# Stop losing source links in research: the Perplexity workflow most consultants miss

Source: https://getstartedwithai.net/tutorials/perplexity-for-research-workflow
Site: Get Started With AI
Tagline: Build AI automations that actually run your business.
Audience: solo professionals, freelancers, consultants and AI implementers in training
Contact: hello@getstartedwithai.co.uk

This Markdown companion is generated from the prerendered page for AI crawlers.
# Stop losing source links in research: the Perplexity workflow most consultants miss

A 600-word client briefing with full citations, ready to paste into a discovery call agenda, finished inside 30 minutes. £16 a month for Pro covers the 20 daily Deep Research queries plus the choice of GPT-5.x or Claude 4.x as the underlying model. The lever that makes research stick is citation hygiene, the two-minute discipline of clicking every linked source. Perplexity is more accurate than 2024 but still misattributes occasionally, and unverified citations are the error that ends up in client deliverables.

## When to reach for Perplexity over ChatGPT or Google

- You need answers grounded in current sources with visible citations.
- You need to compare claims across multiple sources, not just one source.
- You are scoping a new sector or company you do not already understand.
- You need a research output ready to paste into a briefing doc.

## Pro pricing in April 2026

- Free: limited Pro Searches per day, Perplexity Sonar as the default model with no access to frontier models.
- Pro at £16 a month: unlimited Pro Search, choice of GPT-5.x, Claude 4.x, Gemini 3.x, plus 20 Deep Research queries a day.
- Max at £160 a month: same Pro features plus heavy Deep Research caps and faster runs.
- Education Pro: 12 months free for verified students or staff.

## Step 1: Set up the account (5 minutes)

1. Sign up at perplexity.ai and upgrade to Pro.
2. In Settings, set the default model to GPT-5.x or Claude 4.x. The default Sonar model is fast but writes flatter than the frontier models.
3. Set the default Pro Search to On. Without Pro Search, you only get a single-source answer; with it, you get a multi-source comparison.
4. Set citations to Always show full URL so you can verify sources at a glance.

## Step 2: Run the four-step research workflow (30 minutes)

This is the workflow that produces a briefing document for a discovery call or a sector scope. Run it once for any new client area.

1. Step A. Scoping question. Ask Perplexity a broad question to get the lay of the land. Example: "What are the top operational pain points for boutique law firms in the UK with 20 to 50 staff in 2026, with citations?"
2. Step B. Drill into the top three threads. Pick the three most useful from the answer and ask follow-up questions on each one. "Show me three case studies of mid-sized law firms that automated client intake in the last 12 months."
3. Step C. Cross-check. Take any specific claim that surprises you and ask "What sources contradict this?" If no sources contradict, the claim is more reliable. If sources do contradict, you know the area is unsettled.
4. Step D. Output. Ask Perplexity to "Produce a 600-word briefing document with the three pain points, the three case studies, the open questions and a list of follow-up readings, with full citations." Paste the result into your briefing doc.

## Step 3: Build a Space for the next sector scope (10 minutes)

Spaces are the Perplexity way of grouping research. Each Space has its own custom instructions, uploaded files and history, keeping multi-week research separate.

1. Click Spaces, Create new Space, name it after the client or sector.
2. Set the Space instructions: who you are, what the research is for, the level of detail you want, the citation style.
3. Upload up to 25 files: prior research, briefs, reports.
4. Run all queries inside the Space. The history stays scoped, the instructions apply on every query.

## Step 4: Two-minute citation hygiene (5 minutes)

A research output is only useful if the citations are real and recent. Two-minute discipline that prevents embarrassing mistakes.

- Click every citation that appears in the output. Confirm the URL loads and the claim is on the page.
- Check the date of every cited source. Flag any source older than 18 months unless the claim is historical.
- Run any contested claim through a second Pro Search to see if the same source list comes back.
- Never paste an unedited Perplexity output into a client deliverable. Always read, edit and verify before sending.

## What you ship

A configured Pro account, one Space loaded with 25 reference files, and a 600-word cited briefing on a real client area. Total time invested: 50 minutes. Time replaced on the next sector scope: roughly half a day of manual research.

## What to build next

Pair this with the Claude or ChatGPT setup tutorials to draft from the research output. The two-tool stack of Perplexity for research and Claude for drafting is what most consultants converge on.

## Related builds

More tutorials in Core Assistants. Each one runs in 60 to 90 minutes and ships a finished asset.

### Replace 3 hours of admin a week: the 60-minute ChatGPT build

30 to 60 minutes. Solo Pro.

[Open tutorial](/tutorials/chatgpt-setup-solo-professionals)

### Stop losing client context: organise every engagement in ChatGPT Projects

60 to 90 minutes. Solo Pro.

[Open tutorial](/tutorials/chatgpt-projects-client-work)

### Turn one repeatable task into an assistant: build your first Custom GPT in 30 minutes

60 to 90 minutes. Solo Pro.

[Open tutorial](/tutorials/build-first-custom-gpt)

### Cut research and drafting time in half: the Claude Project build

30 to 60 minutes. Solo Pro.

[Open tutorial](/tutorials/claude-setup-solo-pro)

## Questions people ask before they start

### Is Free enough for client work?

Free covers occasional personal research only, while client-facing work needs Pro for the 20 daily Deep Research queries and the model choice. The £16 a month is one of the easiest subscription decisions in the stack.

### How does Deep Research differ from Pro Search?

Pro Search returns an answer in 30 seconds from 20 to 30 sources. Deep Research takes 5 to 10 minutes, runs 50 to 100 sources and produces a structured report. Use Deep Research for scoping a new sector; use Pro Search for everything else.

### Can I share a Space with a client?

On Pro, Spaces are personal. Sharing requires Enterprise Pro at £40 per seat, which adds team workspaces and admin. For solo work, Pro is fine.

### Will Perplexity replace Google search?

For research with citations on a sector or company you do not understand yet, Perplexity wins. For everyday lookups like weather, addresses or simple facts, Google or any standard search is faster.

### How do I use Perplexity in a sales call?

Run a Pro Search on the client sector live in the call and show the side-by-side citations. Most clients have not seen the output next to a ChatGPT search; the comparison sells the engagement on its own.
