Your customers are making decisions about your business before they ever visit your website. They’re asking ChatGPT for recommendations. They’re getting AI-generated summaries instead of clicking through to your carefully crafted landing page. They’re comparing you to competitors in conversations you can’t see, measure, or influence.
And your analytics dashboard? It shows none of this.
This is the quiet shift that’s catching Irish businesses off guard. The customer journey hasn’t just changed; entire stages of it have moved off your property and into AI-powered environments where traditional tracking simply doesn’t reach. Discovery, consideration, even shortlisting now happen in places your Google Analytics can’t see.
For Irish SMEs operating in competitive markets, this matters enormously. Customer acquisition costs are climbing. GDPR means you can’t just hoover up data and hope for the best. And Irish consumers, frankly, have high trust expectations. They’ve been burned by pushy marketing before, and they can smell inauthenticity a mile off.
So what do you actually do about it? That’s what this post is about. Not vague futurism. Practical steps you can take this quarter to understand and adapt to how AI is reshaping the way your customers find, evaluate, and choose businesses like yours.
How Is AI Rewriting Every Stage of the Customer Journey, From Awareness to Retention?
Let’s walk through the full journey and look at where AI is now embedded. Because it’s not just one stage. It’s all of them.
What’s Changed at the Awareness Stage?
This is where the biggest disruption is happening. People are discovering brands through AI search results, algorithmic social feeds, and recommendation engines rather than through the traditional ten blue links. Google’s own AI Overviews now synthesise answers at the top of search results, often without the user needing to click through to any website at all.
Social algorithms on TikTok, Instagram, and LinkedIn decide what content surfaces. YouTube’s recommendation engine drives discovery. Your potential customer might learn about a solution to their problem without ever typing a branded search query.
What’s Happening During Consideration?
This is where it gets properly uncomfortable for marketers. AI assistants are now doing the comparison shopping. A potential customer asks an AI tool to compare accounting software options in Ireland, and they get a synthesised answer drawing from reviews, pricing pages, and third-party content. The user never visits your site during this entire evaluation phase.
Review synthesis is another shift. Instead of reading twenty individual reviews, people get AI-generated summaries of what customers think. The nuance of your five-star reviews and <a href=”#”>customer success stories</a> might be flattened into a single sentence.
How Is AI Affecting Conversion and Beyond?
At the conversion stage, AI is driving predictive offers, intelligent chatbots, and guided selling experiences. The businesses doing this well are seeing genuinely better outcomes. Those doing it badly are annoying people with robotic chat pop-ups.
Post-sale, AI enables churn prediction, next-best-action recommendations, and personalised lifecycle messaging. It can identify which customers are likely to leave and trigger retention campaigns before they do. For advocacy, AI tools can identify user-generated content, surface community insights, and help brands understand what their happiest customers are actually saying.
The critical point: this is always-on. AI-driven personalisation runs continuously across web, apps, email, and marketplaces. It doesn’t clock off at 5pm. Your customer journey is being shaped by algorithms 24 hours a day, whether you’re actively managing that or not.
Why Can’t Traditional Analytics Show Most of Today’s Customer Journey?
Here’s the problem I see with almost every client we work with at BeFound. They’re making decisions based on data that captures maybe 40% of what’s actually happening. The other 60%? Invisible.
What Is the “Journey Invisibility” Problem?
When someone asks ChatGPT “best physiotherapist in Dublin” and then visits your website directly, that shows up as direct traffic. There’s no referral data. No campaign tracking. No indication that AI played any role in the discovery.
Dark social compounds this. Someone shares your link in a WhatsApp group. A colleague mentions your service on Slack. A friend sends a voice note recommending you. None of this appears in your analytics as anything other than mysterious “direct” traffic.
Where Do Standard Tools Fall Short?
Last-click attribution was already a poor model. Now it’s actively misleading. If someone’s journey was: saw your LinkedIn post, asked ChatGPT about your category, read a Reddit thread mentioning you, then Googled your brand name, your analytics will credit 100% of that conversion to brand search. The other three touchpoints? They don’t exist in your data.
Walled gardens make this worse. Meta, TikTok, and Google each have their own measurement ecosystems that don’t talk to each other properly. AI-driven referrals from tools like Perplexity or ChatGPT are often miscategorised or simply not tracked.
What Does This Look Like in Your Reporting?
You’ll notice some telltale signs. Direct traffic is inflated beyond what makes sense. Brand search is increasing but you can’t identify what’s driving top-of-funnel awareness. Page paths are shorter because users arrive already informed. Time on site might drop because people need less convincing.
