ior. And when asked, we’re very like­ly to miss attribute or post-ratio­nal­ize our behav­ior based on what we think to be true and we might actu­al­ly fer­vent­ly believe that to be true, but that’s not nec­es­sar­i­ly what’s dri­ving our behav­ior. And you will all almost cer­tain­ly be asked mul­ti­ple times a day, ques­tions about prod­ucts or ser­vices. To rate them out of 10, to see whether you would rec­om­mend some­thing to your fam­i­ly and friends.

Tim:
In fact, you may even be asked to rate this pre­sen­ta­tion after the event, but you’ll be doing that in a post-ratio­nal­ized way. It won’t nec­es­sar­i­ly be what you actu­al­ly feel or what is behind the dri­vers of your behav­ior. So, in that regard, we want­ed to under­stand more about col­or behav­ior. In answer­ing these busi­ness ques­tions, we look to mon­i­tor peo­ple’s behav­ior rather than ask them ques­tions. We put par­tic­i­pants in envi­ron­ments, which are like the things that we’re inves­ti­gat­ing. So, like the con­text of the behav­ior we’re inves­ti­gat­ing. So we can find out what they do and why they do it. And in this par­tic­u­lar study, we’re obvi­ous­ly look­ing at what col­or influ­ences peo­ple, where they shop, how much they’ll be pre­pared to pay and how there­fore, by under­stand­ing col­or and using it bet­ter, busi­ness­es could gen­er­ate more sales, more income, more prof­it. And as in all our work, it’s about under­stand­ing behav­ior rather than ask­ing questions.

Tim:
So, this par­tic­u­lar study was for a retail opti­cians group and was an online behav­ioral study. They want­ed to rebrand their stores in order to increase their mar­ket share amongst a num­ber of big name com­peti­tors. And there’s a pletho­ra of small, inde­pen­dent opti­cians, which I’m sure you know about. They’ve done some tra­di­tion­al, “Which col­or do you like best?” Kind of research, by ask­ing peo­ple, and had some doubts over the results. So, they want­ed to under­stand bet­ter from a behav­ioral per­spec­tive, how these col­ors influ­ence what peo­ple did. Specif­i­cal­ly, they want­ed to know whether col­or influ­enced first choice pref­er­ence, which is clear­ly a key cri­te­ria in dri­ving busi­ness in this mar­ket­place, and also whether col­or influ­enced peo­ple’s per­cep­tion of val­ue for mon­ey with­in a par­tic­u­lar store. And whether one of those par­tic­u­lar can­di­date col­ors that they had was pre­ferred by people.

Tim:
We did an exper­i­ment via the very excel­lent Goril­la online behav­ioral plat­form and recruit­ed 1,200 peo­ple to take part in this exper­i­ment with a bal­anced mix of demo­graph­ics. We screened peo­ple for cor­rect­ed vision, so we knew they were in mar­ket. And we also want­ed to find out where they last bought their glass­es or con­tact lens­es, so we could under­stand who was their par­tic­u­lar incum­bent brand. And they car­ried out a num­ber of tasks online so that we can under­stand a bit more about how col­or was influ­enc­ing what they were doing. I just want to give you an overview of the kind of tasks that we got peo­ple to do. So, first of all, we designed and built a rep­re­sen­ta­tion of a typ­i­cal shop­ping mall with three dif­fer­ent opti­cians in it. And we asked par­tic­i­pants to indi­cate which one they would pre­fer to shop in by click­ing on the store­front. We rotat­ed and coun­ter­bal­anced the order of the posi­tion of all these stores and we had our client brand com­pet­ing against the two lead­ing high street brands, and rotat­ed the col­ors, the can­di­date col­ors accord­ing­ly, so that we could see how they com­pare to each other.

Tim:
We also forced peo­ple to make a sec­ond choice by telling them when they clicked on their first choice, that the store was closed and there­fore, where would they go sec­ond? So again, we could see how things stacked up in a sec­ond choice pref­er­ence. Now, inter­est­ing­ly we found that no one par­tic­u­lar can­di­date col­or came out on top, but what we did find was that our client brand was notable in that it was cho­sen least often amongst its com­peti­tors. High­light­ing the crit­i­cal need for the rebrand­ing and the exer­cise that they were car­ry­ing out to try and boost that first choice pref­er­ence and to boost con­sid­er­a­tion and brand aware­ness. We also dis­cov­ered that this was true even amongst the brand’s exist­ing cus­tomers, 40% of whom did not choose their own brand first. High­light­ing anoth­er issue in terms of the actu­al cus­tomer expe­ri­ence and some­thing that would also need to be addressed.

