Perplexity has quietly rolled out 10 significant features since December 2025 that fundamentally change how the AI search engine works, transforming it from a reactive search tool into a proactive research and intelligence system. Unlike flashy product launches, these updates arrived without press events or announcements, but they represent meaningful shifts in what the platform can do for leaders, researchers, and professionals who rely on it for decision-making .

What Changed in Perplexity's Memory and Search Capabilities?

The most significant upgrade involves Perplexity's memory engine, which now recalls the right details in 95% of cases, up from 77% previously, while creating half as many memories overall . This means the system learns what matters to you without cluttering itself with irrelevant information. If you've been researching a specific market, tracking a competitor, or working on a recurring project, Perplexity now carries that context into your next conversation without requiring you to re-explain your background or goals.

The search function itself has also improved dramatically. Previously, Perplexity required close keyword matching to find past conversations. The updated engine now handles fuzzy recall, meaning you can search by topic, approximate wording, or what you were trying to figure out at the time, rather than remembering exact phrases. Results show text previews of matching content, and clicking a thread takes you directly to the relevant message instead of the top of the conversation .

How Can You Set Up Perplexity to Work Smarter for Your Workflow?

• Personalization Settings: Open Settings and select Personalize to set your occupation, custom instructions, preferred response length, and whether you want essays or scannable outputs. This takes approximately 60 seconds and applies to every answer from that point forward without requiring you to restate your context each session.
• Memory Reference Toggle: Enable "Reference search history" in the Memory section so Perplexity uses your previous searches when answering new questions, allowing the system to accumulate your working context the way a good analyst would.
• Learn Step by Step Mode: Access this feature through the "+" menu in the input bar by selecting More and then "Learn step by step." Instead of receiving a direct answer, Perplexity walks you through a topic with questions, hints, and scaffolded explanations tailored to your knowledge level, which is particularly useful for understanding competitive dynamics, financial concepts, or technical architecture.

For leaders who want concise, scannable outputs rather than lengthy essays, setting both personalization dropdowns to match your preference prevents the frustration of receiving detailed responses when you asked a quick question .

How Do Scheduled Searches Turn Perplexity Into a Proactive Intelligence System?

Perhaps the most transformative feature is Scheduled Searches, which are recurring queries that run automatically and notify you with results. This shifts Perplexity from reactive to proactive, allowing you to set a search to run daily, weekly, or monthly and receive a notification when results are ready . You can include a prompt that shapes the format and focus of the output, essentially creating a custom intelligence briefing that runs without your involvement.

To set up a scheduled search, navigate to Notifications in the left sidebar, select Scheduled Searches, create a new scheduled search by entering your prompt, set the frequency, and toggle notifications on. For example, a weekly AI briefing could be configured to fetch fresh web and Reddit content from the last 72 hours, prioritizing new AI models, papers, benchmarks, agentic systems, alignment debates, policy moves, and open-source releases from reputable sources like AI labs, research groups, and well-known researchers .

What New Connectors and Data Sources Are Now Available?

Perplexity has introduced connectors that pull live context directly from your existing tools, including Gmail, Slack, and Google Drive . This integration means your research can incorporate real-time information from your email, team communications, and stored documents without manually copying and pasting information between systems. Additionally, Perplexity Finance now includes SEC-linked financial data, making company research actually auditable and traceable to official sources rather than relying on aggregated or summarized information .

The combination of these features represents a fundamental shift in how Perplexity functions. Rather than treating it as a fancier version of Google Search, users can now build it into their actual research workflows, letting it accumulate context, run intelligence briefs automatically, and pull live data from the tools they already use daily. For leaders and professionals who spend significant time researching markets, competitors, or technical topics, these updates mean Perplexity can function as a dedicated research assistant that learns your priorities and delivers relevant information without constant prompting .

The practical implication is clear: Perplexity has moved beyond being a search alternative and into territory where it can genuinely augment how teams gather, organize, and act on information. The features are not all equally impactful, but collectively they represent the kind of incremental, workflow-focused improvements that often matter more than flashy announcements.