AI in Digital Commerce (Part 2): How AI impacts on-site journeys

From AI-assisted shoppers to emerging agentic traffic

In Part 1 of this series, we looked at how the 2025 Christmas trading period highlighted AI’s growing influence on digital commerce. Higher-intent traffic, shaped by AI-assisted research and recommendations, is already changing conversion dynamics.

But that’s only part of the story.

As AI becomes more embedded in digital commerce, its impact doesn’t stop at discovery or decision-making. It is increasingly beginning to influence how traffic behaves on site, both AI-assisted shoppers arriving with clearer intent, and emerging agentic traffic interacting with digital commerce platforms directly.

In this article, we look beyond the AI hype to focus on the practical, on-site implications for digital commerce teams: how journeys are changing, and why variability is becoming harder to ignore.

Two AI-driven forces are shaping digital commerce journeys

To understand what’s changing, it helps to distinguish between two related but distinct patterns.

1. AI-assisted shoppers

Many customers now use AI tools to research, compare and shortlist before visiting a site. By the time they arrive, much of the decision-making has already happened.

Available research and early platform data suggest these users often:

• engage more deeply once on site

• explore with clearer intent

• have less patience for friction or ambiguity

These journeys aren’t necessarily shorter or simpler but they are less forgiving. When the experience aligns with expectations, conversion can be fast. When it doesn’t, abandonment often happens more quickly.

In this sense, AI raises the bar for journey clarity and consistency rather than changing what a “human” journey looks like.

2. Agentic traffic interacting with sites directly

Alongside AI-assisted humans, some ecommerce sites are beginning to see more agentic interaction, where AI systems access sites directly to gather information, validate options or perform defined tasks on a user’s behalf.

Where agentic interaction exists, most of it currently happens through real browsers. From the site’s perspective, this traffic can look familiar, JavaScript execution, cookies, session state, but the behaviour is different.

Agentic traffic typically exhibits patterns such as:

• faster and more deterministic interaction

• less exploratory behaviour

• lower tolerance for poor performance

• high sensitivity to ambiguous UX

Broken buttons, inconsistent pricing, unexpected modals or unclear states that humans might work around can cause an agent to fail or abandon the task.

AI agents don’t introduce new problems, they surface existing fragility more consistently.

Why AI in digital commerce changes how traffic behaves

A common assumption is that AI traffic simply “starts deeper” in a site. In practice, the more important change is variability.

As AI-assisted shoppers and agentic traffic begin to coexist, ecommerce teams should expect:

• more fragmented journeys

• fewer predictable paths

• mixed human and non-human behaviour

• faster decision cycles

Journeys become harder to describe as a single, linear funnel. What matters more is whether outcomes remain achievable, even as behaviour becomes less uniform.

Why browsers still matter for AI-driven traffic

Because most AI systems interact through browsers, they inherit the same frontend complexity humans do and then stress it.

That includes:

• JavaScript execution timing

• consent banners and pop-ups

• conditional UI logic

• security and anti-bot tooling

Combined with faster, more deterministic behaviour, this can surface issues that only appear intermittently for human users, or not at all.

From a journey perspective, AI doesn’t bypass the frontend. It tests its reliability more aggressively.

What this shift means for modern ecommerce teams

The challenge for ecommerce, CX and digital teams isn’t predicting exactly how AI traffic will behave.

It’s recognising that:

• journeys are becoming more variable

• assumptions about “normal” behaviour are weakening

• edge cases matter more than averages

This has implications for:

• conversion reliability

• release confidence

• peak trading performance

• how success and failure are understood

AI doesn’t reduce the need for monitoring, if anything, it raises expectations of what monitoring must cope with as behaviour becomes faster, more automated and less forgiving.

The impact of AI on your online store

Agentic commerce isn’t a single leap. It’s an incremental shift, from AI-assisted discovery today to more agent-driven interaction over time.

At thinkTRIBE, we’re actively exploring, alongside clients and partners, how these changes could affect real-world digital journeys, and what they mean for how teams think about reliability, experience and risk as AI becomes part of everyday ecommerce traffic.

If this is something you’re starting to see, question or plan for, we’d welcome the chance to compare notes and explore what it could mean for your journeys.