Five touches over ten days is a sensible default for an open remodelling quote: a day-1 text, a day-3 email answering the usual questions, a day-5 text about the start date, a day-8 email with a similar project, and a day-10 text closing the loop. Stop on any reply, and have AI draft each message for you to approve.
"Sensible default" is a judgement, not a statistic. Nobody can honestly tell you the right number of touches for every household, and anyone quoting a magic figure is guessing. What matters more than the count is that each message gives the household a reason to answer, that the sequence stops the moment they do, and that nothing goes out unseen. AI is good at the drafting and the personalising. The owner approves every message before it goes, and that split is what keeps the sequence honest.
The five touches and what each one is for
Day 0 is the day the quote is sent. The gaps run two, two, three and two days, so the household hears from you often enough to remember you and not so often that it feels like pressure.
| Touch | Day | Channel | The job it does | Opening line or subject |
|---|---|---|---|---|
| 1 | 1 | Text | Confirms the quote arrived and opens a two-way channel | "Did the [kitchen] quote reach you OK?" |
| 2 | 3 | Answers what households ask before signing: allowances, start date, payment stages | "Your [kitchen] quote: the three questions people ask most" | |
| 3 | 5 | Text | Raises the start date, the first real deadline | "Our [month] start window is open" |
| 4 | 8 | Shows a similar finished project and offers ways to adjust scope | "A project like yours, and two ways to adjust the quote" | |
| 5 | 10 | Text | Closes the loop politely and offers a later date | "I'll close your quote file on [date]" |
Each touch has exactly one question or one offer. A message that asks three things gets answered with silence, because the easiest reply to a crowded message is none. The order matters too: touch 1 lowers the barrier to replying, touch 2 answers objections before they are voiced, touch 3 introduces time, touch 4 offers a way forward if price is the obstacle, and touch 5 gives a graceful exit.
Wording to copy: texts under 160 characters, emails under 120 words
The wording below is deliberately plain. Placeholders sit in square brackets, and the samples use "[first name]" and "[your name]" because the real ones belong to your leads and to you. Every sentence is a promise the owner has to be willing to keep, so change anything that is not true for your firm.
Touch 1, day 1, text (140 characters as typed):
Hi [first name], [your name] at [company]. Did the [kitchen] quote reach you OK? Happy to talk through any line of it. Call or text anytime.
Touch 2, day 3, email. Subject: Your [kitchen] quote: the three questions people ask most
Hi [first name], three questions come up on most quotes like yours, so here are short answers. First, what is the allowance for? Allowances (cabinets, countertops, tile) are budgets for items you have not chosen yet. If your choices cost less, you get a credit. If they cost more, we show you the difference before ordering anything. Second, when could we start? Our next opening is [month], and cabinets currently take [number] weeks to arrive. Third, how do payments work? In stages, as set out on page [number] of the quote, and never ahead of the work done. If a call is easier, reply with a time that suits and I will ring you. [your name], [company]
Touch 3, day 5, text (136 characters):
Hi [first name], [your name] again. Our [month] start window is open. Want me to hold a date while you decide? No commitment either way.
Touch 4, day 8, email. Subject: A project like yours, and two ways to adjust the quote
Hi [first name], here is a recent project that is close to yours: [link to project page]. It had a similar [layout change], and the page shows what turned up behind the walls and how we handled it. If the total on your quote is the sticking point, I can price two or three variations, for example [option A] or [option B]. Tell me which matters most to you and I will send revised figures. What is the one thing you are still weighing? A one-line reply is plenty. [your name], [company]
Touch 5, day 10, text (146 characters):
Hi [first name], I'll close your quote file on [date]. If timing changes, reply LATER with a month and I'll check back. Thanks for considering us.
Two notes on the texts. Keep to straight quotation marks and no emoji. Twilio's documentation explains that one character outside the standard SMS set, such as an emoji, switches the whole message to a different encoding, which cuts the limit for one message from 160 characters to 70, and that longer messages are split into segments of 153 characters (67 in the other encoding), each billed separately. Curly quotation marks sit outside that set too, unless your platform converts them. If you text from your own phone, none of that matters. If a platform sends your texts, it does. Count characters after the names are in, because a long first name or company name can tip a message over the line.
The Complete Excel Course with AI Integration
From formulas and pivot tables to Power Query, DAX, VBA and AI with Copilot, ChatGPT and Claude — hands-on from the first lesson.
