how to create automated email templates with AI (saves hours weekly)

I used to spend at least 5 hours every week writing emails. welcome emails for new customers, follow up emails for leads, proposal emails for prospects, invoice reminders, you name it. most of these emails followed the same basic structure with minor variations, but I was writing each one from scratch like some kind of email masochist.

then I started using AI to create email templates and automated sequences. now I spend about 30 minutes per week on email, and the emails actually perform better because they are more consistent and optimized. in this guide, I will show you exactly how I set up my entire email template system.

why AI email templates work better than manual writing

let me be honest about something. when I first heard “use AI for emails,” I thought it would make my emails sound robotic and generic. the opposite happened. here is why.

when you write emails manually under time pressure, you cut corners. you skip personalization, you reuse the same tired phrases, and you sometimes forget important details. AI generated templates, when done right, are more complete, more personalized, and more consistent than what most people write in a rush.

the key is using AI as a starting point, not the final product. you still need to add your voice and personality, but the AI handles the structure, the key information, and the optimization.

step 1: build your core email template library

you need templates for the most common email types you send. here are the ones that save me the most time.

the essential email templates

email type	when to send	frequency
welcome email	new customer/subscriber	per signup
follow up (sales)	after initial contact	2 to 3 per lead
proposal	after discovery call	per opportunity
invoice	after project completion	per project
payment reminder	overdue invoice	as needed
thank you	after purchase/meeting	per event
re engagement	inactive customer (60+ days)	quarterly
newsletter	regular content update	weekly/monthly
meeting request	outreach or scheduling	as needed
feedback request	post delivery	per project

let me walk you through creating each type using AI.

step 2: create email templates with Claude or ChatGPT

the prompting technique that works

the secret to good AI emails is giving enough context. here is my template prompt structure:

“write a [email type] email for my business [business description]. the recipient is [who they are]. the goal of this email is [what you want them to do]. tone should be [describe your tone]. include [specific details to include]. the email should be [length]. sign off as [your name/title].”

template 1: welcome email

here is the prompt I use:

“write a welcome email for new customers of my data analytics consulting business. the recipient just signed up for our newsletter. the goal is to introduce our services and direct them to our most popular resource. tone should be warm, professional, and conversational. keep it under 200 words. sign off as Xavier, founder.”

Claude gives me something like:

subject: welcome aboard, glad you are here

hi [first name],

thanks for signing up. I am Xavier, the founder of [business name], and I wanted to personally welcome you.

I started this business because I saw too many companies sitting on valuable data without the tools or knowledge to use it. if that sounds familiar, you are in the right place.

here is the best place to start: [link to popular resource]

it covers the basics of turning raw data into decisions that actually move the needle for your business.

over the coming weeks, I will send you practical tips, case studies, and the occasional tool recommendation. no fluff, just stuff I have actually used and tested.

if you ever have questions, just hit reply. I read every email.

talk soon,
Xavier

I then tweak this to add specific details about my actual business and resources. the AI nails the structure and tone, I add the substance.

template 2: sales follow up sequence

for follow ups, I create a 3 email sequence:

email 1 (sent 1 day after initial contact):

prompt: “write a follow up email for a prospect I spoke with at a networking event. we discussed their need for data analytics. keep it short, reference our conversation, and suggest a 15 minute call.”

email 2 (sent 4 days after email 1, if no reply):

prompt: “write a second follow up email for the same prospect. this time share a relevant case study or result. no pressure, just value.”

email 3 (sent 7 days after email 2, if no reply):

prompt: “write a final follow up. acknowledge they are busy, give them an easy out (reply no if not interested), and leave the door open for future contact.”. for related reading, see automate customer feedback.

template 3: proposal email

proposals need to be personalized, so I use AI to generate the structure and fill in the specifics myself.

prompt: “write a proposal email template for a data analytics consulting project. include sections for: project understanding (2 sentences), proposed approach (3 bullet points), timeline, pricing placeholder, and next steps. professional but not stuffy.”

template 4: invoice and payment reminders

prompt: “write a friendly invoice email for a completed consulting project. include the invoice amount, due date, payment methods accepted, and a thank you for the business. keep it professional but warm.”

for overdue reminders, I have three escalation levels:

reminder	timing	tone
gentle	3 days overdue	casual, assume they forgot
firm	10 days overdue	professional, reference terms
final	30 days overdue	serious, mention next steps

AI generates all three with appropriate tone adjustments.

step 3: set up email automation in Mailchimp or ConvertKit

once you have your templates, you need to automate the sending.

email automation platform comparison

feature	Mailchimp	ConvertKit	Brevo	ActiveCampaign
monthly price (1K contacts)	free	$25/mo	free	$15/mo
monthly price (5K contacts)	$50/mo	$50/mo	$25/mo	$49/mo
automation builder	good	excellent	good	excellent
email templates	many	minimal (by design)	many	many
landing pages	yes	yes	yes	yes
tagging and segmentation	good	excellent	good	excellent
deliverability	good	excellent	good	very good
ease of use	moderate	easy	easy	moderate
best for	general marketing	creators, newsletters	budget option	advanced automation

my recommendation

for most solopreneurs, ConvertKit is the best fit. the automation builder is visual and intuitive, deliverability is consistently high, and the tagging system makes segmentation easy. I switched from Mailchimp to ConvertKit two years ago and my open rates improved by about 8%.

