How to Automate Quote Follow-Up Without Sounding Like a Robot

Many quotes die from silence. The customer got busy, the quote sat in an inbox and nobody followed up because everyone was on a job. Following up is easy to describe and easy to forget, which makes it a good candidate for automation. The risk is that a badly built sequence sounds like a bot and puts people off. This guide shows how to automate the timing and the admin while keeping the voice human. If you run a trade business, automation for trades and construction covers more of the picture.

Step 1: Get a clean record of every quote

You cannot follow up on quotes you cannot see. Start with a single list showing, for each quote:

• Customer name and contact details.
• What was quoted and the amount.
• Date sent.
• Status: sent, viewed, accepted, declined, expired.
• Owner: the person who knows this customer.

If you quote through software such as Xero, MYOB, ServiceM8, Tradify or a CRM like HubSpot, this list already exists. If you quote by email and PDF, add a simple spreadsheet row each time you send one. The follow-up system needs this record to trigger from. If your quoting process is not written down yet, map it first.

Step 2: Decide your follow-up rhythm

Keep it short and spaced. A plain schedule might be:

When	Purpose
2 to 3 days after sending	Check it arrived and offer to answer questions
About a week after	Add something useful: a relevant photo, availability, a common question answered
About two weeks after	Final polite check, with a clear way to say no

Adjust to the size of the job. A small repair quote needs a quicker rhythm than a kitchen renovation that the customer will think about for weeks. Ask your team how long customers usually take to decide.

Step 3: Write the messages yourself

This is the step that stops the robot feel. Write each message by hand, in your normal voice, as if you were writing to one person you liked. Then replace the personal parts with fields: first name, job description, the date you sent the quote.

Two examples for a hypothetical landscaping business:

Day 3

Hi {first_name}, just checking the quote for the {job_description} came through okay. Happy to talk through anything, including the materials or the timing. Cheers, {sender_name}

Day 14

Hi {first_name}, I'll close off the quote for {job_description} on the {expiry_date} so I can keep my diary accurate. If the timing isn't right, no worries at all. Just let me know either way and I'll keep your details for when it suits. {sender_name}

Both are short, use plain words and give the customer an easy way to respond or decline. Compare that with "I am writing to follow up on the previous correspondence regarding your quotation".

Step 4: Add real information where you can

Generic nudges are easy to ignore. Fields from the quote make messages feel specific: the job, the suburb, the date, a particular line item. Where it fits, add one genuinely useful piece in the second follow-up, such as the answer to the most common question customers ask about that kind of job.

An AI model like ChatGPT or Claude can help you draft variations of these templates for different job types. Read each one and edit it until it sounds like you. Treat the AI draft as raw material and finish it yourself. See also can AI write quotes and proposals.

Step 5: Set the triggers and the stop rules

At the concept level, the workflow is simple:

1. Trigger: a quote's status becomes "sent".
2. Wait the set number of days.
3. Check the status. If it is still "sent" with no reply, send the message.
4. Repeat for the next step in the rhythm.
5. Stop the moment anything changes.

Zapier, Make and n8n can all run this, and many quoting tools and CRMs have a built-in version (compare the three). Check what your existing software offers before building something separate. For the underlying approach, see how to set up an enquiry triage workflow.

Stop rules matter more than the schedule. End the sequence when:

• The customer replies by any channel.
• The quote is accepted, declined or marked closed.
• A person pauses it for that customer.
• The customer unsubscribes or asks you to stop.

Step 6: Send from a real person and make replies easy

Send from the owner's name and address, not from a "no-reply" address. Make sure replies land in a real inbox that someone checks. Nothing sounds more like a robot than a message you cannot answer.

For work done by a person on a first-name basis, send plain text with a normal signature. Skip banners, logos and heavy formatting. A message that looks like a note from a colleague gets read more often than one that looks like a campaign.

Step 7: Keep a person in the loop for the valuable quotes

Not every quote deserves the same treatment. Add a rule: for quotes above a certain value, or from repeat customers, the system creates a task for the owner to call instead of sending an email. Calls on larger jobs build trust in a way a message cannot.

Where AI drafts any part of a message, a person should read it before it sends. See keeping a human in the loop on AI output.

Step 8: Respect the rules on commercial messages

Australia's Spam Act covers commercial electronic messages. It has requirements around consent, identifying the sender and offering an unsubscribe option. Whether a follow-up on a quote someone requested falls inside the rules can depend on the circumstances. Check with the regulator's guidance or an adviser. This guide is general information, not legal advice.

Step 9: Review what happened

After a month, look at the list:

• How many quotes were accepted after a follow-up compared with before?
• Did any customer complain about the messages?
• Which message in the sequence drew the most replies?

Adjust the wording and timing. If you have a baseline from before you started, comparison is simple. If not, our guide to measuring whether an AI change is working explains how to build one.

Copyable checklist

[ ] Every quote recorded with status, date, owner
[ ] Follow-up rhythm agreed (e.g. day 3, day 7, day 14)
[ ] Messages written by a person in their own voice
[ ] Fields inserted: first name, job, date
[ ] Stop rules set: reply, accepted, declined, paused
[ ] Sent from a real address that is monitored
[ ] High-value quotes create a call task instead
[ ] Spam Act requirements checked
[ ] Review booked for one month after launch

What to do next

If you want the sequence built into the tools you already use, with the wording tuned to your business, workflow automation from 21Labs begins with a free discovery. Send the form below and we reply within 24 hours.