How to Stop Losing Customer Enquiries in Your Inbox

KaamWithAI Editorial · 10 November 2026 · Approximately 8 minute read

A customer asks for a quote on Monday. You read the message between two jobs, tell yourself you will reply after lunch, and then the afternoon disappears. On Wednesday, you find the message again. You meant to follow up. The customer has already booked someone else.

This happens in businesses that are otherwise good at their work. Enquiries arrive through a website, email, Instagram, WhatsApp and phone calls. Each channel has its own inbox. The owner or one busy employee is expected to remember which person needs a reply, which quote is waiting for a decision and who should be contacted next.

You can improve this with a small system. It does not need a CRM project or an AI agent. First make sure every enquiry has a place, an owner and a next step. Once that works, AI can help with the parts that involve reading and writing.

Start by deciding what counts as an enquiry

For a home cleaning company, an enquiry might be a website form, a WhatsApp message asking about a deep clean, or a missed call from a new number. For a tax consultant, it might be an email asking about filing fees. Your definition should include every route through which a potential customer can ask for work.

Write those routes down. If two people answer the same WhatsApp account, decide who owns new messages and how the other person can see what has already been handled. A shared inbox is useful, but even a simple agreement about ownership is better than assuming someone else replied.

Give every enquiry one home

At low volume, a Google Sheet is enough. Create one row for each new enquiry and use columns like these:

The two fields people tend to skip are owner and next action. “Follow up” is not a useful next action. “Call Priya on Thursday to confirm whether the 3-bedroom clean is still needed” is.

Keep the sheet boring. Use one header row, avoid merged cells and don't casually rename or move columns if an automation depends on them. Zapier's Google Sheets setup guidance explains why those changes can break connected workflows.

Agree on what happens after a message arrives

Here is a workable routine for a small service business. Adjust the timing to your opening hours and the kind of service you sell.

1. When an enquiry arrives: Add it to the sheet, record where it came from and assign an owner. If you cannot reply properly yet, send a short acknowledgement that sets an honest expectation about when someone will respond.
2. When someone replies: Record what was discussed and the next action. If you need the customer's address, measurements or preferred date before quoting, write that down.
3. When you send a quote: Record the amount, date sent and when you plan to check back. Do not leave the follow-up date in someone's memory.
4. When the follow-up date arrives: The owner contacts the customer, updates the status and records the result. If the customer asks for no further contact, respect that and close the loop.
5. At the end of each day: Check that new enquiries have an owner and that nothing due today is sitting untouched.

Follow-up should be useful, not relentless. A customer who has not replied may be busy, comparing options or no longer interested. One relevant reminder can help. Repeated generic messages can make your business look careless.

A message you can actually send

Suppose a customer asked for a kitchen renovation quote. You sent an estimate two days ago and have not heard back. A reasonable follow-up is short and gives them an easy way to respond:

Change the message to reflect the actual conversation. Do not send it if the customer has already replied, declined, or asked you not to contact them.

Where a simple automation helps

Once the sheet and routine are working, remove the small bits of copying that happen every day. For example:

• A website form creates a new row in the enquiry sheet.
• The owner receives a notification when a new enquiry is added.
• A follow-up date creates a task or reminder.
• When a row is marked “won” or “closed”, the reminder stops.

These are ordinary if-this-happens, do-that rules. In a tool such as Zapier, a workflow starts with an event—such as a new email or spreadsheet row—and then performs actions you choose. Its guide to how a Zap works explains the trigger-and-action model. You can also see a worked example that creates a lead row and email drafts.

Do not automate every channel on day one. Start with the source that brings in the most enquiries. Test it with a few real submissions. Check that names, phone numbers and message details land in the right columns. Make sure a duplicate submission does not create a mess. Keep a manual way to add an enquiry if the connection stops working.

What AI can take off your plate

AI is useful when the incoming message is messy or written in a different way every time. A customer might write, “Need someone to come look at the leaking tap, preferably Friday afternoon, I'm near Indiranagar.” A simple form may not capture all of that neatly.

You can ask an AI tool to turn the message into a draft record:

A person should check the details before the record is saved or the message is sent. AI can misunderstand a location, turn a tentative preference into a confirmed appointment, or make up a detail that was never provided.

Keep its job narrow: organise the customer's words, identify missing information and prepare a draft. Your booking calendar should determine availability. Your price list should determine prices. A person should handle complaints, unusual requests and promises that could cost the business money.

If messages contain home addresses, phone numbers or other customer details, check the privacy terms and access settings of any AI or automation service before sending that information to it. Collect only what you need, and do not use a personal account that former staff can still access.

When you need a CRM

A spreadsheet starts to become awkward when several people need to work on the same enquiries, customers contact you repeatedly, or you need a reliable history of calls, quotes and conversations. That's when a customer-management tool may be worth paying for.

Move when you can name the problem the new tool will solve. Perhaps two employees keep contacting the same customer. Perhaps quotes are going out but nobody knows which ones are waiting for a decision. Perhaps you need to see the full history when a customer calls back six months later.

Don't buy software because a sales page says every growing business needs a CRM. If a sheet gives everyone a clear view and reminders are being completed, keep using it.

Check whether the system is helping

For the first month, track four simple numbers each week:

• How many new enquiries arrived?
• How many received a useful first response within your chosen service window?
• How many had a recorded next action and owner?
• How many became bookings or paid work?

Also note how many enquiries were missed, duplicated or followed up after the customer had already replied. These are signs that the process needs fixing.

Do not assume that a higher booking rate came from AI or automation. Compare with your own earlier results, account for busy and quiet weeks, and look at whether the new routine is actually being followed. The first win may simply be that fewer people are forgotten.

Try this before you buy anything

Take the last ten enquiries your business received. Put them in one sheet. For each one, fill in the owner, current status and next action. Then ask: could someone else look at this sheet tomorrow morning and know exactly what needs to happen?

If the answer is no, fix that first. Once the process is clear, automate the repetitive hand-offs. Add AI only where it handles language or sorting that would otherwise take a person meaningful time.

This guide uses a hypothetical home-services example. Tool capabilities and privacy terms change; check current documentation before connecting customer data. For a deeper look at setting boundaries around AI actions, read Your First AI Agent Should Have Less Authority Than You Think.