How a 300-year-old brand tests what helps customers decide, without scarcity tactics and countdown timers.
Read the full transcript
Matt: it’s not just gonna be a Google search anymore. It’s gonna be a 10-minute conversation with an LLM before they even reach the site, if they reach the site.
Sarah: Yeah, 100%.
Introduction
Matt: Hi, everyone. Welcome to this podcast between REO Digital and Fortnum & Mason. My name’s Matt, head of, CX at REO, and I’m joined today by Sarah.
Matt: Thank you for having me.
Sarah: I’m Sarah Wilkinson. I am director of online for Fortnum & Mason.
Sarah: I’ve been there for nearly two years now, um, looking after everything digital. So I look after the, uh, digital marketing team, the merchandising and trading teams, and also the optimization team. And that team’s been working really closely with you.
Matt: And so I know most people will probably be familiar with the brand, but for anyone that isn’t, can you give us a bit of a background on Fortnum’s?
Fortnum & Mason’s background and brand heritage
Sarah: Yeah, of course. Um, so Fortnum & Mason was started in 1707, so it’s been, a brand for over 300 years. And Piccadilly really sits at the heart of that. So, um, Piccadilly is our central London retail store. And over time, we’ve expanded that retail footprint out, so we have stores now in Bicester Village, um, we have stores in London, St Pancras, also in Heathrow, um, and Rex.
Sarah: So we’re expanding all the time, um, which is really exciting. But what’s especially exciting for me is that online has now become a third of the business- Wow … so really significant chunk. We’ve seen incredible growth post-COVID, but then we’ve continued to keep momentum, which is fantastic. So, um, so yeah, it’s a really exciting place to be.
Matt: Amazing. And with kind of ever-changing needs, I guess, from digital customers particularly, how are you balancing the work you’re doing online in terms of optimization and elsewhere with those kind of 300 years of brand values and heritage? How are you balancing those priorities?
Sarah: I think when you’re custodian of a brand like Fortnum & Mason, you are really conscious that you have a responsibility to maintain that amazing customer service at every single touch point. And so everything we do really puts the customer at the heart. and so making sure that that experience online feels as incredible as the store, um, is top of the list for us.
Matt: Mm.
Sarah: And over time, I think we’re, we’re kind of moving in that direction where there’s, the website starts to feel more immersive, starts to r- really replicate that experience in the store, which is amazing.
Matt: Mm-hmm. Cool. And the, the sort of history before we started working with REO, so end of last year, how have, how has your kind of strategy changed in terms of online and, um, utilizing your existing tech stack or new tools?
Matt: How has that changed since the end of last year?
Sarah: Oh, it’s changed hugely. I think as a team, we’re incredibly good at using quant data. Mm-hmm. So we have access to Contentsquare, we use Adobe Analytics, and we had some really good processes in place to utilize that data to drive our AB testing program.
Sarah: But I think go, kind of going back to the customer point, I think we missed, um, that piece, that huge piece of the puzzle really, which I think you brought to the table, around really understanding that customer experience. And so- We were able to pull together that customer research, watch- Mm … in real time what customers were doing on the website, and make some quite strategic decisions on the back of that, which was fantastic actually.
Sarah: And I think just small things like, um With our, with our hampers, we had, um, we, we used a variety of different, um, terminology, I guess-
Matt: Mm-hmm …
Sarah: to talk about hampers. So we would talk about a hamper, we would talk about a wicker. And I think one of the really fascinating things that very early on you came to me and said was, “Nobody knows what a wicker is, Sarah.”
Matt: Mm-hmm.
Sarah: So when you’re saying what’s inside the wicker, you’re, confusing for customers. And so just really small things like that became really interesting for us to understand, and really start to update the way that we speak about our products so that customers really understand. Um, and I think we’re really keen to, again, kind of extrapolate some of that data to think about some of the AB testing that we’ve put in place.
