The Future of Shopping
How personal AI agents and data vaults will transform consumer behaviours, brand value, and experience strategies.
Goodbye Status Quo.
Also: Read to the end to see why you shouldn’t always follow best practices
When AI Agents and Data Control Weaken Traditional Marketing..
You’ll need to rethink:
Who you are
How you do it
What you mean to a consumer
Shifting consumer power is driven by 4 accelerating trends:
The rise of autonomous AI agents for consumers
The breaking of the traditional purchase funnel
The development of personal data vaults
The exponential speed of AI agent’s capability
I’m focusing on the consumer/brand impact, leaving the technical stuff to others more expert than me.
Intro
When it comes to tech breakthroughs, you may feel like the future is here. But it’s not. Next week it’ll be there. And in a few months, somewhere else. Now, it’s exponential.
ChatGPT and other current foundation models are great.
But although tireless and enthusiastic, they need guidance. They require direction, monitoring, and vigilance against half-truths.
And there’s one challenge they’ve yet to conquer - efficiently completing a task made up of distinct steps, without you.
That is changing.
TREND 1. Step up, autonomous AI agents.
Consumers will have Personal AI agents to do things for them autonomously. They will take their instructions, break it into tasks, tackle them step by step, store and learn, and reuse for new steps in the process. And that includes buying things.
And these agents will not just be smarter versions of today’s models.
They open up a new dimension: ‘Slow thinking” (Kahneman’s ‘System 2’) as they ‘think’, solve complicated buying tasks, and crawl to their goal.
It’s a fundamental change in how you’ll interact with consumers.
These AI agents will use their owner’s personal preferences and data when buying things.
(I say buying to make a distinction between a process that we want to run smoothly and requires as little of our attention as possible, whilst shopping is aspirational, for fun, for experiences, for the self, and more exciting, with different behaviours, and which when freed from buying will become more focused).
An Agent will be a new kind of consumer.
And just like any consumer, you’ll need to learn their capabilities, preferences, and build relationships, demonstrate value, and earn recommendations.
TREND 2. The Fluid Purchase Funnel
The traditional purchase funnel - the stories and narratives of your brand, the value and emotional investment you create as your consumers travel through awareness, interest, desire, to action - won’t protect you against the objectivity of agents, because the traditional funnel largely doesn’t exist anymore.
The funnel has changed.
It isn’t linear anymore. Gen Z and Millennials have already stamped their behaviours on our status quo.
They total more than half of all global consumer spending power. And these behaviours are leaking into other demographics, too - including Boomers.
Now, the funnel is an endless loop of inspiration and exploration, and it skips much of the brand connections we rely on to drive conversion by shifting it post-conversion.
When over 50 per cent of global consumers are doing most of their brand research after they buy, and 78 per cent uncovering things that attract and make them loyal to a brand after purchase, pre-purchase is increasingly dedicated to price and feature research.
And this is the buying stage where autonomous agents will do most of their work.
Instead of vying for consumer attention through your advertising and marketing promises - a one-to-many model - with the loop you must now centre more on the customer and their specific needs and wants - a many-to-one model - with loyalty experiences focused post-purchase.
Back at the top of the funnel, your relationship with your consumer will be mediated by their AI agent, a gatekeeper filtering out anything and any noise that doesn't meet their buying or data-sharing criteria.
It will use its ability to complete a transaction and restrict the sharing of data to the minimum necessary. In the process, you’ll lose access to much of the customer data you've relied on up. On the flip side, if the consumer permits you data access, targeted marketing and personalised offerings become 1-1.
TREND 3. Personal Data Vaults
This is because Personal Data Vaults are coming - owned and managed on the individual customer side - that will empower consumers to create new, disruptive, digital relationships. That’s when consumers can then decide what to share with you, and in what context: ‘You can only have this much information now, for this action, and I will revoke it when completed’.
TREND 4. Exponential Capability
Things change when AI assistants develop what some call AQ, their “actions quotient.”
This is their ability to get stuff done on autopilot in the digital and physical world.
Right now the spotlight is focused on a new tool called Assistants API. This OpenAI tool empowers applications capable of understanding natural language, executing functions within their apps, and utilising services with cross-collaboration. Assistants become agents, communicating with each other to complete tasks.