Tim:
So, that was the first task, to rotate peo­ple through these dif­fer­ent oppor­tu­ni­ties, these dif­fer­ent choic­es. We sec­ond­ly want­ed to look at whether col­or had an impact on cost per­cep­tion. So, whether col­ors and brands com­mu­ni­cate val­ue for mon­ey. So, we showed par­tic­i­pant dif­fer­ent store­fronts, and we got them to indi­cate how much they would expect to pay for their pair of glass­es with­in this par­tic­u­lar store. And we got each par­tic­i­pant to look at two dif­fer­ent stores. Again, we rotat­ed the can­di­date col­ors of our own brand, our client brand, against the two lead­ing com­peti­tors. And we found that col­or indeed did have a sig­nif­i­cant impact on what peo­ple expect­ed to pay. With one of our client can­di­date col­ors offer­ing the best val­ue for mon­ey and com­pa­ra­ble indeed with its com­peti­tors. And this was con­trary to the con­scious ques­tion that we asked peo­ple at the end of our study, in a more con­ven­tion­al way, which indi­cat­ed that they saw this par­tic­u­lar col­or as both expen­sive and indeed pre­mi­um. And there­fore, these two things are run­ning counter to each oth­er. What peo­ple say and what peo­ple actu­al­ly do, being very different.

Tim:
So, that’s again, how that exper­i­ment was run. It looks very com­pli­cat­ed, but again, each per­son only had two choic­es to make. In the third task, we again asked peo­ple to look at some­thing online. We pre­sent­ed them with a 3D vir­tu­al rep­re­sen­ta­tion of a store and asked them to browse with­in it, to have a look round. They could alter the angle that they were access­ing, that they could walk around, swiv­el and change the shape of the store itself. And the only dif­fer­ence between these stores was in fact, the col­or of the back­ground, as you see in these images here. What was inter­est­ing was that again, in our post-exper­i­ment ques­tion­ing, some three-quar­ters of peo­ple did not real­ize that these stores were exact­ly the same apart from the col­or. And some of them went into the very elab­o­rate rea­sons why they found the stores to be dif­fer­ent, even though they were only dif­fer­en­ti­at­ed by color.

Tim:
Again, the asso­ci­a­tions that peo­ple make with these par­tic­u­lar things. And in this par­tic­u­lar part of this task that we ran, again, one col­or emerged as a clear pref­er­ence for peo­ple. With these lat­er ques­tions post-ratio­nal­iz­ing that pref­er­ence to do with the fact that they saw this envi­ron­ment as being more pro­fes­sion­al and more relax­ing. With 30% more peo­ple in fact, pre­fer­ring this par­tic­u­lar col­or scheme. And final­ly, for the final task, we want­ed to cre­ate a kind of proxy for this idea of clin­i­cal excel­lence, some­thing which is very, very impor­tant clear­ly in the opti­cians world. That the idea of trust­ing your eyes to some­one who is clin­i­cal­ly excel­lent is clear­ly very, very impor­tant. Long-term eye health being absolute­ly crit­i­cal to almost all of us.

Tim:
So, in this vir­tu­al shop­ping mall we took peo­ple back to and we pre­sent­ed them this time with two stores, which were sim­ply just a den­tist, and again, very clear­ly marked with just the col­ors being dif­fer­ent. But we told them to imag­ine that they were in urgent need of treat­ment, they’d just devel­oped a ter­ri­ble toothache and they need­ed to get it sort­ed straight away. So, where would they go? And again, we were keen to see how col­or influ­enced the deci­sion that they made, because it was the only dif­fer­ence between the stores. Again, the posi­tions were rotat­ed amongst the two col­ors. And again, we saw that one col­or was sig­nif­i­cant­ly pre­ferred to the oth­ers based on this test. So again, we were see­ing that col­or was hav­ing an influ­ence on the deci­sions peo­ple were mak­ing in this environment.

Tim:
So, what did we find? In the world of retail opti­cians, which col­or offers the best com­mer­cial return? In this par­tic­u­lar case, what was the col­or of mon­ey? And the answer was pur­ple. So, from our behav­ior exper­i­ments, the dif­fer­ent tasks that we ran in terms of val­ue for mon­ey, head to head choice, and this idea of clin­i­cal excel­lence, pur­ple was the clear win­ner in the bat­tle of the dif­fer­ent can­di­date col­ors, cut­ting through a lot of the noise that had been gen­er­at­ed by tra­di­tion­al research, ask­ing peo­ple ques­tions about which they liked. And despite the com­plex­i­ty of the exper­i­men­tal design and the num­ber of peo­ple, par­tic­i­pants that we had, and the rotat­ing a bal­anced nature of the study, which Goril­la was able to han­dle very eas­i­ly, I have to say, we were able to pro­duce this insight from ini­tial brief to the pre­sen­ta­tion in just over three weeks.

