How does scrapensend work?

You describe your target audience (e.g., "25 software houses in Europe"), write instructions for the AI email, pick the tone and style, and hit Execute. scrapensend finds companies, scrapes their contact emails from real websites, generates a unique personalized email for each one, and sends them from your inbox. Setup takes about 5 minutes.

What do I need to fill in to create a campaign?

Two main things: (1) who you want to reach — a description of your target companies, and (2) what the email should say — instructions for the AI. Then you pick the tone, length, opening hook, call-to-action, and language from dropdown menus. You can also add an HTML signature and set the send interval. It takes about 5 minutes total.

Is scrapensend safe to use with my Gmail?

Yes. scrapensend uses official Google OAuth2 — we never see or store your password. Only an encrypted refresh token. Emails are sent from your actual account with natural timing intervals (3-6 minutes between each) to protect your sender reputation. You can revoke access anytime from your Google Account settings.

How personalized are the AI-generated emails?

Each email is uniquely generated. The AI reads the target company's actual website, understands their business, services, and context, and writes a message based on your instructions specifically for them. No templates, no mail merge — two companies in the same industry get completely different emails.

Where do the company emails come from?

scrapensend scrapes them from the company's own website — contact pages, footer, about pages, mailto links, even Cloudflare-obfuscated emails. These are public business emails (info@, contact@, hello@), not personal addresses. If no email is found on a website, that company is skipped.

Is cold emailing legal?

Yes. Sending B2B cold emails to publicly available business addresses is legal under GDPR (legitimate interest, Article 6(1)(f)) and CAN-SPAM Act. scrapensend only targets companies, never private individuals. Our prompt validation system blocks requests for personal data.

How much does scrapensend cost?

Three plans: Starter at $19/month (10 campaigns, 100 leads each), Pro at $39/month (40 campaigns, 200 leads each, full style controls), and Business at $79/month (unlimited campaigns, 250 leads, no branding in emails). No setup fees. Cancel anytime.

Can I control the tone and style of emails?

Yes. You can customize: objective (sales, partnership, link building, research, recruiting), opening hook (website visit, industry match, direct question, compliment), tone (casual, direct, formal, bold), length (short or standard), call-to-action (soft check, find person, offer resource, book call), and language (auto-detect, English, Polish, German, Spanish, French).

Can I preview emails before they are sent?

Yes. scrapensend has a Preview button that generates a sample email using your instructions and style settings, addressed to a fictional company. You can review the tone, content, and structure before launching the campaign.

What email providers does scrapensend support?

Currently Gmail (via OAuth2), SendGrid (via API key), and Resend (via API key). Microsoft Outlook, custom SMTP, Mailgun, and Amazon SES support is coming soon.

Can I stop a campaign after it starts?

Yes. One click on the Stop button and no more emails will be sent. You can also resume a stopped campaign later. All progress is saved.

Do I need any technical knowledge to use scrapensend?

None. If you can fill in a form and click a button, you can use scrapensend. There is no code to write, no API to configure, no CSV files to import.

AI Cold Emails vs Templates: Which Actually Gets Replies?

The cold email world is split into two camps right now. On one side, you have the template purists: proven frameworks, merge fields, and systematic A/B testing. On the other, you have the AI believers: let a language model write a unique email for every single prospect.

Both camps have loud advocates. Both camps cherry-pick their best results. So let us actually break this down with data, nuance, and honesty about when each approach works — and when it does not.

The Template Approach: How It Works

Template-based cold email is the method that built the outbound sales industry. You write a master email, insert variables like {first_name}, {company_name}, and {pain_point}, then send it to your entire list. Better teams create multiple templates for different ICPs and run A/B tests on subject lines and opening lines.

Pros of Templates

Predictable and testable — When you have a fixed template, you can isolate variables. If version A gets 2% replies and version B gets 4%, you know exactly what changed.
Fast to launch — A good copywriter can produce 5 template variants in an afternoon. You can be sending by tomorrow.
Easy to scale — Once you find a winning template, you can send it to 10,000 people without additional effort.
Proven frameworks — Formats like AIDA, PAS, and BAB have decades of direct response marketing behind them.

Cons of Templates

Fatigue is real — When every outbound team uses the same frameworks, prospects start recognizing the patterns. "I noticed that {company} is..." has become a punchline in sales communities.
Personalization is shallow — Merge fields only go so far. Swapping a first name and company does not make an email feel personal.
One size fits none — A template written for 1,000 people cannot speak to anyone's specific situation.
Diminishing returns — Template reply rates have been declining industry-wide as inboxes get more crowded and prospects get more skeptical.

The AI Approach: How It Works

AI-generated cold emails use large language models to write a unique email for each prospect. The system takes input data — the prospect's name, company, role, recent activity, and any other context — and generates a complete email tailored to that individual.