Sarah: So, um, going back to your brand heritage point, for us, it was really important that anything we tested brought the products to life in a more effective way, or h- really supported customers making quite difficult decisions on our website. So if you think about customer behavior on our site versus perhaps a supermarket-
Matt: Mm-hmm
Sarah: customers are coming to us making quite big decisions from a discovery perspective. So they’re not really coming with a, with a view of what they want to buy necessarily. Some customers are of course, but a huge wave of customers come wanting to be inspired, wanting to find something exciting, wanting to find the best present for their mom or dad-
Sarah: or grandad. Um, and for us it’s about how do we support them in that decision-making process in a way that gets them really excited and inspired. And so working with you to test things like
Hamper contents on the product page
Sarah: how do we bring to life hamper visibility on PDPs, how do we showcase what’s inside hampers much more effectively than we have before, has enabled us to kind of m- meet, meet that brief really and really showcase what we can do.
Matt: Yeah. I think from our perspective, that’s why it’s been so exciting because it would’ve been very easy with a retail site to kind of come in and look at the kind of traditional CRO tactics of countdown banners and social proof and that kind of stuff. But I think the results of the program have still been really positive without having to do that and kind of compromise the values of the brand along the way.
Sarah: No, absolutely
Sarah: … it’s always been at the core of it.
Sarah: And I totally agree. I don’t think, I don’t think Fortnum & Mason are gonna have a countdown timer-
Sarah: No
Sarah: anytime soon.
Matt: No.
Sarah: But we definitely have simplified processes as well-
Matt: Mm …
Sarah: I think, which has been really important.
Matt: Yeah.
Item count on listing pages
Sarah: So we had a te- we had the test where we showcased the number of items that are in each hamper, which, on, on PLP, which really did help customers-
Sarah: Mm-hmm …
Sarah: then make that decision. Um, and so it’s kind of balancing the two, isn’t it?
Sarah: And I do remember we tested… We went too far once with our, um, with our branding piece.
A bigger image versus two products per row on mobile
Sarah: So we tested, instead of having f- um, two products and side by side on mobile, we tested one full product image.
Sarah: Mm-hmm.
Sarah: Um, and actually we saw that added more friction than having the two. So- I think the benefits of testing just demonstrate how important it is.
Sarah: You can make a brand decision, test it, and actually- Mm … see that it has a really negative effect commercially.
Matt: Yeah, even if the evidence is, is there, ’cause I think that, going back to your point on the hamper contents, that was the same evidence behind that hypothesis of let’s just showcase what’s actually in those hampers, and actually making the image larger made things worse.
Sarah: Yeah.
Matt: So yeah, we could have very easily just deployed that as a change without testing it, assuming that a bigger image was better. Um, but it’s not always the case.
Sarah: It’s really interesting, isn’t it? Yeah. And it, it is that kind of constant brand versus function- Mm-hmm … friction that I think every business faces into- Yeah
Sarah: but I think a luxury brand faces into more than any other actually. Just making sure that you are creating a beautiful experience, but functionally it’s working well enough to really drive those commercial gains that you want to see.
Matt: Yeah, absolutely. Um, in terms of personalization fitting into that mix, obviously we started very much on the kind of CRO side, looking at optimizing the experience for all visitors.
Matt: That’s shifted over the course of the last six months more into personalization. What are the kind of segments of customers that you’re focused on? Obviously, we’ve, we’ve started to look at international customers as an example. Um- How does that differ for Fortnum than it might do for, for another retailer?
Sarah: Good question. Um, and I think personalization is something that we’ve been really keen to get off the ground, and we’ve been working in the background in terms of tech stack- Mm-hmm … looking at CDP, et cetera, to kind of get us in a better position from a personalization perspective. I think where you really helped actually was helping us reevaluate our current tech stack and looking at where we might be able to optimize what we had already.
Sarah: So we were using Adobe Target. We also had a third-party tool called Ones we were using for a variety of other things. We didn’t actually, um, recognize that we could utilize that tool in terms of real-time personalization on the website until- Mm-hmm … you brought that to us. And so it’s given us a huge opportunity to kick off personalization in a much bigger way, which is very exciting for us. and I think in terms of the kind of activity that’s really worked for us, we’ve definitely been thinking much more about things like new versus returning customers. Mm-hmm. So we know that a new customer, as an example, really does want to understand the brand and wants to fall in love with the brand.
