Blog · Published · Updated
AI Agent for Email Personalization: Research and Review
Short answer
An AI agent in outbound gathers facts about the recipient's company from public sources, writes a personalised email draft with the source of each fact and hands it to a salesperson. A person approves the sending, and you settle the legal basis for contact before the start. At Syntalith the agent costs from €3,000 net, and with 600 emails a month at 10 minutes each it pays back in 1.3 months.
How we workDetails ↓
- 01Call30 minutes with an engineer, free
- 02PrototypeA working system on your process and data, free
- 03ImplementationTarget and fixed price in the contract; 50% at signing
- 04Acceptance50% once the agreed target is met
- net per agent for one process, build paid once
- from €3,000
- from signing the contract to acceptance
- 6–16 weeks
- free prototype on your cases
- €0
We belong to the OpenAI and Anthropic partner programmes and learn through their official courses. What this means ↓
Author: Syntalith · 3 min read
Statistics
Sales and AI: statistics 2026
1.3 mo.
payback of a €3,000 agent at 600 emails a month: €3,490 of work today, €1,257 with checking and the model. Scenario.
40%
According to a study published in Science (2023), ChatGPT cut the time 453 professionals spent on writing tasks by 40%.
Source: Noy and Zhang, Science (2023) ↗
34.7%
According to Eurostat (2025), 34.7% of EU companies that use AI use it for marketing or sales.
Source: Eurostat, use of artificial intelligence in enterprises (2025) ↗
48.8%
According to Eurostat (2025), 48.8% of EU companies that considered AI and did not use it named concerns about data protection and privacy.
Source: Eurostat, use of artificial intelligence in enterprises (2025) ↗
90%
According to the European Commission (2026), 90% of companies using AI say that people take the decisions, and 2% of decisions are fully automated.
Source: European Commission, Flash Eurobarometer 578 (2026) ↗
Figures checked: 22 October 2026
Email personalisation at scale: the law and consent first, then the prompt
A salesperson writes 30 emails a day to new companies. Each one needs a look at the website, the latest announcements and the person, so the message does not read like a template. After three weeks the salesperson switches to a template because there is no time, and replies drop.
Before the agent writes its first email, you settle the legal basis. In Poland, for example, commercial information sent electronically to a person who has not agreed breaches the Electronic Communications Law, and processing contact data needs a basis under the GDPR, for example a legitimate interest assessed with a balancing test. In B2B practice this often means a first contact that asks for consent, or contact with company addresses within limits agreed with a lawyer.
- 01Listcompanies and people with a documented basis for contact
- 02Factspublic sources: the website, announcements, job ads, with a date
- 03Draftan email that refers to one verifiable fact
- 04Checkthe salesperson checks the fact, the tone and the consent
- 05Sendingwith a daily limit and opt-out handling
- No fact without a source and a date
- Two reminders at most, an opt-out respected at once
- Data from public sources deleted after the campaign closes
- No automatic sending without approval
What an outbound email agent costs and when it pays back
- Today
- 600 emails × 10 min = 100 h (€3,490)
- With the system, a person checks the output in 35% of the time
- €1,222
- Model, 3 calls per case, Claude Sonnet 5.5
- €35
- Cost of one case
- €5.82 today, €2.10 with the system
- Saving
- €2,233 a month
Payback of €3,000
1.3 months
The volume, the time and the checking share are the scenario's assumptions. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat. Model: 8,000 input tokens and 600 output tokens per call at Anthropic's rates, converted at the ECB rate of 1.1205 USD per euro on 22 October 2026.
The count covers the work around each case: searching, reading, writing and handing over. What remains is the time to check the result, 35% of today's work in the scenario. Before the quote we measure both times on 20 to 100 cases the team has already handled, in a free test that takes 5 working days.
When not to roll it out
With a few dozen emails a month, a salesperson will write them better alone. When you have no legal basis for contact, or a list of unclear origin, the agent only speeds up a breach. When the offer does not fit the recipients, personalisation will not help. According to a study in Science (2023), ChatGPT cut the time spent on writing tasks by 40%. In outbound, checking the facts takes more time than the writing.
A few dozen emails a month
a salesperson with a good template
no build
A few hundred emails, one group of recipients
ChatGPT or Claude with instructions and a list of sources
training from €1,150
Hundreds of emails, several segments, records in the CRM
an agent that prepares drafts for approval
from €3,000
Where to start
The pilot covers one segment and 100 drafts compared with the salespeople's emails for the accuracy of the facts, replies and opt-outs. We build this kind of system as part of our AI agents. For the free process review, a description of one case from the last few weeks and the tools it went through is enough. You do not need to send confidential documents, and you receive a written plan within 2 working days.
Worked example · Scenario
How we count the payback: CRM in B2B sales
We run the same count for your team in the free process review.
€698
a month for entering 300 CRM updates by hand
Payback: 5.1 mo.
4 minutes per CRM update and checking in 15% of the old time are our assumptions. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat, for companies with 10 or more employees. Model cost: Claude Sonnet 5.5, about 3,000 + 500 tokens per update, at the ECB rate of 1.1205 USD per euro on 22 October 2026. The extra margin is a scenario: we assume that 1 in 400 updates that get lost today ends in an order, with a margin of €459. In the review we count it on your sales process.
See the full calculation →
