Episode 5: Contextual Marketing with Kenzo Selby from GumGum

Welcome to the Location Insights Podcast.
Guest: Kenzo Selby (Managing Director, GumGum Japan)
Host: Kevin (Global Marketing Manager, unerry Inc.)

GumGum Japan: https://ja.gumgum.com
GumGum US: https://gumgum.com

In this episode I speak with Kenzo Selby of GumGum Japan. 
Topics Covered:

• From Contextual Targeting to Mindset Marketing
  Kenzo explains how GumGum has expanded beyond keyword-based contextual advertising to include signals such as article topics, images, video, sentiment, timing, attention, and user mindset. 
• Measuring Online Advertising’s Offline Impact
  The conversation highlights how unerry’s visitation analysis helped GumGum show that digital ads can drive measurable store visits, creating a clearer link between media campaigns and real-world outcomes. 
• Shabu-Yo Campaign Result Case
   Then our discussion explores the joint GumGum and unerry campaign for Shabu-Yo, including how rich creative, food-related context, value messaging, and location data contributed to strong customer acquisition performance. 
• Japan’s Strong Offline Retail Culture
  Kenzo shares how Japan’s high-quality retail and restaurant experiences make foot traffic measurement especially important compared with more digital-first markets. 
• Beyond Simple Geofencing
  Our conversation looks at how movement intelligence can help brands understand how far people are willing to travel for the right retail experience, rather than limiting campaigns to narrow geographic boundaries. 
• The Future of Outcome-Based Media Planning
  We discuss the potential for location data, POS data, and third-party measurement to support more outcome-driven advertising models tied to visits, sales, and customer journeys.

Like to be a guest on the show or have questions drop us a mail: [e-mail] 

Creators and Guests

Host

Kevin

Global Marketing Manager at unerry

Guest

Kenzo

Kenzo Selby is the Managing Director of GumGum Japan

What is Location Insights?

Welcome to the Location Insights Podcast
Where we bring you information, news, and insights from Partners, Clients, and Data Scientists about: Location marketing, Retail Media, POI, Data Analysis, and other interesting topics.

Hi, and welcome to the Location Insights Podcast.

I am your host Kevin, the Global Marketing Manager at unerry Inc.

unerry is a location analytics platform and data insights company based in Tokyo, Japan.

And with me today is a colleague and member of the Gum Gum Japan organization.

Gum Gum provides contextual marketing services around advertising and has been working closely with unerry in of late.

Welcome to the show, Kenzo Selby.

He's the managing director of Gum Gum Japan.

Welcome.

Thank you so much, Kevin, for having me on.

Great to have you.

So why don't we start with some background, introduce yourself.

How did you get into the marketing industry?

What kind of background do you have?

Yeah, so actually I had a fairly

late entry into the marketing industry, so to speak.

I was actually raised in both Japan and Australia with all of my school years being in Australia right through to university.

I actually worked in Australia as both an accountant and then into digital media after about six years of working in liquidation and turnaround as an accountant and then I became a tax accountant after that

I think those were definitely some of my uh formative years in terms of getting an idea of how businesses work, going out and speaking to business owners.

I then actually joined AOL when they had a a

product which was a viral video syndication network.

And that actually took me to Japan.

Yeah.

So that was the first time that I was really working in Japan.

I think that was in

early 2014, so that was just over 12 years ago now.

Blows my mind.

Worked for their product was called Beyond at the time.

I did that for about two and a half years and they actually brought me over to Japan uh and so they said, look, you know, we we wanna grow this business within Japan as well.

And then about a year and a bit later, I got asked

to join Teeds, which was then was another ad network, but it wasn't, it was very, very different in the sense that it was really promoting video

on mobile phones, which was quite quite a new thing at the time.

So I think people obviously had forgotten just how revolutionary that was at the time, but it was really it was fantastic to be part of an industry that was moving so fast at the time.

and I was with them for about seven years a ride until the last four years where I actually headed their sales team in Japan

So that became a very, very significant business in this so-called rich media ad network space here in Japan and the rest of the world.

And because of the success of that business, I was lucky enough to be called upon gum gum.

which had been in Japan for at the time about six years and they said, look, you know, we've got some exciting things on the future.

And I've been with them for the last two and a half years leading all of their efforts.

So we have a small but mighty team here in Tokyo

pushing a lot of not only open web browser formats or creative formats from there on, but we also are now coming into the CTV space and we're actually seeing that mindset thinking

which we've recently evolved to is really showing up in other interesting areas like UNERRY with their location services.

Wow, that's a great background overview there.

If I can just take a step back and pick your brain a bit.