Sarah: I think certainly I feel, the longer I’ve been at Fortnum & Mason, that I really do start to fall in love with the brand, and, and there’s more to discover every single day in terms of the variety of our products, in terms of the provenance behind them. And so I think that personalization piece is just really understanding how we can bring that to life for somebody that doesn’t know us versus give somebody that knows us quite well an experience that’s suitable to them, and also demonstrates to them, brings to life for them categories that they’ve already purchased- Mm
New versus returning customer homepage
Sarah: or new categories in the right way. So the homepage tests and it is a really good example of that, I think, where we updated the homepage depending on whether you’re a new or returning customer. Yeah. We saw some really good results from that.
Matt: Yeah, I think especially for you guys with that new and returning mix because the majority of customers aren’t purchasing weekly on the site, it’s a kind of annual purchase couple of times a year gifting, personalization almost come, becomes even more important because we don’t have a lot of chances to show them that content a- after visit one.
Matt: So, um, everything we’ve done with kind of new visitors has, has been really successful, so I guess now it’s about what do we do when they do come back and, and utilizing some of that technology you’ve, you’ve got around affinity, um, to specific product categories. What have they looked at? What have they searched as well?
Matt: Um, so I think that’s, that’s something we can definitely start to go after.
Sarah: certainly we do have a challenge that a lot of retailers don’t have, which is we are definitely still by some seen as a kind of Christmas brand- Mm-hmm … so people will come to us for their annual Christmas. And we know speaking to customers that they have it as part of their Christmas ritual, which is amazing Mm-hmm
Sarah: and beautiful that they do that. But I think what we’ve been working really hard on is building that all year round business, and so we’ve really kind of nailed all year round gifting in terms of Valentine’s Day, Mother’s Day, Easter, et cetera. But what we’re also starting to see is that we’re leaning into those other kind of key moments in people’s lives.
Sarah: So that might be picnics or dinner parties or events that they’re holding. Um, and so bringing that to life in the testing that we’re pulling together will be really interesting. Yeah. But I think even the test that we ran with the search results actually in terms of kind of personalizing those to, um, categories that customers had shopped previously- Yeah
Sarah: was incredibly successful. So I think that was a great start of 10, and I think we can continue to build on that.
Matt: And like you say, even showcasing categories they might not know exist on Fortnum’s, if they’ve just come for hampers year after year, how do we showcase everything else that you’ve got, um, in that moment as well?
Matt: That’s been really successful.
Sarah: Yeah, 100%, ’cause you’re absolutely right. I think you can, you can personalize to the nth degree- Mm … and actually lose that discovery piece. And since we are such a discovery brand, it’s finding that balance, isn’t it? Making sure that we’re inspiring as well as giving customers what they need or what they might be coming back for.
Adaptive and personalised search
Matt: So I think one of the examples that we talked about around search was we’ve introduced some personalization around what we’re calling adaptive search. So I think the original, um, search functionality that you had has a selection of categories and image tiles that are all static for all customers- Yeah
Matt: which showcase your most popular categories, which makes sense. But I guess going back to what we were saying before about showcasing the breadth of product and trying to become that
Matt: all year round
Matt: gifting, um, that really helped identify by personalizing those based on what customers had looked at before, not necessarily purchased, but just what they’d looked at or navigated to.
Matt: Um, the fact that we could increase engagement with that kind of functionality. So I think previously the click rate within those, that static search was about 1%. Mm. With those personalized ones, it’s 5%. That’s obviously a huge increase, but also what customers were engaging with changed dramatically as well.
Matt: So we were seeing much more engagement with things like picnics, birthdays, occasion-based gifting, rather than just take me back into hampers straight away. So that was really interesting for us as a kind of building block of if we know customers are now interested in those categories, what can we do when they do come back next time- Yeah
Matt: around that. Yeah. So it’s not all just, um, get them back into that hamper journey as quickly as possible.
Sarah: But also a really great example of kind of using that personalization to meet internal objectives as well. Mm. Yeah. So we want to really help customers understand and discover some of the amazing other kind of customer missions that we might be able to serve them with, whether that is a party or a barbecue-
Matt: Mm-hmm
Sarah: or, an event. And so I think being able to both- Really support that, um, that objective in terms of raising visibility of those categories whilst increasing commercial value as well was just a really great example of- Yeah … those two things coming together, which was brilliant to see.
Matt: Yeah. I think aligned to those seasonal campaigns as well is definitely a next step for that, um, to showcase categories customers might not be aware of.
