Word of mouth has evolved into a complex ecosystem of reviews, creators, communities, search and AI, yet many marketers are still working from an outdated playbook. Paul Sinkinson, managing director (Asia and Australia) at Analytic Partners, reveals how word of mouth is now one of the most important competitive advantages a brand can have.
Word of mouth has an image problem. It sounds like something pulled from a marketing textbook two decades old – something traditional and fluffy, and a term that’s almost been retired because it’s too soft to put in front of a CFO.
That instinct is a mistake. The thing word of mouth describes hasn’t faded, it’s exploded.
Think about the last thing you bought that wasn’t an impulse purchase. Chances are you didn’t hear about it from one person and headed straight to the checkout. Perhaps you saw a creator mention it in a video, or it came up in Reddit. You may have even skimmed reviews on Amazon or asked ChatGPT what it thought.
None of that would have been called word of mouth 10 years ago. Today, all of it is – and the brands that still picture it as a friend’s recommendation over coffee are planning for a version of it that no longer exists.
Word of mouth has evolved from a recommendation into an influence system
In 2016, Analytic Partners research found word of mouth drove approximately 13 per cent of sales on average, with offline word of mouth generating around twice the impact of online, and one impression driving at least five times more sales than a paid ad impression.
If we ran that study today, I have no doubt the overall contribution would be far higher.
Much of that growth would be digital, but that does not make offline any less important. A recommendation from someone you know may still be the moment that puts a brand into consideration. What changed is what happens next.
That conversation might send someone to check reviews, search Google or ask AI, and the eventual sale could happen somewhere else entirely. Equally, a positive offline experience can become an online review influencing thousands of people.
The question is no longer offline versus online, but how influence moves between the two and accumulates before it appears as a sale.
IAB Australia research shows 68 per cent of online shoppers use social media to discover, research or compare products, rising to 88 per cent among 18 to 39 year olds. Add reviews, marketplaces and AI search into the mix, and recommendation is no longer something that happens after advertising has done its job – it’s woven throughout the journey.
AI is turning word of mouth into direct recommendations
Tools like Claude, ChatGPT, Gemini, and more have fundamentally changed word of mouth because they have become active participants in recommendation in a way Google, YouTube and Instagram never did.
It’s the difference between searching “laptop sale” versus “what laptop should I buy that fits my budget and lifestyle?”.
Rather than returning links to evaluate, an LLM synthesises reviews, commentary, community discussion and product documentation into what feels like a considered recommendation.
As a consequence, all those other sources become far more important. That’s why brands are desperately looking to build their GEO strategies, because brands need to understand how they’re being represented when AI is asked for an opinion, not simply whether they appear at all.
Which brings us to another important change. The internet may have enabled far more ‘word’, but it has reduced trust of mouth. Recommendation has become easier to generate while credibility is harder to establish. Consumers build confidence by looking for multiple signals pointing in the same direction so trust today is built through consistency, not one recommendation.
Permanence matters too. A review written years ago can influence a purchase today, and AI models draw on that entire body of publicly available information to answer queries. Every positive customer experience, useful review and credible recommendation has the potential to keep working long after the campaign that helped generate it has finished.
Word of mouth influences the sale but rarely gets the credit
Just because a recommendation is more visible doesn’t mean it’s easier to measure. Modelling from Analytic Partners found 30 per cent of paid search clicks are generated by other marketing activity, with a further 30 per cent to 60 per cent influenced by non-marketing factors like loyalty and seasonality.
Amazon is perhaps the clearest example of how much influence can disappear between two moments. A consumer might use its product pages to read reviews and decide whether they trust a product, then buy it from the brand’s own website, another retailer or a physical store. Amazon helped shape the decision but the sale appears somewhere else.
In one Amazon Ads apparel case study, 64 per cent of ad-attributed purchases happened outside Amazon and 33 per cent of those off-Amazon purchases showed a wider brand halo, generating sales beyond the categories being promoted.
That is where attribution becomes dangerous. Models built only around the last click can overstate the contribution of clickable activity by two to 10 times. Word of mouth is vulnerable to the same misreading – the recommendation creates the intent, while search or retail gets the credit.
Credible measurement needs several layers: exposure signals like reach and review volume, behavioural signals like search and conversion, and, where possible, controlled tests, fed into a model that also accounts for media, price and seasonality. Skip that discipline and marketers are simply attaching an intangible influence to whatever metric happens to be nearest.
Word of mouth belongs in the media plan
Marketers need to stop treating word of mouth as something that sits outside the media plan. Earned and paid have long been viewed as separate disciplines but that distinction is increasingly unhelpful because recommendation doesn’t sit besides media, it interacts with it.
Brands should be building better customer experiences, stronger products, creator partnerships, communities, advocacy programs and review generation because those activities increase the likelihood that customers will have something worth talking about.
Word of mouth is now bigger than any one platform, creator or conversation. It’s an interconnected system that influences what consumers discover, what they trust and ultimately what they buy.
The sooner marketers start treating it that way, the better.