So for the video side of things, was that like early iPhone or early Android days, that kind of time period?

Or was it like

Android had already been around for a while.

Yeah, so actually on the videos side of things, I think there was too big a difference.

Up until then, if you can we're talking about at least 12, 13 years ago, I think at the time most advertising was what I would say is a static banner, and it wasn't particularly interesting, especially on the open web

at the time.

And what we were actually able to be part of was see video be syndicated not just on PC but on mobile.

And that shift really started to happen during about, I would say, 2016 in Japan, maybe a year earlier, uh in the UK and and the US.

and maybe other areas from there on.

But up until then, the main way that video was syndicated on advertising was on a pre-roll.

So that, you know

uh most publishers would have content that they would run on their site and then before that they would have so-called pre-roll content

But at the time, obviously we didn't have internet speeds as fast as they are now, so most people would obviously connect to Wi-Fi to watch that content.

I think it was a combination of the fact that

the technology, the infrastructure had really evolved at that time and we were able to really syndicate a lot of video advertising, you know, within just standard articles and ads through that time.

So yeah, it was

Definitely an exciting time to be in adtec and I'm glad to have been part of it.

Wow.

So were the commercial types like

Movie trailers or more like TV ads that were redesigned for the mobile format for the mobile screens

Yeah, so it was mainly TV commercials at the time, so they didn't actually have anything particularly specially made for the formats.

I think at that time, you know,

If you think about social media at that time they were just starting to come up in in Snapchat and obviously you know Instagram came in at that same period.

So

we were finally starting to see the so-called vertical video format come through and really use the the mobile phone.

But at the same time was a lot of the same content that was being used on television content effectively blown up a little bit for the vertical format.

Yeah, that makes sense.

Better to repurpose and reuse the contents, the media that you already have, rather than shooting something from scratch.

Makes sense.

And then if we pivot into your current role in Gum Gum, so Gum Gum is all about contextual marketing.

And for listeners, in simple terms, how would you define what contextual marketing is?

And what do companies use your services for?

Yeah, so I think you know GumGum started off as a company that was able to scan and analyze

articles, not just what was written in the articles, but look at all different points.

So obviously if you had an image, it would also scan that image and get an idea of what was in that image.

So

One of the I think misunderstood concepts about contextual targeting is it's only looking at like keywords, for example

But languages are very nuanced, and especially, you know, uh with greater uh understanding through AI and all of these LMs at the moment.

understanding languages become, you know, has grown hugely in the last 15, 20 years.

I mean, I think when I first came to Tokyo, we were still using Google Translate and especially with languages like Japanese and Korean

the translations were almost unusable in most most circumstances.

And so I think contextual targeting was also, if if you think about it, was also

have had that evolution go through it to make sure that it understands exactly what the environments, what the con you know, what exactly is the key points of

articles on the open web, but it's not just limited to the open web.

If you think about it, anything can be potentially contextual.

You put your mind to it.

I mean some of the stuff that we're obviously doing at Gum Gum is not

just on the open web, but looking at you know scre screen CTV, for example, content on there to kind of get an understanding of what that content is.

And then

serve an ad which is most appropriate for that content at that time.

So one of the key things that, you know, which is very, very simple from something that I had in my upbringing is whenever we would see, say, a cooking

show for example, you'd get served perhaps like an ad for anything like soy sauce or anything food related

And it would kind of come up at that time.

And that is contextual targeting as most basic way.

So at what we're effectively doing is we're

with the uh benefits of technologies, we're actually able to scale that what was once a manual process and make that, you know, use it all across the web, use it all across

different formats, CTV, for example, but it's not limited just to that.

I mean if you think about it in terms of uh

even out of home, for example, you could potentially have more scalable architectural ads that would deliver ads that would be correct for that current, how can I say, buildings or or um

any kind of big stadium around it that you you have all of those ads which are relevant for that particular event.

So for example, if you had a big soccer event, you know, obviously Adidas would like to play ads quite close.

for that stadium, but at the same time, if you had other advertisers that were also interested in that area, you know, contextual targeting is is a really, really viable option in terms of ensuring that you get the right message at the right time.

Now we've actually evolved from just being a contextual company.

Gum Dum actually rebranded to the mindset company actually about a month ago.

And the mindset company is

making sure that we still keep our contextual targeting as a big identifier, but we also look at other signals as well to get more of an idea of the user's mindset at that time.

So for example, we look at things like article topics

you know the videos, images, keywords, sentiment, environment, and other things to look at the intent of that user at that time.

So

You can look at time, for example.

We can look at attention metrics.

