Why Your Search Results Are Nothing Like They Were and How to Improve Them

This video explains how modern search results have changed and offers practical ways to get more relevant information when a standard Google search is cluttered with ads, AI answers, SEO-heavy pages, and broad matches. The presenter focuses on using more targeted search behaviors rather than relying on the default results page. One concrete tactic discussed is Google Lens, which can be especially useful when you need to identify an object, plant, product, or item in a photo. Instead of typing vague keywords, you can snap a picture and use the visual search to narrow down what you are looking at, then use the resulting links to find purchase options or identification details.

The video also warns that AI-generated search summaries can be useful for quick orientation but should not always be trusted for factual precision. The presenter recommends clicking through to the source pages included in the AI result and checking what the original website actually says, especially when the question involves accuracy, product capability, or technical details. Another practical strategy is to add platform-specific terms to your query. For example, adding “Reddit” can surface results from Reddit itself or from pages discussing firsthand experiences, which is often useful for product research or troubleshooting. Similarly, going directly to YouTube can be a better route for how-to content and demonstrations, since video search results may be closer to what you actually need.

A final point is the risk of SEO manipulation and search-result gaming, where optimized pages may rank highly despite being low quality. The advice throughout is to be intentional about where and how you search, rather than accepting the top-ranked results at face value. Practitioners looking to improve everyday web searching can apply these techniques immediately: use visual search for identification, verify AI summaries against original sources, search within specific platforms when appropriate, and remain cautious about overly optimized results that may not be the most useful or accurate.