Matt: Yeah. Um, and given your international customer base is, is large and growing, there’s all kinds of things we can start to look at now based on where they are in the world as well. So some occasions aren’t relevant in some countries. Absolutely. Yeah. you know, we’ve got the technology to do things based on weather this weekend, you know, and, and pushing categories in, in the short term as well.
Matt: So yeah, still lots to go after in that area, I think around as well.
Matt: Yeah,
Sarah: and I think international is a huge- Mm … objective of ours as well, so we really are looking at… We, we are kind of slightly constrained by the makeup of the website- Mm-hmm … in that we don’t have individual locales for individual locations across the world, and so we are trying to work with a website where all customers apart from the EU come to one site.
Sarah: Um, and struggling with things like every product isn’t available in every country. Mm. To your point, calendars are incredibly different in terms of even Mother’s Day as an example
Matt: Mm-hmm …
Sarah: is different. We’ve got, you know, different events like the 4th of July in the US as an example, and also different reasons for international customers to come to us.
International personalisation
Sarah: So as an example, in the US, customers come to us because they love our British heritage, and so we want to speak about that specifically to them, and we may do less so for another region. So us working together to really understand those customer requirements and needs and personalizing to that will be a, a great evolution, I think, for us.
Matt: Mm. Yeah. Yeah, I guess on international,
Delivery options in the basket
Matt: one of the things we’ve looked at, um, which we haven’t quite nailed the execution yet on, I think, is delivery information. Yeah. So we’ve done a bit of testing in the basket area around basically showcasing all the available options to customers, which there are a lot, particularly in the UK, Click and Collect, Next Day, Name Day.
Matt: Huge amount of options. Collecting from the store, which I guess goes back to that experience piece as well about people wanting to come in. Um, what we’ve done to date is kind of help customers, um, identify what those options are- Yeah … in the basket, particularly in the UK. But we haven’t seen any movement in conversion, so I think there we can focus more on the, the international piece as well, understanding what those shipping times are, what the cost is, um, just as an informational piece, which we haven’t quite hit yet.
Matt: So- Yeah … I think there’s still a lot we can, we can do in that area as well.
Sarah: I think it’s a good example as well of, you know, sometimes just overwhelming customers at different points in the journey. Mm. And whilst we may be thinking we’re really helping those customers make decisions at that point, I think it just illustrated that they weren’t quite ready for that level of data at that point.
Sarah: And so it might be that us showcasing it later in the journey or in a slightly different format that’s less overwhelming supports them in their future journey. But I think- It just illustrates why testing for us is really important, isn’t it? Mm. We may well have just made that decision and not- Yeah
Sarah: tested it historically, so.
Matt: Yeah, show delivery options at basket makes complete sense. Kind of makes complete sense- Yeah … doesn’t it? But- Yeah … but not for us. I guess c- it’s come back to that, some of the research we did initially as well. There was two very different types of gifters, which we could probably see in our own lives as well, of people who are last minute, don’t care what the cost is- 100%
Matt: I just need it tomorrow. Yeah. Versus planning in advance. Oh, okay, I can order it right up until the day before and still get it. So that’s where the kind of personalization piece can come in, which we haven’t touched yet, I think
Matt: on,
Matt: on delivery overall outside of the basket as well.
Sarah: Yeah, yeah. Mm-hmm. And I do think we’re probably not thought of for those last minute gifts, but actually- Yeah, yeah
Sarah: we’ve just improved our delivery, operationally, so now we can offer, um, delivery right up to, next day delivery right up to 8:00 PM the night before. So it feels like a big opportunity- Mm … for us to bring that to life. I
Sarah: guess to your point, where in the journey does it make- Yeah … most sense to do that?
Sarah: And we can certainly have a look at how we test it.
The future of digital and AI
Matt: So in terms of where you see the kind of digital experience of Fortnum’s going over the next 12, 18 months, obviously the world is changing weekly at the moment- … in terms of tech- Daily. … but also customer expectations of, of what they want from a, from a site or how they’re shopping even, if not through the website.
Matt: How do you see that strategy sort of evolving in the short term?
Sarah: we definitely are, um, out there really looking at how the world’s changing as much as we possibly can. And so just in terms of if you think about search behavior in the last year and how significantly that’s changed, and the requirement now to be highly visible in LLMs.