Modern AI email generators go beyond simple name insertion. They can reference a prospect's recent LinkedIn post, comment on their company's product launch, or connect their hiring patterns to a pain point your product solves.

Pros of AI-Generated Emails

Genuine personalization at scale — Each email is unique. Even if two prospects are at similar companies, they get different messages that reference their specific context.
Better pattern interruption — AI-written emails do not follow the formulaic structures that prospects have learned to ignore.
Higher reply rates — In our data across thousands of campaigns, AI-personalized emails average 5% to 8% reply rates compared to 2% to 4% for templates in the same market segments.
Scales with data — The more context you feed the AI, the better the output. Rich lead data directly translates to better emails.

Cons of AI-Generated Emails

Quality variance — AI does not always nail it. Some generated emails miss the mark or sound slightly off. You need quality controls.
Harder to A/B test — When every email is different, isolating what works requires larger sample sizes and more sophisticated analysis.
Dependent on input data — AI can only personalize with the data it has. Bad lead data produces generic-sounding AI emails that are worse than a good template.
Cost and speed — Generating unique emails takes more computational resources and time than sending a fixed template.

Side-by-Side Comparison: The Data

We analyzed performance data from campaigns run through ScrapenSend over a six-month period, comparing template-based campaigns against AI-generated campaigns targeting similar audience segments.

Here is what we found:

Open rates: Templates averaged 52%, AI emails averaged 58%. The difference comes primarily from AI-generated subject lines, which tend to be more varied and less "salesy."
Reply rates: Templates averaged 2.8%, AI emails averaged 5.4%. This is the most significant gap. AI emails nearly doubled the reply rate.
Positive reply rates: Templates had 1.1% positive replies, AI emails had 2.9%. Not only did AI get more replies — a higher percentage of those replies were interested rather than annoyed.
Unsubscribe and spam reports: Templates had 0.8% combined, AI emails had 0.3%. Personalized emails feel less like spam, even to people who are not interested.

The numbers favor AI, but context matters. Let us dig into when each approach actually makes sense.

When Templates Still Win

Templates are not dead. There are specific scenarios where they outperform AI:

Event-triggered outreach. When you are emailing people who just took a specific action — signed up for a trial, attended a webinar, downloaded a whitepaper — the trigger itself is the personalization. A clean, well-crafted template that references the trigger can outperform AI because the relevance is already built in.

Very narrow audiences. If you are targeting 50 CFOs at Fortune 500 retail companies, you can afford to hand-write killer templates for that specific audience. The AI advantage diminishes when a human writer deeply understands a tiny niche.

Heavily regulated industries. In fields like healthcare or financial services, where every word in an outbound email may need legal review, templates that have been pre-approved by compliance teams are safer than AI-generated content that varies every time.

Simple, transactional offers. If your offer is dead simple — "We will audit your Google Ads for free" — a clear, direct template might outperform an AI email that adds unnecessary complexity.

When AI Clearly Wins

Large, diverse audiences. When your lead list spans multiple industries, company sizes, and use cases, AI can adapt the messaging for each segment without you creating dozens of template variants.

Competitive markets. When your prospects receive 10+ cold emails per day, AI personalization is often the only way to stand out. The template patterns are too recognizable.

Complex products. If your product solves different problems for different personas, AI can emphasize the relevant pain point for each prospect based on their role and company context.

Follow-up sequences. AI really shines in follow-ups. Instead of generic "bumping this to the top of your inbox" messages, AI can generate fresh angles that add new value in each touch.

The Hybrid Approach: Best of Both Worlds

The most effective teams we work with use a hybrid approach. They use templates as a strategic framework — defining the overall structure, tone, and CTA — then let AI handle the personalization within that framework.

For example, you might define a template structure like:

Opening line: AI-personalized reference to the prospect's company or role
Problem statement: Fixed (because you have tested and validated it)
Social proof: Fixed (specific case study)
CTA: Fixed (because you know what converts)

This gives you the reliability of templates where it matters (tested CTAs, proven value propositions) and the personalization of AI where it matters most (opening lines that earn attention).

ScrapenSend supports this hybrid approach by letting you lock certain sections of your email while AI personalizes the rest. You keep control over your messaging strategy without sacrificing relevance.

Making the Switch: Practical Steps

If you are currently running templates and want to test AI, do not switch everything at once. Run a parallel test:

Take your best-performing template and your target audience
Split the audience into two equal groups
Send the template to group A and AI-generated emails to group B
Run for at least 200 sends per group to get statistically meaningful data
Compare reply rates, positive reply rates, and meeting book rates

Let the data decide for your specific market. Industry averages are useful, but your audience may respond differently. The goal is not to prove one method superior in theory — it is to find what books more meetings for your business.

If you want to run this test yourself, create a free ScrapenSend account and set up both a template campaign and an AI campaign targeting the same ICP. Most teams see clear results within 2 weeks.