Sales
AI for sales follow-up emails
Give the AI three things: your notes from the call, the next step you agreed, and two or three of your best past follow-ups as style examples. Ask for a short email that confirms what was discussed, answers open questions and proposes one clear next step. Edit it, check every fact and date, and send it the same day.
Effort:
Low
Time saved:
Often 30–50% less writing time per email; measure it in a pilot
The problem today
Follow-ups after calls and meetings slip because a good one takes real effort and the next meeting is already starting. Late or generic follow-ups let warm leads go cold.
How it works, in plain words
It works like a junior colleague who heard the call, read your best emails and hands you a draft to finish.
What goes in: your rough notes from the call, the agreed next step, and three to five of your best past follow-ups as style examples.
What the AI does: it turns the notes into a short email in your structure and tone: thanks, a summary of needs, answers to open questions and one next step with a date.
What a person checks: you check every name, price, date and promise against your notes and the CRM, then adjust the wording to sound like you.
What comes out: a follow-up ready to send the same day, logged in your CRM.
It only knows what you give it. If a detail is missing from your notes, it may fill the gap with something plausible but wrong.
How it could work for businesses like yours
Illustrative examples, not client case studies.
A 10-person content marketing agency in Pagrati, Athens
Account managers run several discovery calls a week and often send the recap two or three days later. The agency would save one shared prompt with its best recaps and ask each manager to send the follow-up before the end of the day. They would track how many recaps go out within 24 hours and how many prospects reply.
A furniture showroom with 7 staff in Larissa
Customers visit, get a verbal quote for a sofa or kitchen, and rarely hear back. Staff would jot three bullets after each visit and use AI to draft a short email with the options discussed and a suggested second visit. A realistic first target could be a follow-up for every quoted visit, then a check on how many return.
A 20-person freight forwarder in Piraeus
Sales staff speak with importers about routes and timings, but follow-ups are long and arrive late. They would ask the AI for a short version that confirms the route, the open questions and the date for a formal quote. They would measure the time from call to email and the share of calls that lead to a quote request.
How to do it, step by step
Collect your best follow-ups. Pick 3–5 emails that led to a reply or a signed deal, remove client names and save them as style examples.
Write one reusable prompt: your role, tone, maximum length and structure (thanks, summary of needs, answers, one next step with a date). Save it where the whole sales team can find it.
Right after the call, write rough notes as bullets. Include the client's own words for their main problem. That is what makes the email sound specific.
Ask for two versions: a short one for busy decision-makers and a fuller one that lists the open points. Pick one and edit it in your own voice.
Check every promise, price, date and name against your notes and CRM before sending. AI fills gaps with plausible inventions.
Log the sent email in your CRM and set a reminder for the next contact. If there is no reply by the agreed time, ask the AI for a short, polite second follow-up.
Tools to consider
Microsoft 365 Copilot in Outlook: drafts and refines replies inside the email thread, using its context. Caveat: it only knows the thread and your files, not what was said on the phone.
Gemini in Gmail: the "Help me write" feature drafts emails inside Google Workspace. Caveat: features differ by plan and language; test the quality of Greek drafts.
HubSpot: AI email drafting inside the CRM, next to the contact record and deal stage. Caveat: drafts stay generic unless you add your own notes and examples.
ChatGPT or Claude: flexible for building one follow-up prompt with your examples and reusing it. Caveat: use a business plan that does not train on your data, and share only the client details the email needs.
Data protection and GDPR
Follow-ups contain personal data about named contacts and often confidential deal details. Use business versions of AI tools with a data processing agreement and no training on your inputs. Paste only what the email needs. Leave out personal phone numbers, internal margins and remarks about the person. B2B emails to existing contacts still fall under the GDPR and the e-privacy rules, so automated sequences need an easy opt-out. Review any automatic sequence before it runs. If you are unsure about a specific practice, check with your DPO or legal advisor.
Mistakes to avoid
Sending the draft without checking promised dates, prices or scope.
Letting every email sound the same, so clients recognise the template.
Pasting full call transcripts, including off-the-record remarks, into a chatbot.