Poor website search drives Australian shoppers away

Australian online shoppers are abandoning retailers after poor on-site search experiences, with new survey data suggesting expectations are shifting fast as consumers increasingly use natural-language queries similar to those used with generative AI tools.

Research commissioned by Elastic among 1,020 Australian online shoppers found 72% had abandoned a brand because of weak website search. More than one in 10 said they would permanently stop using a brand after a single failed search.

The findings put search at the frontline of eCommerce performance. The survey suggests it influences conversion and repeat purchasing, with shoppers willing to switch brands quickly if they can't find products or information fast.

Loyalty pressure

Brand switching often follows an unsuccessful search. The survey found 62% of shoppers turn to an external search engine when a retailer's on-site search fails; 78% of those consumers are then directed to competing brands and products.

Shoppers also appear willing to pay more elsewhere if the journey feels easier. Elastic's research found 55% would pay more on a competitor's website for a better search experience.

Together, those behaviours create a direct revenue risk-especially for retailers that invest heavily in brand awareness and loyalty schemes but rely on older search approaches that struggle with how people now ask questions.

Elastic ANZ country manager Jeremy Pell described search as a competitive exposure point for online retailers.

"Your search bar has become your biggest competitor's best salesperson," Pell said.

Expectation shift

The study points to a widening gap between how consumers search and how many retail sites interpret requests. It found 62% of shoppers now expect-or wish-brand search bars were as intelligent as ChatGPT.

Younger customers are driving that change. Elastic found 52% of shoppers aged 25-34 now use full-sentence, natural-language queries rather than keywords. This cohort is also among the most likely to pay more to shop elsewhere or to leave a brand because of search performance.

The research also suggests a reputational impact. It found 60% of customers instantly judge brands as "technologically behind" when search returns irrelevant results.

Rising frustration

Poor search is no longer a minor irritation. The survey found 56% of respondents had abandoned a brand temporarily or permanently because they couldn't find what they wanted within seconds.

The data also points to heightened anger: 69% of Australian shoppers said they had shouted at a website search bar when results failed them. Shoppers aged 35-44 reported the highest frustration levels, while 17% of shoppers aged 65 and over said they had shouted when filters failed.

Poor search quality now ranks as the third biggest reason customers never return to a website, behind data breaches and bad customer service.

Features in demand

Shoppers also want new ways to search, not just more accurate results. The survey found 76% want photo or voice search options, and 28% described them as a potential "game-changer."

The findings come as retailers reassess digital investment priorities amid rapid consumer adoption of AI tools in everyday tasks. In online shopping, that appears to be translating into less patience for rigid keyword matching, slow discovery, and irrelevant results.

Pell said retailers should treat search as core to commercial performance, not a background website function.

"The brands that will win in 2026 and beyond are those that treat search as a strategic revenue driver, not a technical afterthought," Pell said. "AI-powered search that understands natural language, learns from behaviour, and delivers instant relevance is table stakes. Customers are fleeing now. The question is whether you're ready to stem the tide."

Retailers also face growing pressure to interpret customer intent across structured information such as product catalogues and reviews, and unstructured sources such as PDFs, emails, images and text messages-an area where traditional systems can fall short, Elastic said.

Elastic said it offers a unified context layer that brings together commerce data-including product details, customer reviews and high-resolution images-with AI outputs. It said the approach can produce personalised results and answers in response to shopper queries.

The survey was conducted in February 2026 via Pureprofile, with quotas set to reflect Australian Bureau of Statistics population proportions based on the 2021 census.