Tim:
So, it can be done very, very quick­ly. And I guess, it’s just one exam­ple of the sort of online behav­ior exper­i­ments CX Lab have done and would like to do more of in the real world to mea­sure this authen­tic behav­ior in dif­fer­ent con­texts, which are rep­re­sen­ta­tive of what peo­ple would actu­al­ly do in the real world. And mea­sur­ing their behav­ior, the out­put, rather than what they say they would do or did do. And I guess, it offers all busi­ness­es the oppor­tu­ni­ty to car­ry out this kind of rapid qual­i­ta­tive and quan­ti­ta­tive mix of exper­i­men­ta­tion with big, robust sam­ple sizes to pro­vide a good degree of con­fi­dence in com­mer­cial deci­sions. So, that’s my rather rapid take through of one par­tic­u­lar exper­i­ment that we’ve done. So, thanks very much for your time and I look for­ward to answer­ing any ques­tions that you might have.

Speak­er 2:
Tim, that’s amaz­ing. Look, just in case every­body else … Oh, I can’t share my screen while you’re shar­ing your screen, but if I under­stand it cor­rect­ly, Tim, Vision Express came to you, asked you, “What col­or should we rebrand our com­pa­ny with?” And rather than just going, I like these col­ors for what­ev­er rea­sons and read­ing the lit­er­a­ture, you ran three exper­i­ments and found what will actu­al­ly dri­ve them the most busi­ness. And they have imple­ment­ed it. I can’t share my screen, but if you want to, you can go to Vision Express. Look at this. This has hap­pened in the real world.

Tim:
Indeed.

Speak­er 2:
Tim has changed the world with his sci­ence and with behav­ioral sci­ence, which I just think is bril­liant. It’s so rare, I think, in acad­e­mia you often feel like you’re doing the research, but it nev­er has any impact in the real world. And Tim’s here run­ning exper­i­ments in the real world and inform­ing busi­ness­es who actu­al­ly change what they can do in real-time, so that you can see the impact.

Speak­er 2:
I’m real­ly very inspired by that. And for every­body in the chat, say if you’re inspired or what you to think about how your research could be applied in indus­try. “Love world-chang­ing sci­ences,” we’ve got here. [inaudi­ble 00:14:33] mes­sages com­ing in as we chat. Now, I have a ques­tion for you. So del­e­gates, if you’ve got a ques­tion, put them in the Q&A and Tim answer them once he’s fin­ished chat­ting to me. But my ques­tion for you, Tim, is do you think there are oth­er appli­ca­tions for this kind of online, fast turn­around research?

Tim:
Yeah, absolute­ly. We’ve done a num­ber of things our­selves. I mean, the appli­ca­tion is only real­ly lim­it­ed by the ques­tions that were being asked by busi­ness­es, of which there are many and diverse ones, as Dan has showed you. Some inter­est­ing ones, as well as some rather dull ones, “What’s the right col­or?” And also, the skill of the sci­en­tists in design­ing exper­i­men­ta­tion to be able to answer those ques­tions and the only lim­i­ta­tion is by what you can actu­al­ly do online. So, we’re all about this track­ing of behav­ior. So, out­put in the con­text of the mar­ket­place cus­tomers are actu­al­ly oper­at­ing in and buy­ing in. So, we want to emu­late the choice archi­tec­ture of their deci­sion mak­ing that we make as close­ly as pos­si­ble and there­fore, its effect on their behav­ior. So, obvi­ous­ly this requires us to be pret­ty smart in the kind of exper­i­ments we design, with­out blow­ing our own trum­pet too much.

Tim:
And it also often requires a bit of mis­di­rec­tion, which you can get away with a bit more in com­merce than you can in acad­e­mia, in terms of what peo­ple under­stand to be what you’re try­ing to do. The rea­son that they’ve come to you in the first place. So, we’ve done a whole bunch of things, from mon­i­tor­ing dri­ver reac­tions to video sim­u­la­tions, actu­al­ly alter­ing some vari­ables in terms of caf­feine and hunger and things like that, to change, to see how they react. How cus­tomers choose, pay for, and play online games. So, we’re doing some work with Camelot and indeed, how much peo­ple are pre­pared to pay for car insur­ance based on the cus­tomer jour­ney they go through online and how much they’re pre­pared to save. So, lots of finan­cial ser­vices appli­ca­tions. We’re cur­rent­ly look­ing into stuff on wait­ing time and how that affects how much peo­ple are pre­pared to spend in a sce­nario where a delay is inevitable.

Speak­er 2:
Right.

Tim:
So, where you have to slow peo­ple down, what is the opti­mal time to slow peo­ple down so that they still spend what you want­ed them to spend? If that makes sense?