We can look at how much the user is actually reading the article, whether they're reading it slow, whether they're reading it fast.

There's all these other signals that we can actually make sure that we have a stronger image or identify exactly where that user's mindset is so that we can display exactly the right ad that's going to deliver the strongest message at that time

So I think it's really simple in the sense that with contextual targeting, I think it was one of the earlier parts of our service and it's been the most fundamental signal that we've

continue to use, continue to advance.

However, I think the goal is the same, is to ensure that our brands appear in the most relevant moments and where those consumers are most receptive.

So I think that's been the big change, yeah.

Wow.

Yeah, that that covers a lot.

So it's it's all about the

Customer at the right time, the right place, the right message, and taking in all those different signals that the customer is

consuming, let's say watching videos or uh reading blog posts or reading tweets and Instagram posts and taking all that and making sense of it, saying like what is relevant, what is most interesting.

for this customer.

Are they interested in the sporting event that's going to be happening this weekend?

Are they interested in the music event that's going to be happening across the city on the other side of the city?

which of them is a better targeted ad for them.

Correct, correct.

So it's going a bit deeper than just that moment because as human beings, we're definitely on the move.

We're definitely uh, you know, certain times I mean even today, um woke up and, you know, it was raining, or was looking like it was gonna rain.

And by yesterday it was raining a little bit as well.

And when you have rainy days, you tend to be a little bit more gloomy

And you just don't have that switch.

And so Jim and I started running for about two or three minutes and I just said, look, you know, I've had enough.

I'm going to move to walking

Uh people obviously change with their moods, they have mood swings during the day, then had to uh get changed, come to work

and change my mindset again.

But one of the things that's obviously really important is in that mindset, given the circumstances, or given a lot of circumstances, you can really change on, you know, if you just

generally tired at the time, but you can serve have the right ad serve to you um potentially, but if it doesn't take into account all those factors

then the ad will just simply be another one of the 3,000 ads that that hit users every day that doesn't get noticed, that gets completely forgotten.

And unfortunately for the brand is it's just a waste of money.

Right, right.

Yeah, I think that raises some important points about the right timing for the customer, you know

And I think that leads into our recent press release that Gum Gum and unerry jointly announced.

So recently we had some exciting press releases, both in Japanese and in English.

So I'll kind of highlight the Japanese one with the Shabu Yo from Skylark and what Skylark said is that through this collaboration between Gum Gum and unerry, they were able to verify that

that they were able to achieve a cost per customer acquisition that surpasses that of social media advertising.

So through the use of Winaries location data and analytics and

Gum gum's contextual marketing skills and targeting, they were able to reach customers better.

Can you share some of your thoughts on this

Yeah, well I think before I answer that, my family and I are are big fans of Shavuyol.

It's the hot pot uh restaurant which is all through Japan.

I very much recommend it

Um for any listener that's you know interested in trying what I would say shabbu shabbu it's uh a very, very kind of light pot.

You know, you've got a range of sources and and the kids love it.

you can drink and you've got all sorts of desserts as well.

However, this was um particularly important for us as a business.

So being in the what I would say is ad network open internet space, what we've seen in the last few years

is that a lot of advertisers, a lot of advertiser dollars have been moving towards social platforms and uh, you know

the the big major platforms.

Um I think a lot of advertising agencies have seen that the open web is a lot more complex, there's a lot more players to it, and in the end a lot more diverse.

Diverse.

A lot more diverse.

Yeah.

There's a lot of different platforms.

There's a lot of different signals.

Which ones are more important to take in and capitalize on and show the right message to the potential customers?

Totally, totally.

And and because of that, um what we've we've struggled to uh prove as a business is is we've delivered strong results for our uh brands and agencies on brand lift studies.

But that hasn't necessarily shown what every advertiser wants to see is they want to see their revenues go up.

They want to see what's driving, what's pulling, you know, um

bringing clientele to their stores.

They want to see what what factors are really pushing the needle.

for their revenue.

And so one of the things that we've always struggled with is this measurement and how we then display or how we show the value of our services

So I think one of the proof and the pudding was when we actually ran this campaign with Unetti and for Shab.

We were against all of the social platforms, which I would say were were fairly significant, not just in uh

global social platforms, but also the Japanese local platforms, which are seen as the what I'll say is the stronger players within this market because they offer

you know, a range of discounts for retail companies and they have a much bigger um uptake and and especially in rural Japan, some of the social platforms

that we're actually using this this campaign are particularly dominant because not only are they made for Japanese locals, but at the same time the services that Shabuyo has

a lot of their stores are out in remote locations.

