The majority of companies that have invested in AI-assisted support have done so in one place: the website.

That's an excellent starting point. Web chat agents are relatively straightforward to deploy, easy to measure, and visible to leadership. 

However, two channels in particular stand out for how consistently they're overlooked, and how significant the opportunity is for organisations that move on them.

Channel 1: Voice

Voice remains the highest-volume customer service channel for most organisations by contact count. Yet it's the channel where AI assistance is most absent.

That gap has persisted partly because of perception. Voice AI has a history of frustrating experiences - rigid menus, misheard commands, dead ends that force callers to start over. 

Those systems left a lasting impression on both customers and the organisations that deployed them.

But the reality is that the tech has moved on considerably. Modern voice AI can handle natural, conversational speech in English (even with an Australian accent), manage interruptions and corrections mid-sentence, and escalate to a human agent with full context intact - so the customer doesn't have to repeat themselves.

The technical barriers have largely been resolved. 

What remains are organisational ones: telephony integration, contact centre alignment, governance and compliance, and operational change management.

These are solvable problems, and organisations that work through them are unlocking significant capacity.

Consider the scale of the opportunity. A contact centre handling 50,000 calls per month at an average handling time of seven minutes - even a 30% containment rate through voice AI represents a material reduction in cost and queue times.

Voice AI also operates after hours, capturing demand overnight and supporting outbound use cases: appointment reminders, payment follow-ups, booking confirmations, satisfaction surveys.

‍

Channel 2: In-Store

Physical retail is the channel AI strategies most consistently ignore, which is surprising given how much of the buying journey still happens in person.

In-store AI agents, deployed on kiosks, tablets, or staff-facing tools, are now capable of handling the same product knowledge queries, stock checks, and comparison tasks that a web agent handles online.

For categories like electronics, appliances and automotive, having AI available at the point of decision can directly improve conversion.

There's also a customer preference dimension worth acknowledging. 

Some customers actively prefer asking an AI for factual product information rather than waiting for staff or navigating what can feel like a sales interaction.

An in-store agent answers the question without pressure, at the customer's pace.

The internal productivity case is equally strong. Staff who can instantly access product specifications, stock levels, or compatibility information through an AI tool spend less time searching and more time with customers. 

That's a compounding gain across every store and every shift.

What Makes Omni-Channel AI Actually Work

Extending AI across voice and in-store isn't just a matter of deploying the same chatbot in more places. A few things have to be true for it to work well.

Shared context, not just shared branding. When a customer who chatted with a web agent later calls or visits a store, the AI-assisted interaction at that next touchpoint should have access to what was already discussed. Without a unified data layer, each channel starts from zero.

Channel-appropriate interfaces. The experience that works in a web chat widget is different from what works in a voice call or an in-store kiosk. Voice AI has to handle the rhythms of natural conversation: pauses, interruptions, corrections. In-store AI has to provide quicker support and task-based interactions. Designing for the channel matters as much as the underlying intelligence.

Consistent knowledge across every touchpoint. Product information, pricing, policies, and availability need to be the same regardless of where the customer asks. This requires discipline in how knowledge is managed and updated - but it's the foundation that makes omni-channel AI trustworthy rather than contradictory.

The Compounding Effect

There's a business case for each channel individually. But the more significant opportunity is what happens when they work together.

Customers who get consistent, accurate answers across every channel they use don't need to re-contact or escalate.

They don't call after visiting the website because they didn't get what they needed. They don't walk out of a store because no one could answer their question. They complete their journey.

McKinsey research consistently finds that customers who engage across multiple channels are at least 1.25 times more valuable than single-channel customers.

The mechanism is straightforward: friction reduction compounds across the journey.