A consumer’s request to book a holiday will trigger a series of coordinated actions between multiple AI agents: one to book a flight, another to secure hotel reservations, another to arrange transportation, and another to plan activities.
++++++++++++++++++++++++++++++++++++++++++++++++++++++
How Soon Will This All Happen?
To build a truly personal customer AI, the customer’s agent must be able to access its user’s data. Otherwise, the AI will remain a generalist, not a personal agent.
OK, you ask, but how soon is soon?
If someone did nothing but read 24 hours a day for their entire life, they'd consume eight billion words.
That's a lot of words.
The most advanced AIs today consume more than eight trillion words in a single month of training.
And it’s not just set to continue - it’s exponentially accelerating.
November 2022 saw GPT-3 launch. GPT-4 was March 2023, already Claude 3 is outperforming it, Meta with Llama 3 is expected in weeks, and GPT-5 in mid/late 2024 will be a big jump, far better at reasoning, and much more capable at performing complex, multi-step such as finding and booking flights, hotels and transport arrangements.
OpenAI is promising GPT-5 will be ‘an amazing new model’ this year and ‘do many different things’. It’s likely to have autonomous agent features, and Microsoft’s move into ‘Consumer AI’ means autonomy will be baked into its consumer products.
Critically, GPT-5 promises 3rd party integrations. So, our autonomous agent could buy groceries based, say, on our budget or dietary preferences - auto-searching for recipes, buying the ingredients, and having them delivered by working with all the ‘agents’ on the brand side.
Apple and Google are talking about integrating Google’s Gemini into the iPhone. This will be a massive move - bringing advanced AI capabilities on-device as soon as iOS 18, due soon.
At least 100 serious projects are working to commercialise agents, said Matt Schlicht, who writes a newsletter on AI.
Combined, all 4 trends will transform brand and consumer interactions.
And what happens then?
Then, you’re on the cusp of the ‘Intention Economy’ where consumer/brand relationships reverse and your CX and marketing must be different.
You’ll compete to meet the customer’s intentions, not to grab their attention.
The Intention Economy
In the ‘intention economy’ consumers will use the tools together to:
Build personal loyalty programs with you
Dictate personal terms of service to you
Tell your whole category what they want, how they want it, where and when they will take it, and how much they will pay for it. You may have to ‘bid’ for the customer’s intent.
Doc Searls calls this The Intention Economy in his book of the same name.
In it, he describes an economy where the consumer is in charge, one where brands must respond to the actual intentions of a single customer instead of vying for the attention of many. As power shifts to the consumer, it’s logical they become more independent and powerful.
Instead of brands vying for attention, they will compete to meet the specific intentions of each customer. The customer's agent will negotiate for the best offers to meet their needs.
Imagine you’re landing at an airport. Your agent connects to multiple airport car rental brands requesting responses with the best offer (price, terms, loyalty rewards, etc) in readiness for your arrival.
Imagine agreeing on a personal loyalty programme with you, in return for shopping with you exclusively in your category; or agreeing personal terms for the service/s levels they want in return for exclusive purchasing of your product or service.
(More on The Intention Economy in a later newsletter from me)
Do Nothing and Emotional Connections to your brand will weaken.
There are two base types of loyalty:
Attitudinal. Loyalty driven by a customer’s brand preference. Usually rooted in emotion.
Behavioral. Loyalty driven by a customer’s actions. Usually rooted in repetitive behaviors, simplicity, product placement, convenience, etc.
How have you built your brand?
Are your loyal customers emotionally attached or behaviorally attached?
If you’re a purchase in a category your customers use regularly and don’t feel emotionally tied to your brand, you become an easy target for them to trade down from, or for their agent to ignore.
The Threat of Commoditisation
With the rise of agents, if you didn’t already know, you’re going to discover that your customers are not all faithful.
Often, the assumption of loyalty is flimsy, based on a combination of consumer laziness, convenience, your brand placement power, and your brand salience.