By Sagnik Bhattacharya and Virginia Thorn, Primetime Emmy Award winner
Buy nowStop rules, written down before the first message goes out
A follow-up sequence is only as good as the moment it stops. Write the stop rules first, because a message that arrives after the household has said no is the one they remember. These are the rules the sequence follows, in the order a person would check them.
| What happens | What the sequence does |
|---|---|
| Any reply, on any channel | Stops at once. The owner answers personally, and the sequence does not resume unless the owner restarts it. |
| A no, "not interested" or "we went with someone else" | Owner sends one thank-you, logs the reason and the sequence ends. |
| "Not yet", "after the holidays", "waiting on finance" | Sequence ends. The owner diaries the date the household gave and sends one message shortly before it. |
| STOP, unsubscribe or similar | All messages on that channel stop immediately, and the request is logged where the next person can see it. |
| Quote accepted or deposit paid | Sequence ends and the onboarding messages begin. |
| Bounced email or a number that fails | That channel stops. The other channel gets one try, then the lead is closed. |
| Upset or complaining reply | Everything automated stops. The owner phones. |
| Day 10 passes with no reply | The file closes. No sixth message. One diary reminder at about 45 days for a personal note, only if something has changed. |
Add timing rules to the same page. Set quiet hours (no texts before 8 a.m. or after 8 p.m. in the household's own time), allow at most one message a day, and skip any day the household said they would be away. Texting rules on consent, quiet hours and opt-out wording differ from place to place, and they depend on whether a message counts as marketing. Ask your adviser or your messaging provider's compliance guide, and make sure a STOP reply halts the sequence in whatever system sends it.
The prompt that drafts the sequence from a lead record
The prompt works from a lead record, not from memory. The record holds only facts you have already collected, and the rules forbid the AI from adding any. That single constraint is what keeps a draft honest.
You draft follow-up messages for a remodelling company's open quote. Use ONLY the fields in the LEAD RECORD.
Write five messages following the SEQUENCE. For each: the day, the channel, the message, and one line headed "If they reply" saying what the owner should do.
Rules:
- Texts: 150 characters or fewer including spaces, straight quotation marks only, no emoji.
- Emails: 120 words or fewer, with a subject line.
- One question or one offer per message.
- Never open with "just checking in", "touching base" or "hope you are well".
- Never invent scarcity, deadlines, discounts, awards, start dates or project facts. If the record has no start window, write [OWNER TO CONFIRM].
- Mention an objection only if the record lists it.
- Use -ise and -our spellings (organise, colour). Sign off with [your name] and [company].
- Finish with a table: touch, day, channel, character count (texts) or word count (emails).
SEQUENCE:
1. Day 1, text: confirm the quote arrived.
2. Day 3, email: answer three common questions using the record.
3. Day 5, text: raise the start window.
4. Day 8, email: link the similar project and offer scope options.
5. Day 10, text: close the file and offer a later date (reply LATER).
LEAD RECORD:
First name: [first name]
Project and quote total: kitchen remodel, $46,000
Asked about: countertop allowance, weeks without a kitchen
Objections so far: none
Preferred channel: email
Start window the owner can honour: January
Similar project link: [link]Here are the lines from a typical first draft that the owner would not send, written the way these tools tend to write them. They are illustrative, but each kind turns up regularly.
Touch 2 (email): "Sign this week and we will include under-cabinet lighting at no charge."
Touch 3 (text): "Hi [first name], our January slots are filling fast, so I wanted to check in. Shall I reserve one for you?"
Touch 4 (email): "I have attached photos of our recent work for you to look at."
Each one breaks a rule the prompt stated, which is the reason to read every draft:
- The free lighting is an invented incentive. The record contains no offer, so the owner deleted the sentence. A discount that is not yours to give becomes a promise the household will quote back.
- "Filling fast" is invented scarcity. The record says January is open, and nothing says it is nearly full. The owner replaced it with "Our January window is open", which is true and still gives a reason to reply.
- "Check in" sneaked into the text even though the prompt banned the phrase. Drafts drift, so scan for the banned openers.
- The attachment does not exist. The record held a link and no files. Emails that mention attachments which are not there make a firm look careless, so the owner changed the sentence to use the link.
The approval step catches all four in a minute, and that is the argument for approval in one sentence: AI drafts quickly and plausibly, and plausible is not the same as true.
Quote size changes the channel, not the count
Five touches suits most jobs. What changes with the size of the quote is which touches are voice calls, because a household spending $60,000 expects to hear a person.
| Quote size | Touch 1 | Touch 3 | Touch 4 addition |
|---|---|---|---|
| Under $15,000 (a bathroom update) | Text | Text | Link to a similar project |
| $15,000 to $50,000 (a kitchen, a larger bathroom) | Text | Phone call, with a voicemail and a text if there is no answer | Offer a short site visit to talk through options |
| Over $50,000 (a basement, a whole floor) | Phone call, then a text | Phone call | Offer a second visit with the site lead or designer |
Take three quotes as examples. A $14,500 bathroom follows the first row exactly: two texts, two emails and the close. A $46,000 kitchen swaps touch 3 for a call, which is where a question like "how long will we be without a kitchen?" tends to come out. A $63,500 basement starts with a call on day 1, because a quote that size has usually followed a conversation, and a text alone feels thin.