If you’re seeing these patterns and interpreting them as “things are getting worse,” you might be wrong. The journey hasn’t disappeared. It’s just moved somewhere your tools can’t follow.
What Does AI-Driven Traffic Look Like, and Why Can Onsite Metrics Be Misleading?
When someone arrives at your site after an AI-assisted research process, their behaviour looks fundamentally different from traditional organic traffic. Understanding this is crucial to avoid misinterpreting your data.
How Do AI-Informed Visitors Behave Differently?
These visitors tend to be more engaged but less exploratory. They’ve already done their research. They know what they want. They might visit just one or two pages, but those visits have significantly higher intent.
Think about it. If someone has already asked an AI assistant to compare three Dublin accountancy firms and yours made the shortlist, they’re arriving at your site to confirm a decision, not to browse. Fewer pageviews, shorter sessions, but much higher conversion potential.
Why Are Traditional Engagement Metrics Now Misleading?
Reduced pageviews doesn’t mean reduced consideration. It means the consideration happened elsewhere. If you’re judging content performance purely by pages per session or average session duration, you’re measuring the wrong things.
A visitor who lands on your pricing page, spends 90 seconds there, and then fills in a contact form has had an incredibly efficient journey. But by traditional engagement metrics, that looks like a low-engagement session.
What Should You Measure Instead?
Focus on outcomes, not activity. Assisted conversions in GA4 give you a better picture than last-click. Post-click quality matters: are the leads from “low engagement” sessions actually converting to customers?
Lead-to-sale rates tell you more than click-through rates ever did. And honestly, one of the most underrated measurement tools is simply asking. A “how did you hear about us?” field on your contact form, or a brief question during the first sales call, can reveal journey intelligence that no analytics platform captures.
We’ve started recommending this to all our clients. The qualitative data from these simple questions has been more valuable than months of attribution modelling.
How Can Brands Build “Journey Intelligence” Beyond the Data to Understand Real Customers?
If the data can’t show you the full picture, you need to supplement it with genuine customer understanding. This isn’t just a nice-to-have any more. It’s essential.
How Do You Combine Quantitative and Qualitative Insight?
Your analytics tell you what happened. Customer conversations tell you why. You need both.
Voice of Customer research doesn’t have to be expensive or complicated. Five customer interviews can reveal more about your actual journey than a month of staring at dashboards. Ask questions like: “When did you first hear about us? What else did you consider? What nearly stopped you from getting in touch?”
The answers will surprise you. I guarantee it.
What Role Does AI Play in Understanding Your Customers?
Here’s where AI helps rather than hinders. AI-powered social listening tools can mine reviews, forum posts, and social mentions to identify patterns in what customers say about your category. Not just about you, but about the problems they’re trying to solve.
Journey mapping needs updating too. Your customer journey map probably doesn’t include “asked ChatGPT” or “watched AI-generated comparison video” as touchpoints. It should. Map the journey as it actually exists, including the AI-assisted stages you can’t directly measure.
Why Should You Look Beyond Your Own Website?
Market-level intelligence matters more than ever. What’s the overall demand for your category? Where are competitors showing up that you’re not? Are there marketplace or directory listings driving traffic you’re missing?
The output of all this work should be clearer segment needs, identified friction points, and specific content improvements you can make. Not a pretty report that sits in a drawer. Actionable intelligence.
How Should Marketers Redesign Trust, Transparency, and the Human Side of AI Experiences?
Just because you can automate something doesn’t mean you should. And if you do automate it, you need to do it in a way that builds rather than erodes trust.
How Do You Design Trust Into AI-Driven Experiences?
Be upfront about when AI is involved. If a chatbot is handling initial enquiries, say so. Don’t pretend it’s a human. Irish consumers are particularly sensitive to this; there’s a cultural expectation of honesty and directness that you undermine at your peril.
Always provide a human fallback. Nothing destroys trust faster than being trapped in an AI conversation loop when you need to speak to an actual person. Make the escape route obvious and easy.
What Are the Privacy and Compliance Considerations in Ireland?
GDPR isn’t optional, and it’s not just about cookie banners. If you’re using AI to personalise experiences, you need lawful bases for processing that data. The Data Protection Commission in Ireland has been increasingly active, and the fines for getting this wrong are significant.
There’s also a “creepy line” that’s worth respecting even when you’re technically compliant. Just because you legally can use someone’s browsing history to personalise their experience doesn’t mean you should make it obvious you’re doing so. Subtlety matters. Helpful personalisation feels like good service. Obvious personalisation feels like surveillance.
What About Bias and Brand Safety?
