How to Set Up Your First AI Prospecting Workflow in 5 Minutes

Can You Really Set Up AI Prospecting in 5 Minutes?

Yes. AutoReach is designed so you can go from signup to your first AI-researched, qualified leads in under five minutes. The process has three steps: create a workflow, add prospects, and start the run. The AI handles research, qualification, and email drafting automatically.

This guide walks you through each step with exact instructions so you can follow along in real time.

Minute 1: Create Your Account and First Workflow

Sign Up

1. Go to [autoreach.work](https://www.autoreach.work) and click "Get Started"
2. Enter your email and password
3. Verify your email address

Create a Workflow

1. Click "New Workflow" from the dashboard
2. Name it something descriptive (e.g., "SaaS Companies - First Test")
3. Select your workflow stages:

- Research — AI analyzes company websites and gathers data - Qualify — AI scores leads and estimates close value - Contact (optional for first test) — AI drafts and sends outreach emails

1. Click "Create"

Your first workflow is ready.

Minute 2: Define Your Target

Set Your ICP Criteria

In the workflow settings, define who you are targeting:

Company characteristics:

• Industry (e.g., "SaaS", "Fintech", "E-commerce")
• Employee count range (e.g., 50-500)
• Geography (e.g., "United States", "Europe")

Contact characteristics:

• Target titles (e.g., "VP Sales", "Head of Growth", "CTO")
• Departments to target

You do not need to fill in every field. Even basic criteria help the AI focus its research and qualification.

"For your first workflow, keep it simple. Pick one industry, one size range, and one geography. You can always refine later based on results." — AutoReach Team

Minute 3: Add Prospects

Option A: Manual Entry

Click "Add Leads" and enter company details:

• Company name
• Website URL
• Contact name and email (if you have them)

Start with 10-20 companies for your first test. This gives you enough data to evaluate quality without waiting long for results.

Option B: Bulk Import

Upload a CSV file with columns for company name, website, contact name, and contact email. AutoReach maps the columns automatically.

Option C: Search-Based Discovery

Enter your ICP criteria and let AutoReach find matching companies. This is the fastest option but works best after you have validated your criteria with a manual test first.

Minute 4: Start the Workflow

Click "Start" to begin processing your leads.

What Happens Next

The AI agent processes each lead through your selected stages:

Research (30-60 seconds per lead):

• Visits and analyzes the company website
• Extracts company information (products, team size, tech stack)
• Identifies potential pain points and personalization hooks
• Stores structured research data

Qualify (10-20 seconds per lead):

• Evaluates the lead against your ICP criteria
• Generates a quality score (1-100)
• Creates a fit assessment explaining the score
• Estimates close value based on company characteristics

Contact (if enabled, 15-30 seconds per lead):

• Drafts a personalized email based on research findings
• Applies your tone and style preferences
• Queues the email for review or auto-send

For 10-20 leads, the entire process takes 3-5 minutes.

Minute 5: Review Your Results

Open the Review Panel

Click "Review" to see your processed leads. For each lead, you will see:

1. Quality score with color indicator (green = high, yellow = medium, red = low)
2. Company summary from the research stage
3. Fit assessment explaining why the lead scored the way it did
4. Close value estimate predicting potential deal size
5. Key signals (positive and negative indicators)

Take Action

For each lead:

• Accept — Lead is qualified and moves forward
• Reject — Lead is not a fit; removal from pipeline
• Flag — Save for later review

Your accept and reject decisions start training the AI agent memory, making future results more aligned with your preferences.

What to Do After Your First 5 Minutes

Review and Refine (Next 30 Minutes)

1. Look at quality distribution — Are most leads scoring high, low, or mixed?

- Mostly high (80+): Your ICP criteria are well-defined, or possibly too broad - Mostly low (below 40): Your ICP criteria may not match the prospects you added - Good distribution: The scoring model is working well

1. Check research accuracy — Open a few research summaries and verify against the actual company website. Is the AI capturing the right information?

1. Evaluate fit assessments — Do the reasons for high/low scores make sense to you?

Scale Up (Next Week)

Once you are satisfied with quality:

1. Add more prospects (50-100 in the next batch)
2. Enable the Contact stage if you have not already
3. Connect your SMTP account for email sending
4. Review and accept leads to build agent memory
5. Consider enabling auto-review for high-confidence leads

Go Continuous (Week 3+)

After you have reviewed 50+ leads and the AI is aligning with your preferences:

1. Switch the workflow to continuous mode
2. Set daily volume limits
3. Configure auto-review thresholds
4. Let the agent prospect around the clock
5. Shift your time to reviewing results and handling replies

Tips for First-Time Users

Tip 1: Start Small

Process 10-20 leads first. This lets you evaluate quality quickly without committing significant credits.

Tip 2: Be Deliberate in Reviews

Your first 25-50 reviews are the most important. Accept and reject thoughtfully — these decisions train the AI agent memory and shape all future results.

Tip 3: Check Research Quality

Open a few research reports and compare them to the actual company website. This helps you understand what the AI can and cannot detect.

Tip 4: Do Not Skip the Qualify Stage

Even if you think you know which leads are good, run the Qualify stage. The quality scores and fit assessments often surface signals you would not have noticed in manual review.

Tip 5: Connect SMTP Early

If you plan to use the Contact stage, connect your SMTP account early so you can test email sending with your first batch.

Common First-Run Questions

Why did a lead score low when the company seems like a good fit?

The AI may have detected signals you did not notice — a very small team, missing technology indicators, or other negative signals. Check the fit assessment for details. If the AI is wrong, accept the lead anyway — this feedback improves future scoring.

Why is research incomplete for some companies?

Some company websites have limited information (no team page, minimal product details). The AI can only extract what is publicly available. For these companies, you may need to supplement with manual research.

How many credits does a first run use?

Processing 10-20 leads through Research and Qualify typically costs 30-80 credits. Adding the Contact stage adds 20-40 more credits. Check your credit balance in Settings before starting.

Can I undo a workflow run?

You cannot undo research that has already been performed, but you can delete leads or reject them at any time. No outreach is sent until you approve it (unless you enable auto-send).

Your First Workflow Is Just the Beginning

Five minutes gets you from zero to qualified leads. The real power comes from iteration:

• Each review cycle makes the AI more accurate
• Agent memory learns your preferences over time
• Continuous mode maintains steady pipeline flow
• Multiple workflows target different segments simultaneously

Start simple, validate quality, build trust, then scale. Your first workflow is the foundation for an AI-powered prospecting system that improves every day.