Sarah: If we think about the long-term changes in customer behavior around customers potentially checking out using agentic commerce as opposed to even coming to the site- Mm … in the first place. Um, if you think about the way that customers might start to interact with our website. So as an example, we’ve already seen in Algolia customers starting to use conversational search as opposed to keyword search.
Sarah: Um- Mm-hmm … and it really brings to life how much that behavior’s gonna change almost overnight. Yeah. And so we’re looking very quickly at tools that might be able to support that kind of search. Um, but what that means is that we’ve never really needed to be more at the heart,
Sarah: of the
Sarah: customer than we have now, and really understanding that customer behavior becomes absolutely paramount because that’s gonna change, not just for everyone, but for individual cohorts as well.
Sarah: Mm-hmm. So we were thinking about this fairly recently, you know, so we’re gonna have… We’re definitely gonna have customers that still want to have the same customer experience that they had, um, and, and come to a website in the same way that they have before to have that confidence and, authority, I guess.
Sarah: Our job then is to make that experience the best experience we possibly can and to personalize that experience so it feels incredibly relevant to them, whilst thinking about customers that may not even check out on the website- Mm … or customers that want a different conversational experience and expect a lot more.
Sarah: And so I think understanding those individual cohorts will become really important, really speaking to those customers so we know what they’re looking for and we can get ahead of the game a little bit. And then the AB testing piece becomes incredibly important again as well. So I think-
Sarah: It’s e-
Sarah: it’s interesting, isn’t it? ‘Cause we’ve thought a lot about tech stack, obviously- Mm … and about what we might need to support that changing behavior. But I think ultimately understanding that behavior first- Yeah … is gonna reap the best rewards, and I think custom- companies that continue to put customers at the heart of everything that they do will be the ones that win in the end.
Sarah: And so as a brand, we will always put the customer first, understand what they’re looking for and what they want, and try to deliver on that.
Matt: And
Matt: as you, you said before, most customers are skewed for you guys more towards that research phase initially than perhaps a typical retailer.
Matt: Absolutely, yeah. So understanding what research they’ve gone through prior to arriving at the site, it’s not just gonna be a Google search anymore. It’s gonna be a 10-minute conversation with an LLM before they even reach the site, if they reach the site. 100%. So understanding what they’ve done is gonna be more important than ever, I guess, rather than just channel.
Sarah: Yeah, 100%. And you might find as an example that, you know, historically direct customers have come with really high intent. Mm. But that might change if they’ve found you kind of top of funnel on an LLM and just coming for a little bit more information as an example. So even understanding channel mix becomes really important.
Sarah: Um, it’s just fascinating, isn’t it- Yeah … kind of how quickly it’s changing, actually. Yeah. And for us, it’s thinking about marrying up how we can be highly visible in an LLM environment and make sure that, as an example, on our PDPs, we’ve got enough information to make sure that, that a bot can read and can understand or an agent can read, um, but still maintaining that customer experience.
Sarah: Mm. So an example for us, um, is on PDPs, we have a lot of information that’s concertinaed up- Yeah … um, which really supports customers because we know then they can very easily click down to see the information that they want rather than scrolling throughout the page. But agents can’t read that information- Mm
Sarah: because it’s behind a dropdown. So we’re having to really think about how we, um, design pages to suit both needs in the future, and then A/B testing becomes really- Yeah … important because we’re gonna make some quite radical changes- Mm … where historically we might not have needed to. Um, and really understanding the impact of those is incredibly important.
Matt: Yeah, I guess it goes back to that test we were talking about at the beginning around showcasing what’s in the hamper on PDP. We’ve done that successfully for customers shopping on the site. At the moment, though, probably LLMs aren’t able to access that data, so what can we change about that to improve that process?
Matt: Yeah. Um, as you say, while still making it a great experience for those that are still on the site. Yeah, exactly. Um, And I think we can st- we can use the research with those customers offline as well to, to kind of influence how we, how we do that ’cause I think there’s… Obviously it’s still a customer at the end of the, end of the day looking for, looking for the product.
Matt: Um, but just adjusting it based on the channel they’ve, they’ve come in from as well.
Matt: Cool.
Closing advice
Matt: So based on what we’ve just talked about, what would be your kind of advice, recommendations for retailers, either in a similar space to you or thinking about getting started into experimentation, given how fast things are moving at the moment?