AI outputs can contain biases, inaccuracies, or content that doesn’t align with your brand values. If you’re using AI to generate customer-facing content or responses, you need guardrails. Human review, fact-checking processes, and clear guidelines for what the AI can and cannot say.
The principle I keep coming back to is “automation with empathy.” Speed plus care. Efficiency without losing the human touch that makes people want to do business with you in the first place.
What Must Marketers Do to Stay Discoverable in AI Search and a Changing Discovery Landscape?
This is where the practical work begins. If AI is now a significant discovery channel, you need to be visible in AI-generated results. Here’s how.
How Do You Strengthen Your Brand Fundamentals?
AI tools recommend brands they can clearly understand and categorise. If your positioning is muddled, if it’s unclear what you do and for whom, AI will struggle to recommend you. Clear category positioning, specific proof points, and a well-defined value proposition aren’t just good marketing. They’re now technical requirements for AI discoverability.
This is actually good news for businesses that have done the foundational work. If you’ve invested in building a genuine brand with clear positioning, you have an advantage over competitors who’ve relied purely on SEO tricks.
What Kind of Content Does AI Actually Use?
AI systems pull from content that is clearly structured, factually grounded, and directly answers specific questions. FAQs, comparison pages, definition pages, and original research are all formats that AI can easily parse and reference.
Think about it from the AI’s perspective. It needs to synthesise information from multiple sources into a coherent answer. Content that’s clearly organised with descriptive headings, that states facts plainly, and that provides original insights is exactly what these systems are looking for.
Generic, keyword-stuffed pages that say nothing original? AI has no reason to cite those. And neither does anyone else, frankly.
What Authority Signals Matter for AI Visibility?
Credible citations from authoritative sources help AI systems trust your content. Expert authorship with verifiable credentials matters. Case studies with specific results, like this [B2B SEO case study](link), provide the kind of concrete evidence AI tools love to reference.
Reviews on trusted platforms like Google Business Profile, Trustpilot, and industry-specific directories feed into the data that AI systems draw from. If you’re not actively managing your presence on these platforms, you’re invisible to a growing portion of the discovery process.
What Is Generative Engine Optimisation and Should You Care About It?
Generative Engine Optimisation, or GEO, is the emerging practice of optimising your online presence specifically for AI-driven discovery. It’s not replacing SEO. It’s extending it.
The basics include entity clarity (making sure search engines and AI systems understand exactly what your business is), consistent NAP information (name, address, phone number) across all platforms, structured data markup using Schema.org vocabulary, and creating AI-friendly content assets.
For Irish businesses, your Google Business Profile is particularly important. It’s one of the primary data sources that AI systems use when making local recommendations. Keep it complete, accurate, and regularly updated.
What Are the Positive and Negative Impacts of AI on Marketing, and How Do You Manage the Trade-Offs?
I’m not going to pretend AI is all upside. It’s not. But I’m also not going to pretend you can ignore it. You can’t. The honest answer is somewhere in between.
What Are the Genuine Benefits?
Faster insight generation. What used to take a week of analysis can happen in hours. Better segmentation, because AI can identify patterns in customer data that humans simply can’t process at scale. Improved personalisation, when done thoughtfully. Operational efficiency that frees up your team to focus on strategy rather than repetitive tasks.
Customer support is another area where AI genuinely helps. Handling routine enquiries quickly and accurately, triaging complex issues to the right person, providing 24/7 availability. For small businesses without the budget for round-the-clock staff, this is transformative.
What Are the Real Risks?
Measurement blind spots, as we’ve discussed. Over-automation that strips the personality from your brand. AI hallucinations that generate plausible-sounding but completely incorrect information. Trust erosion when customers feel they’re being manipulated rather than helped.
Brand voice dilution is a big one. If you’re using AI to generate all your content, you risk sounding like everyone else. The thing that makes your business distinctive, your voice, your perspective, your personality, gets averaged out into generic AI-speak. I’ve seen this happen to businesses that went all-in on AI content without maintaining editorial oversight.
How Do You Manage These Trade-Offs?
Build a simple mitigation checklist. Human review standards for all AI-generated customer-facing content. Fact-checking processes, especially for anything that includes claims or statistics. Regular monitoring of AI outputs to catch drift or quality drops. And a clear governance policy that defines what AI can and cannot be used for in your business.
None of this has to be complicated. A one-page acceptable use policy and a quick review step before anything goes live will catch 90% of potential issues.
What Should Irish Marketers Prioritise First When Adapting to AI-Driven Customer Journeys?
Right. Enough theory. Here’s what to actually do, in order.
Step 1: Can You Audit Your Discovery and Measurement Gaps?