if you are on a tight budget, Brevo is excellent. the free plan handles up to 300 emails per day, and the paid plans are very affordable.

setting up a welcome sequence in ConvertKit

go to Automations > New Automation
set the trigger: “Subscribes to a Form” (select your signup form)
add your welcome email as the first step
add a 2 day delay
add a second email introducing your best content or offer
add a 3 day delay
add a third email asking what they need help with (drives engagement)
tag subscribers who open all three as “engaged” for future targeting

setting up an automated follow up sequence

create a new automation triggered by a tag (e.g., “prospect”)
send follow up email 1 immediately
wait 4 days
check if they clicked a link in email 1. if yes, send a different follow up. if no, send the gentle nudge
wait 7 more days
send the final follow up with the easy opt out

step 4: add AI personalization at scale

generic templates get ignored. personalized emails get opened. here is how to personalize at scale using AI.

merge field personalization (basic)

every email platform supports merge fields like [first name], [company name], and [location]. always use these. “hi Sarah” performs dramatically better than “hi there” or “dear valued customer.”

AI powered dynamic content

for deeper personalization, use AI to generate custom paragraphs based on recipient data.

store recipient information in your CRM or spreadsheet (industry, company size, pain points discussed)
use Zapier to pass this data to Claude or ChatGPT via API
have the AI generate a personalized opening paragraph for each recipient
insert the generated paragraph into your email template
send via your email platform

this works incredibly well for sales emails. instead of “I hope this finds you well,” each email opens with something relevant to the recipient’s specific situation.

example of AI personalized opening

for a lead in the retail industry:

“I noticed your team recently expanded your online store. that is a big move, and I have seen a lot of retail businesses struggle with the analytics side of scaling ecommerce. here is something that might help.”

versus the generic version:

“I wanted to follow up on our conversation. I think our analytics services could benefit your business.”

the personalized version takes 5 seconds to generate with AI and performs 3 to 4 times better in terms of response rate.

for more on this, see our guide on automate customer feedback.

step 5: optimize templates with data

do not just set and forget your templates. track performance and improve them over time.

metrics to track for each template

metric	what it tells you	target
open rate	subject line effectiveness	25%+ for cold, 40%+ for warm
click rate	content relevance	3%+
reply rate	engagement quality	varies by email type
unsubscribe rate	frequency/relevance issues	under 0.5%
conversion rate	template effectiveness	depends on goal

A/B testing your templates

create two versions of each template with different subject lines
send version A to 50% of recipients and version B to the other 50%
after 48 hours, check which version had higher open and click rates
use the winning version going forward
repeat with different elements: opening line, CTA, email length, send time

using AI to analyze email performance

every month, I export my email metrics and give them to Claude:

“here are my email performance metrics for the past month. identify which emails are underperforming and suggest specific improvements to subject lines, opening paragraphs, and CTAs.”

Claude consistently catches things I miss, like a subject line that is too long for mobile or a CTA that is buried too deep in the email.

my complete email automation stack

tool	purpose	cost
ConvertKit	email platform and automation	$25/mo
Claude Pro	template creation, personalization	$20/mo
Zapier	workflow connections	$20/mo
Grammarly	proofreading templates	$12/mo

total: about $77/mo. this handles all my email marketing, sales sequences, and transactional emails.

email template best practices I learned the hard way

keep it short: my best performing emails are under 150 words. people scan, they do not read. get to the point
one CTA per email: every email should have one clear action you want the reader to take. not two, not three, one
write like you talk: AI sometimes generates overly formal language. edit it to sound like you. “I wanted to reach out” is better than “I am writing to inform you”
test on mobile first: over 60% of emails are opened on mobile. if your email looks bad on a phone, it does not matter how good the content is
do not over personalize: there is a line between helpful personalization and creepy. mentioning someone’s company name is fine. mentioning that you saw they liked a specific LinkedIn post is unsettling
respect the unsubscribe: make it easy to opt out. hiding the unsubscribe link or making the process difficult hurts your deliverability and your reputation

faq

will AI email templates sound robotic?

not if you edit them properly. the trick is using AI for structure and optimization, then adding your personal voice and specific details. I spend about 5 minutes editing each AI generated template to make it sound like me. the result is better than what I would write from scratch in 30 minutes.

how many email templates do I actually need?

start with 5: welcome, follow up, proposal, invoice, and re engagement. these cover 80% of business email needs. add more as specific situations arise. I have about 25 templates now, but it took a year to build that library gradually.

should I use HTML designed emails or plain text?

for marketing emails (newsletters, promotions), HTML with your branding looks more professional. for personal communication (follow ups, proposals, feedback requests), plain text performs better. plain text emails feel more personal and tend to have higher reply rates.

can I use the same templates for B2B and B2C?

the structure can be similar, but the tone and content should differ. B2B emails can be slightly longer, more detailed, and reference business outcomes. B2C emails should be shorter, more emotional, and focus on personal benefits. I keep separate template libraries for each.

how often should I update my email templates?

review performance quarterly and update any template with declining metrics. also update templates whenever your product, pricing, or messaging changes. I do a full template audit every six months where I test new versions of every template against the current ones.