[Live webinar ] Loyalty Meets AI Agents: The Next Growth Loop

Hi everyone. Thanks for joining us

today. If you've been following AI over

the last few months, you've probably

heard the term agentic AI or or um

aentic commerce and it comes up more and

more often. If you strip it all back

though, the shift is quite simple. Seems

that we're moving from a world where

customers make decisions to one where

software increasingly makes decisions on

their behalf. And that's the real

change. AI is no longer just analyzing

or recommending. It's starting to act.

And that's where the idea of agentic uh

commerce really starts to matter. To

explore what that really means in

practice for those of us who don't get

to spend five or six hours of our day

delving into the realms of AI. I am

joined by Mark Galvin, partnerships at

EMA Blue Reach, and Attilla Catchmar,

CEO of Antabo. They've both been

immersed in AI since its early

development. and if I ever want to know

anything about it, they are the two

people that I go to. So, we've brought

them here together for you today. Um,

and I guess a real question for brands

and something that I really want anybody

watching this to do, please put your

questions into the comments. We want

this to be really useful. Um, so please,

if you've come to this thinking, I want

to go away knowing this today, please

pop it into the comments and questions

for us to answer for you. But the real

question for brands is if an AI agent is

making the decision, how do brands show

up in that moment and what can it

actually influence the outcome regarding

the customer? Um, so over to you Mark. I

know you know so much on this subject.

So I'm going to try and uh kind of chair

this for you and Natillaa, but please um

tell us a bit more about what agentic AI

is on the the surface of it.

>> Sure. Sure. So um I think we're probably

going to drill down on this topic. So

I'll start quite high and we can go more

detailed as we move in. Um but from my

perspective, agentic commerce is the use

of software agents that can act on

behalf of the customer across parts of

the shopping journey. Uh most commonly

including things like the product

discovery, comparison, the products,

pricing, comparison right the way

through in some instances to checkout.

um and how that experience works across

LLMs and agents is kind of varying. But

in practical terms, just to give it uh a

non-technical kind of positioning, uh a

consumer might actually go and prompt an

agent with something like find me the

best black leather ankle boots under

£180 available in a size five that can

arrive by Friday and give me the best

loyalty or member value. And then the

agent will go and handle several steps

that previously required the shopper to

compare and decide manually. And they

have to go through a number of websites,

forums, review sites, all sorts of

things to be able to come to that

decision. So uh it's really um moving

classic e-commerce personalization

uh to a point of autonomy and uh if you

think about the traditional cycle was in

e-commerce was around reacting to clicks

and page views on a branded interface

such as the website to be able to build

that. Aentic commerce from my

perspective actually introduces a

delegated decision making process where

the customer increasingly asks software

to do work for them. So it's doing a lot

of the heavy lifting and the effort and

connecting with the right brand and with

the right product from that brand.

>> I'm on mute.

>> Yeah, maybe I'm adding some

>> Yeah. Do you want to go ahead

>> to this? So let's make it as a

discussion and hi everyone. So I think

uh the way I would sort it, we are

living in a very transformative era and

um if we think about AI which is around

us for like three years actually three

and a half years from uh from now AI is

a technology but the agentic is the the

actual solution that we are experiencing

and going to experience which will shape

our life more and more down the road and

you know agentic I know it's everyone's

hearing about this but basically agentic

is making decisions behalf of the

consumer, behalf of us. That's why it's

so uh different than everything else

before. And just think about this when

everyone went from offline to online

that was a huge change, huge, you know,

um uh directional shift. But I think the

agent is even bigger than this one. And

the reason for that is because when we

went from offline to online, still the

center of the decision maker was the

human. But in this case it's going to be

the agent that's why it's even bigger

shift than the previous one. So and uh

the technology is shaping and challen

developing so fast. So therefore we and

everyone else should uh take a look and

spend lots of time on this technology

and this game changer solution.

>> Great. Thank you. Um so we've touched on

the surface of that. Um, and hopefully

everybody has an understanding of what

agentic AI, agentic commerce is. Um, I

guess really now what do brands need to

start doing now? Because I think like

there's this concept of okay, that's

where you say the future's going to be,

but how do people get started on it? You

know, how do you what's the first steps

to take um to make sure that they're not

getting left behind somehow?

>> Yeah, sure. So I I think um the first

thing is that they've got to start

showing up and in order to show up

that's going to

evolve quite considerably uh the the

stack that they use the technology stack

that that's powering them uh and and

enabling that showing up. I already kind

of mentioned that we're in this kind of

point where we're shifting from browsing

to delegation. So instead of I'll go and

search your website, customers are

increasingly going to go and ask an

agent and that agent's going to perform

a number of different tasks, not just

searching for a product then looking at

reviews and looking at other

information. It's going to combine that

data together automatically and simplify

the process for them. I already myself

personally have moved away from Google

browser terrible as it sounds into this

world where I can do I can do tabular

analysis of all the products that I'm

looking at and compare them side by side

who can deliver the fastest and which

ones have got the best reviews right so

and I think I think that's kind of

shifting quite considerably but the real

question from an architecture

perspective and a stack perspective is

what makes it possible and how do brands

start considering this stack in terms of

how they show up So the thing that makes

it possible is that there are various

protocols are out there. There's there's

actually eight in total but there are

three key ones which I think are there

today that are being widely adopted. The

first one is something called UCP and I

hate acronyms so I'm going to actually

I'll mention them but then explain what

they are because I think

>> there are a lot of acronyms where AI is

concerned. They're very similar as well

so it's easy to get them muddled. So

please do explain when you use one. Yeah

>> exactly that. So, so yeah, UCP is

universal commerce protocol. Uh, in

simplistic terms, it's an emerging way

that compliant agents can discover

products. They can build carts and they

can transact across merchants. You've

got ACP which is the agentic commerce

protocol and that is one of the models

being uh discussed for agent-driven

shopping flows particularly around AI

assistant ecosystems. And then the one

which is actually the fastest trending

one at the moment when you compare all

of these is MCP

uh which is the model context protocol

and that is a way that LLMs uh which is

like chatbt the things that we're using