The thing is the mental shortcuts consumers take when buying won’t overwhelm the AI agent. It will choose the brand after comparing prices, delivery times, product sources and other preferences. Emotional brand persuasion will be diminished. AI agents aren’t lazy. They don’t care about the shelf placement. Convenience isn’t a word they understand. Salience isn’t an objective criterion.
The consequence?
Emotional brand persuasion will be diminished, sometimes irrelevant, until post-purchase - when it then becomes key for loyalty.
With this new purchase funnel, with growing brand polygamy, and with agents buying objectively, the threat of commoditisation for your brand may be very real.
There’s no ready-made playbook for you to follow.
For the consumer, it’s good news. It leads to increased price competition and reduced differentiation, as AI agents go for the most cost-effective options based on the customer's needs.
For you, it will de-emphasise brand loyalty, as agents prioritise the best fit regardless of your top-of-funnel activity and brand name.
They will force you to focus on delivering exceptional post-purchase experiences where your customer is more receptive to a connection - your perfect opportunity to shape perceptions, tell deeper stories and create experiences.
It goes against conventional wisdom, but you’ll need to create, motivate, and excite the hardest after the sale is made.
Impact
It will decrease the overall influence of some brands. You may need to rethink your marketing and sales strategies that rely on mass data collection and broad segmentation.
It will impact the influence of your promise. As objectivity becomes an agent criterion, it’ll commoditise some products and services
It will change your customer’s journey. (There already isn’t a linear journey, and it will only get more messy)
It will either limit data capture (the customer refuses to share with you) or greatly increase understanding (the customer chooses to share with you to gain hyper-personalisation)
This will empower and supercharge the importance of CX.
The Disruption of Consumer - Brand Dynamics
We all love convenience. It’s a human bias. So would we choose to instruct our AI to search options, recommendations, and offers; then buy, perhaps even negotiate the prices, and arrange delivery at a time we’re home?
Of course, we would.
We’ll leave the ‘buying’ to our AI while spending our time earnt ‘shopping’.
Consumers reject choice complexity and demand its reduction, and the ability to outsource it - and in this new paradigm we’ll increasingly let our AI satisfy our buying needs while we engage in shopping experiences that are aspirational or emotional.
That means a large part of free choice (choice often created by branding and advertising) is moved into the domain of the algorithms of the AI assistant.
Band Of Indifference
At Hungryroot, a decade-old online grocery startup, the majority of the groceries it sells and ships to customers are picked by algorithm.
There’s this concept of consumers having a ‘band of indifference’. It runs across many, many segments and products. In the Hungryroot example, if you prefer broccoli over Brussels sprouts, the algorithm gives you broccoli. But if you’re within the ‘band of indifference’, which is where a lot of consumers are for large numbers of products or categories, it chooses whatever is cheapest/available/etc.
Last year, the company grew 40% - growth that far outpaced the broader online grocery market, which declined.
Convenience + Indifference = Your next customer could be an AI rather than a human, buying in this area of indifference. Everything from your advertising and persuasion to services and experiences and after-sales support will have to be refocused accordingly.
I’m sure brand loyalty will still dominate post-purchase in areas where consumers have a strong predisposition - devotees of say, Apple, luxury purchasers, or consumers seeking purpose-led brands - anything mostly shopping with consumer’s emotions.
It’s a shift, from us trying to grab the attention of many to grabbing the attention of one: the AI assistant.
This shift would benefit consumers, levelling the playing field across various buying categories, and favouring those of us delivering quality or emotional shopping experiences. And even if you’re an established brand it may need you to ramp up innovation and deliver more, newer and better experiences that outweigh AI’s objectivity.
What Will It All Look Like?
I’ll stick my neck out.
1. Direct Interaction: Agents will become intermediaries between consumers and your brand, potentially bypassing your advertising and communication channels - challenging you to adapt strategies to engage with consumers in new ways.
2. Brand Loyalty: With AI agents capable of comparing instantaneously, loyalty to your brand may become more polygamous as consumers delegate convenience and indifference to the agent.
3. Data Ownership: As agents start being a gatekeeper for its owners data, preferences, and behaviours, it will impact how you can gather and use customer information.
Here’s a flip side - one broadly positive for brands:
1. By leveraging permissioned customer data, you’ll be able deliver highly targeted and personalised advertising messages, greatly increasing your relevance and effectiveness.