So this was particularly important for us to show that we were, you know, our mindset or our way of ensuring that we

could get the brand's message to people in a receptive mind was proven.

And so the one thing that we're at

to see from these various fairs that Shabby all runs.

So they they have fairs, for example, like they have like Korean fairs, so it's it's Korean noodles.

So every month for month and a half they run these fairs and

we happened to run a few of these campaigns with unerry for the last I think probably about

eight or nine months and we've constantly seen these just fantastic results where Gum Gum has been particularly strong on that mindset

philosophy in terms of ensuring that we get users who are hungry who feel like they want to eat shabu shabu at the right time, shabbuyo at that right time.

um and and deliver that ad.

And so we used an advertisement which was an all-screen advertisement, which we call our hangtime unit, and it featured heavily the actual ingredients that Shavill uses.

So Shavill uses

obviously a a range of meats and vegetables, but it also definitely puts an emphasis on the fact that it gets um its vegetables from certain locations in Japan, that kind of thing, and and really it shows the freshness of it because otherwise

It's only lightly seared when they actually do a lot of the the cooking in this uh shabu shabu um cooking.

And so in that respect

uh it was it's really important they show really fresh ingredients.

And so we had that within the advertisement that we actually have to display.

through our rich media advertisement.

And then we actually then made sure that we displayed the advertising in relevant locations.

um some of them being uh food and and drink around those uh areas as well.

But some of the other locations which was particularly interesting was

um talking about saving money was another category which came up as really really strong simply because people in Japan you know at the moment we've got a weakening yen

Shabuyo is is really, really well priced.

It's it's about, I mean, if you put into US dollars, it's it's about $20 for all you can eat.

for an adult for all you can drink as well.

So it comes off as extremely cost effective and really important for a lot of families that are budgeting in this in this country.

It really showed

whether it's the open web against social platforms, it really demonstrates that if you do good advertising, you can really drive real-world results.

And that's beyond just clicks and impressions

I think that's really interesting kind of insights about this particular campaign.

And it's it's great it emphasized that the food and vegetables and ingredients were locally sourced.

And I think that also helps incentivize the customers to come and support the store, help their local farmers in the community, because they're getting their produce and they're getting their ingredients from the local farmers.

So they're supporting their local community and they're getting a great deal as well.

Yeah, yeah, totally.

It's uh I think one of the things that makes this country really

special in a sense is there's definitely a very, very strong sense of using local ingredients, but at the same time um ensuring that that freshness and even in so-called what I'd say is budget

restaurants as well that the produce is still of high quality.

Right.

Yeah definitely.

And so through this

campaign and this collaboration that we've had, what kind of strengths do you see that UNERRY's data brings to provide these solutions in this campaign?

Yeah, I think, you know, the incredible thing about unerry is just how easy it is to scale.

So just to give you an idea, I think Shabuyo has close to 400

stores just within Japan.

But one of the things that was so easy from our perspective was working with

uh Shabuyo and the agency and been able to make sure that we could show all the results for all of the uplift in terms of shop visits throughout the country.

So Japan has got um

you know, 400 stores, so it's got a significant volume of retail stores for Shabior alone.

But it's also got Shabuyo in locations all over the country and some of them

can be fairly distant from other places, which is quite a bit easier to track.

But we also saw that we had no problems with tracking

those Shabuyo stores within what I would say is department stores where you had huge amounts of people going through at the time.

So one thing that was really key to that was ensuring that the targeting of the location services

was high and you know Shabuyo was that that was coming from the advertiser was obviously working with unerry before they started working with ourselves and they were also

already at that time very very satisfied.

So they were using it across all of their media, not just us.

So from our perspective it was a very, very fair

comparison between the different media.

So if I compare that to some of the other ways that we measure media effectiveness, for example, we cannot, for brandly studies, for example.

we cannot use the same method to compare social media platforms to gum gums, you know, media effectiveness for things like brand awareness.

But for unerry services everyone had the same measurement objectives.

They were measured in a way

exactly the same.

So it was comparing apples to apples rather than apples to oranges, which I sometimes find with the other measurement partners.

So that was definitely

really, really important to us.

But I think the other thing that was so key to that was because it was so fair to have those results

was such a strong, you know, it was not only encouraging for my business to say that we were doing really good things, but we have a lot of providers that say that they can do the same things for a lot cheaper.

However, cheap doesn't necessarily mean better.

And and one thing that we've always done at at Gum Gum is we've helped brands and advertisers with the creative.

We've created creatives which are not just static that have

uh movement that can really try and grab people's attention.

But we've always struggled with saying, you know, whether people are actually buying from our ads because we tend to be a little bit more expensive.