How to Use AI to Take Over Your Workflow (No Coding Required)

Author: Damian Player; Translation: Peggy, BlockBeats

Editor’s note: While most people still see AI as a “more efficient search tool,” Perplexity is starting to put it into work.

This article centers on a repeatedly overlooked difference—why, when using the same AI, some people only get an answer, while others directly receive deliverables. The key isn’t model capability; it’s how you use it: whether you treat it as a conversation window, or as an execution system you can direct and schedule.

A new class of tools, represented by Perplexity Computer, swaps “tasks” for “questions” as the core interaction method. From contract review and competitive analysis to data cleansing and report generation, users no longer describe the problem—they directly define the final deliverable. Combined with connections to enterprise tools, and by solidifying personal background and style examples, this capability evolves from one-time output into reusable, automatically runnable workflows.

More importantly, the boundary of automation is being redefined. It’s no longer just about helping you complete one step—it can run continuously, execute across tools, and even proactively propose follow-up tasks. This means the relationship between humans and tools is shifting from “using” to “managing and delegating.”

In this shift, the real dividing line is no longer whether you use AI, but whether you’ve started using it to “deliver results.”

Below is the original text:

People who get this figured out will gain an asymmetric advantage. Soon, everyone will learn how to do it. But before everything becomes obvious, here’s the way you can start early.

Over the past year, developers have been running autonomous AI agents in the background (like Claude Code, OpenClaw, etc.). They can do research on their own, build products, and deliver complete results directly—without people needing to constantly watch or keep prompting back and forth. But you’ve probably never been able to use this—unless you know how to use a terminal and write code.

And Perplexity Computer changes that. For the first time, non-developers can use the same capability. All you need is a browser—and a task you can hand off to it to complete.

Most people open Perplexity, type a question, get an answer, and then close the page. They miss what matters. Perplexity Computer isn’t for answering questions—it’s for executing tasks.

Stop asking questions. Start handing the real work to it.

Why most people fail

Chief financial officers, lawyers, consultants… They open the tool, type a question, get a decent enough answer, and then think: “Oh, a more advanced Google.” Next, they spend another 90 minutes cleaning the spreadsheet they cleaned last Monday.

The problem isn’t the tool. It’s the way they use it. They treat it like a chat bot.

Question style: “What risks does this contract have?”

Task style: “Review this contract. Line by line, verify whether all statements are supported by publicly available sources; flag vague wording, missing clauses, and sections that could create legal liability; list the 5 most critical risk points with specific clause references; output a Word document with tracked changes.”

Same contract. One approach gives you only a checklist for you to read; the other gives you a finished product you can send to the client.

Just set up this system in 10 minutes

First, connect the tool. Click connectors in the sidebar. Perplexity can connect to 400+ applications: Gmail, Google Drive, Slack, Salesforce, Notion, SharePoint… Connect whatever you actually use.

Then make it know who you are. Type it once: “I work in a certain role at a certain type of company. I regularly produce content X, Y, and Z. Please remember this background in every conversation.” It will retain those details long-term.

Next, tell it what “good” looks like. Find 2–3 of the成果 you’re most satisfied with, upload them, and input: “These are my best work examples. Please learn their format and tone, and use them as references when generating content going forward.”

That way, it isn’t guessing your style—it’s reverse-decomposing the proven success path you’ve already validated.

10 minutes—do this first.

A real example: that Monday that no longer takes 90 minutes

A financial analyst receives a data export every Monday—150 rows—with a messy format: duplicate data, three different date formats, and ratings written as words rather than numbers. Before she can analyze, she spends 90 minutes cleansing the data every week. Same question, repeated every week.

She gave it a single instruction: clean this file, deduplicate, standardize date formats, convert text ratings into numbers; analyze on the cleaned data; generate an interactive dashboard with filtering and provide a share link; output a PDF report comparing before and after cleaning; save all files to the “Monday Reports” folder in Drive.

After 4 minutes: a clean dataset, an interactive dashboard, a share link, and a PDF report—everything appears in her Drive.

Then she asks one more thing: “Is there an improvement I haven’t asked for that would make this even more useful?”

The system suggests two items: first, set this task to run automatically every Monday at 7:00 AM; second, add a task to generate Tuesday management briefings based on the segments that perform poorly.

She sets both, then closes the page.