2. Direct-to-Consumer Opportunities: Agents will be facilitating direct-to-consumer sales channels - so you can establish direct relationships with your customer, collect first-party data, and control the customer experience from end to end.
3. Focus on Customer Experience: As commoditisation intensifies, you can differentiate tthrough exceptional customer experiences rather than marketing/comms.
The Bottom Line
Brands that ignore these trends risk commoditisation and irrelevance.
Embrace: Resisting the rise of agents is futile. They’re coming anyway. Instead, focus on making your products and services AI-friendly with clear, structured data that agents can easily understand and integrate into their recommendations.
Leverage: Agents will be very good at comparing objective features and prices. Avoid commoditisation by emphasizing your differentiation and unique value proposition. Exceptional customer service? Warranties? Innovative or better features? Recommendations? A brand identity that emotionally resonates?
Innovate: Innovative brands thrive, and in an AI-driven world you need to continually develop new features, services, and ways of delivering value that go beyond what an AI agent can commoditise.
Conclusion
The evolution of AI personal assistants into agents will redefine the relationship between you and your consumers, empowering them, and challenging traditional marketing paradigms.
It's not automatic that if you’re a commodity and utility-type products you will lose all power due to AI agents. It will change your advertising and branding landscape, but also offer new opportunities for you to engage in more meaningful ways post-purchase.
Achieving AI agent preference will be more meaningful and valuable for you when you achieve thetype of engagement that will become a new benchmark of success for brands - quality, trust, experiences, personalisation.
“Whoever wins the personal agent, that’s [going to be] the big thing. Because you will never go to a search site again, you’ll never go to Amazon again” - Bill Gates
PS: Want to understand more? If you’re interested in attending and/or speaking at a seminar I’m planning on this topic, please message me:
Learn Best Practices…and Then Ignore Them
“We follow our industry’s best practices.”
I’m sure you’ve heard this statement many times, even done it.
And learning ‘best practices’ in any industry is very useful at the beginning.
You get to understand how things work. But you have to choose which ones.
Because you can be reasonably sure that if you do so blindly, its impact will be average.
Because when everybody follows them, we become copies of each other.
Similar products.
Similar thinking.
Similar brands.
And similar tactics.
The result? A lifetime of competition whilst you want to escape from the competition.
At some stage, you have to ignore best practices to be different.
The better way is to replicate the best practices and ideas from unrelated sectors.
Think Apple Store and design.
Or James Dyson’s early vacuum cleaner designs based on cyclone systems.
Other Examples of Cross-Industry Innovation:
Product: Nike copied the cables from suspension bridges to create Flywire technology that provides support with little weight in athletic shoes.
Marketing: Philip Krim and his friends realized almost all mattresses were sold in physical stores. So they founded Casper Sleep to sell mattresses using an e-commerce model.
Business Model: Dollar Shave Club got inspiration from software companies and created a subscription service for razors. Customers loved the convenience and cost savings. The company was sold to Unilever for $1 billion in 2016.
Experience: Peloton combined live group classes with home fitness equipment to create a unique workout experience.
Now, don’t get me wrong.
There are some operational areas where ‘best practice’ is necessary, helpful, vital even.
Like learning the ropes and gaining learned experience.
Like Security.
Like Safety.
But beyond that, you have to challenge the status quo to escape the status quo.
If not, doing what all others are already doing is the perfect way to become a commodity.
Imagine if Steve Jobs kept following the best practices of his time.
Every expert told him ‘Retail just isn’t done that way’. And where would the Apple Stores be if he listened?
As Picasso said:
“Learn the rules like a pro, so you can break them like an artist.”
The best CX videos and reads
Watch Microsoft AI CEO Mustafa Suleyman, one of the primary architects of the AI models many of us use today. He offers a compelling new vision for the future of AI, proposing an unignorable metaphor - a new digital species - to focus attention on our extraordinary times. TED | 2024
Read the interesting story of how Nike’s slogan “Just Do It” was briefed, informed and created.
Thanks for spending time with me on this edition of UNCX.
See you next time.
If you find this insightful, pass it along to a friend or colleague. And don’t forget to subscribe.