Start by understanding what you can’t see. Look at your direct traffic trends. Has “direct” been growing disproportionately? Check your brand search volume. Is it increasing without a clear driver? These are signals that AI-assisted discovery is already happening for your business.
Search your brand name in ChatGPT, Perplexity, and Google’s AI Overviews. What comes up? Is it accurate? Is it flattering? This takes about 15 minutes and the results can be eye-opening.
Step 2: Have You Mapped Your Current Journey Including AI Touchpoints?
Take your existing customer journey map and add the AI-influenced stages. Where might someone encounter AI-generated information about your category? Which AI tools are active in your space? Where are the gaps between what your analytics show and what’s likely happening?
Talk to recent customers. Ask them how they found you. The gap between their answers and your attribution data is your measurement blind spot.
Step 3: What High-Impact Use Cases Should You Pilot?
Pick one or two areas where AI can make a measurable difference. Maybe it’s an AI chatbot for initial enquiries. Maybe it’s using AI to analyse customer reviews and identify common pain points. Maybe it’s optimising your content for AI discoverability.
Whatever you choose, define clear success metrics before you start. “We want to increase qualified enquiries by 15% this quarter” is useful. “We want to use AI” is not.
Step 4: What Governance and Quality Controls Do You Need?
Before scaling anything, establish basic guardrails. Who reviews AI outputs? What’s the process for catching errors? What data can be used and what’s off limits? These don’t need to be bureaucratic. A few clear rules that everyone understands will do.
Step 5: How Do You Iterate and Scale What Works?
Measure the results of your pilots honestly. What worked? What didn’t? Scale the wins and kill the failures quickly. AI capabilities are evolving fast, so build a habit of quarterly reviews to assess what’s changed and what new opportunities or risks have emerged.
The businesses that will thrive aren’t the ones that adopt every new AI tool on day one. They’re the ones that build a systematic approach to testing, learning, and adapting. Boring? Maybe. Effective? Absolutely.
Frequently Asked Questions
How Do I Measure AI-Influenced Journeys if Attribution Is Getting Worse?
Combine what your analytics can tell you with what they can’t. Use GA4’s data-driven attribution models rather than last-click. Add “how did you hear about us?” fields to your forms. Track lead-to-sale conversion rates rather than just click metrics. Run regular brand searches in AI tools to understand your visibility. And accept that some influence will always be unmeasurable; focus on outcomes rather than trying to attribute every touchpoint perfectly.
Will AI Marketing Replace Traditional Marketing Roles in Ireland?
Some roles will change significantly, but outright replacement is overstated. Repetitive tasks like basic reporting, initial content drafts, and data processing will increasingly be handled by AI. But strategy, creative direction, client relationships, and brand building still require human judgement. The marketers who thrive will be those who learn to work effectively with AI tools, using them to amplify their capabilities rather than viewing them as threats.
Is Personalisation Still Worth It Under GDPR?
Yes, but it has to be done properly. GDPR doesn’t prohibit personalisation; it requires transparency, lawful bases for data processing, and respect for user rights. First-party data collected with clear consent is your strongest asset. Focus on personalisation that genuinely helps the user rather than personalisation for its own sake. If a customer would thank you for making their experience more relevant, you’re on the right track. If they’d feel uncomfortable knowing what you know about them, pull back.
How Do I Know Whether AI Is Helping or Hurting My Customer Experience?
Ask your customers. Seriously. Monitor satisfaction scores before and after implementing AI touchpoints. Track complaint rates and the nature of complaints. Watch for signs of frustration like repeated attempts to reach a human, abandoned chat sessions, or negative reviews mentioning automated responses. If your AI implementations are making things faster and easier for customers, great. If they’re creating new friction, reassess.
What’s the First Practical Step We Should Take This Quarter?
Search for your brand and your primary service category in ChatGPT, Perplexity, and Google AI Overviews. Screenshot the results. Then compare what AI says about your business with what you’d want it to say. The gap between those two things is your immediate priority list. From there, review your Google Business Profile for completeness, ensure your website content clearly answers the questions your customers ask, and add structured data to your key pages. These foundational steps cost nothing and position you for AI-driven discovery.
Where to Start
The shift in customer journeys isn’t coming. It’s already here. Your customers are using AI tools to research, compare, and decide, and that process is largely invisible to your current measurement setup.
The good news? You don’t need to overhaul everything overnight. Start with understanding what’s changed, audit your visibility in AI-driven channels, and make incremental improvements to your content, your data practices, and your measurement approach.
If you want help auditing your customer journey strategy and identifying where AI is already affecting your visibility, we’d be happy to take a look. Get in touch with BeFound and we’ll walk through it together.