Matt: What are the things you would, you would be talking to people about?
Sarah: My personal view is that there’s never been a time when you don’t need experimentation more. Mm. And I think if you start to think about … I mean, I know we’ve talked about LLM traffic and what that might mean and changes behavior in terms of search.
Sarah: I think we’re also seeing that, uh, across social as well, so we’re seeing more and more customers use social as a search tool. Mm. I certainly know, if I think about my children as an example, that they do not use Google at all. All of their search is in social, which is fascinating. Yeah. And so us thinking about how do we experiment based on where customers have, have, um, discovered you is, is really important.
Sarah: But I think also if you think about the way that agentic is starting to impact or will impact websites themselves and how easy it is to move through a development cycle. So I’m sure that in the next couple of years, my trading team will be able to release something just with a simple prompt. Yeah. I think it’s then understanding how do you put the right systems in place in your business so that everybody’s aligned in terms of what are we trying to un- what are we trying to identify, how are we trying to improve the customer experience, and what takes priority?
Sarah: And then I think having that laser sharp lens around return on investment, so I know that- Yeah … one of the other things that you’ve brought to the table is that kind of real- really clear view of priority based on commercial value and what we’re gonna see in terms of return on investment. And I think in the new world, that’s gonna become even more important because you’ll be able to test iteratively much more quickly but also be able to release much more quickly.
Sarah: Being able to set priorities in the right way will be really important. And then going back to that customer piece that we discussed before, really working with a team to understand that shift in customer behavior at a cohort level will, will be a game changer for most businesses, I think.
Matt: Yeah, I think
Matt: it
Matt: re- it reduces the risk certainly in deciding on a test because the, the development cycle is just so much shorter.
Matt: Yeah, absolutely. You can build something in an afternoon rather than a week, for example. So I suppose then it just becomes about how do we empower your team, I guess, to, to prioritize those- Yeah … ideas, um, because it’s not gonna increase your traffic overnight. So the, the number of tests you can run might remain the same, but now you can build 50 tests where you could have built five.
Matt: So how, how do you decide what to test when? And it, it’s a good point about the, the team also releasing that themselves of that, governance, I guess, around who-
Matt: 100%,
Matt: Yeah … who can launch what and when.
Matt: Yeah.
Matt: Are we having conflicting experiments at the same time? That becomes even more- Difficult than it is today, I suppose- Yeah
Matt: now that you can do anything.
Sarah: Yeah, so I think effectively managing that across- Yeah … the business and building in the right kind of ways of working operationally will become really important as well.
Sarah: And then kind of coming back to the custom point,
Sarah: really understanding and working with a partner like yourselves to get underneath- those customer changes becomes really important.
Sarah: And not just for my team, but for the whole business, because With customer expectations changing so rapidly, the entire business is going to need to be agile enough- Mm … to understand that and move quickly enough to deliver
Sarah: the
Sarah: experience that we as a brand want to give to our customers.
Sarah: Um, you know, we’re a 300-year-old brand who will need to move ever increasingly quickly to continue to deliver the experience that we want to for our customers. Um, and it feels incredibly exciting, but also I think we need to maintain or, or we need to ensure that we have that level of agility within in- individual
Sarah: teams.
Listen on Spotify, or join the Beyond Conversion list to get every new episode by email.
How do you improve conversion for a luxury brand that doesn’t use countdown timers?
Sarah Wilkinson, Director of Online at Fortnum & Mason, joins our Head of CX, Matt Maidment, to talk about testing at a brand over 300 years old. They discuss how the team tests what helps customers make big decisions, from showing what’s inside a hamper to a bigger image that added friction, and what AI search could mean for how customers shop.
In this episode.
- Why Fortnum & Mason doesn’t use countdown banners, and how the team still improves conversion.
- What customer research found when shoppers didn’t know the word “wicker”.
- Why a bigger product image on mobile added friction, and why showing delivery options in the basket didn’t move conversion.
- How the team personalises the homepage and search for new and returning customers.
- What AI search and agentic tools could mean for online retail, and Sarah’s advice on experimentation.
Sarah Wilkinson, Director of Online at Fortnum & Mason
Matt Maidment, Our Head of CX