After that, every Monday, it runs automatically—whether her computer is on or not.

This is the same capability developers have been using over the past year. Now, you can use it in your browser.

What people are using it for

@gregisenberg did a live test on the @startupideaspod podcast.

He only gave it one task: find the companies that place ads on competitor podcasts, identify the true sponsors responsible, and write a personalized email to each person.

The system found Ramp’s VP of Growth, pulled podcast content from two weeks earlier where he appeared, wrote a cold email quoting his specific lines from the show, and sent it directly. Greg didn’t say “send”—the system determined the task was complete and executed it on its own.

Then it proactively suggested: monitor competitor podcasts—once a new brand starts placing ads, immediately notify him and include the corresponding contact—“contact them when the budget just kicks off.”

In the end, this workflow completed parallel research on 96 potential customers and scheduled follow-up emails for Day 3 and Day 7.

On the Marketing Against the Grain show, the team used it to audit the entire HubSpot product pages: automatically crawl the whole site, score using custom standards, rank the issues, and generate a shareable website report. What would have taken a week for the team was completed while recording the episode.

These were all completed live—not a demo, not a pre-scripted routine.

How to use it for specific work

In the finance industry, a portfolio analyst issued just one task before Nvidia’s earnings release.

The result was: a real-time interactive dashboard including $130.5B in revenue, a 75% gross margin, a 114.2% growth rate, a full income statement, and a forecast trend of profit margins from fiscal year 2021 to 2028—all with filtering and share links supported.

No Excel. No manually hunting for data. Done in 5 minutes.

Perplexity can directly call data sources like the SEC filings, FactSet, S&P Global, PitchBook, and more—no API key needed, and no extra authorization required, since it’s built in.

Legal scenario:
“Review this contract. Line by line, verify whether all statements are supported by publicly available sources; flag vague wording, missing standard clauses, and sections that could create legal liability under [specific state] contract law; list the 5 most critical risk points with specific clause citations; output a Word document with tracked changes.”

A reviewer had uploaded a proposal claiming that the market year-over-year growth was 43%. Perplexity Computer found the real data showed only 4% and stopped the issue before signing.

Marketing scenario:
“Analyze [competitor 1], [competitor 2], and [competitor 3]’s best-performing content from the past 30 days; identify the content formats and themes with the highest engagement; identify content gaps; based on these gaps, generate a 30-day content calendar and save it as a Google Doc.”

Set it as a scheduled task. Every Monday, it automatically generates the latest competitor analysis—no manual research needed.

Operations scenario:
“This is our Q1 CSV data. Please clean the data; analyze revenue by region and product line; identify the three biggest problems; generate a one-page action summary; create a one-page PPT for reporting; save all files to the project folder.”

Five deliverables. One instruction. While you’re in the meeting, it’s already done.

Model Council (Model Council): 60 seconds for three kinds of judgments

When you face a decision with real consequences, just input the question once. Perplexity calls Claude, ChatGPT, and Gemini at the same time, and a “synthesizer” summarizes their consensus and disagreements.

·Areas where all three agree: high-confidence conclusions

·Areas where there is disagreement: needs further judgment

Someone asks whether product pricing should be set at $297 or $497. The three models give different answers, but the synthesizer finds the only thing they unanimously agree on is: don’t set it below $297. The decision is made there.

Many companies spend money hiring consulting firms to keep analysts in a meeting room until they arrive at conclusions.

Here, you only need one instruction.

The real core capability

To get real value from Perplexity Computer, 80% depends on one thing: can you clearly describe the “final output”?

Not technical setup. It’s whether you’re clear enough about what you need to deliver. Don’t describe the steps—describe the result.

After each task is completed, remember to ask again: “Is there something I haven’t asked that could make this outcome more useful?”

It almost always points out blind spots. Every time. Use it.

Start from here

Open Perplexity (Pro plan $20/month). Go to the Computer page, click connectors, and first connect Gmail and Google Drive.

Input your three-sentence background introduction (just once). Upload 2–3 of your best work examples so it can learn your style. Then choose a task you spent more than 2 hours on last week and whose outputs are similar every time: describe it in terms of the “final deliverable,” and send it. Observe the execution process. If it’s a repeating task, set it to run automatically before closing the page.

Developers have been using this system for a year already. The gap between their output and everyone else’s is real.

This is how to close the gap.