One special case needs different wording: the late start. If a quote is already more than a month old when you begin chasing, the first message must ask whether the project is still live before it says anything else. Something like "Hi [first name], I realise this quote is a few weeks old. Is the kitchen project still on your list?" gives them an easy way to say yes, no or later, and the stop rules then apply as normal.
The replies you will get, and a first line for each
A sequence that works produces replies, and replies are where the owner earns the job. Ask the AI for three versions of an answer, short, warm and detailed, then pick one and edit it. Never let it auto-reply: the owner should be the one who decides whether to move on price, scope or timing. These are the four replies that come most often, with a starting line the owner can adapt.
| They write | A first line to adapt | What the owner decides |
|---|---|---|
| "We are getting other quotes." | "Sensible. If it helps, I can walk through ours line by line, so you can compare like with like, especially the allowances." | Whether to offer a comparison call. Never criticise the other firm. |
| "It is more than we expected." | "Thanks for telling me. Which part feels furthest from what you had in mind: the scope, the allowances or the total? I can price two variations." | Which variations are worth pricing. Avoid an instant discount. |
| "We need to talk to my partner." | "Of course. Would a one-page summary help, with the allowances and payment stages on it, so you can both read the same thing?" | Whether to send a summary or offer to join a call with both. |
| "Can you do it for less?" | "I would rather change the specification than squeeze the labour. Shall I price [option] and [option]?" | What can move without harming the job. |
The second row is where most owners reach for a discount, and it is worth resisting. The household has not said that price is the only problem, only that the total surprised them. A question about which part of the quote is furthest from their expectation usually shows whether the issue is the scope, an allowance that looked low or the figure at the bottom. The AI can draft the variations once you have chosen them, and the tutorial on remodelling proposal templates with scope, allowances and terms shows how to word an allowance so it does not look like a hidden cost.
Sending: by hand, through your software, or both
Do not assume you must choose. Many quoting tools already send reminders, and the sensible move is to count them as touches. Jobber is one example whose rules are public. Its Automations help article says quote follow-ups are available on the Connect plan and above, and that you can set up to two reminders for quotes in "awaiting response" status. Each one goes out a number of days after the original quote, up to a maximum of 90, by the same method as the quote (so an emailed quote gets email reminders), at the same time of day the quote was sent. The household receives them only if "outstanding quote follow-ups" is switched on in their communication settings.
For this sequence that means setting Jobber's two reminders for day 3 and day 8, pasting in your approved wording for touches 2 and 4, and sending touches 1, 3 and 5 yourself. If you use a different tool, find out how many reminders it can send, what channel they use and what the default wording says, because the default is usually generic. Other quoting software is discussed in the tutorials on quote follow-ups that run themselves and following up silent quotes, and a longer nurture for big building projects is in AI nurture sequences for home builders.
Whatever sends, keep an approval column. A simple sheet has one row per message with columns for lead, touch, channel, draft, status (Draft, Approved, Sent, Stopped) and the date. Nothing goes out unless the status says Approved, and a reply or a stop rule changes every remaining row for that lead to Stopped. If a person is sending, the first step on every send is to check the inbox, because the most damaging message in the whole sequence is the cheery nudge that arrives an hour after the household phoned the office.
Ten open quotes, five touches each and one Monday block
Ten open quotes make a realistic week for a small firm, from a $14,500 bathroom to a $63,500 basement. Ten leads at five touches each is 50 messages in ten days. If the owner writes each one from scratch at about six minutes (find the quote, recall the conversation, write it, read it again), that is 300 minutes, or five hours, for a sequence nobody enjoys sending.
| Approach | Step | Working | Minutes |
|---|---|---|---|
| By hand | Write every message from scratch | 50 messages at 6 minutes | 300 |
| AI-drafted | Fill in each lead record | 10 leads at 2 minutes | 20 |
| AI-drafted | Read, fix and approve each draft | 50 messages at 1.5 minutes | 75 |
| AI-drafted | Total | 20 plus 75 | 95 |
| Difference | Time saved | 300 minus 95 | 205, about 3 hours 25 minutes |
Those timings are estimates, so time the first ten yourself and replace them with your own. The routine that makes the saving real is one fixed approval block on Monday morning, about 75 minutes on the figures above. The sheet lists every message due that week, each with its draft. The owner reads each one against the lead record, fixes what needs fixing and sets the status to Approved. Anything not approved by Monday noon does not go, and a quote sent midweek has its five messages approved on the day it goes out. Midweek, five minutes a day go on replies, because that is where jobs are won.
Track three things per lead and nothing more: which touch got the first reply, the outcome (signed, later, no or silent) and the reason given for a no. After a month, three questions are worth asking. Which touch gets the most replies? What reasons keep coming up? How many leads went silent after touch 5? If most replies land on touch 1 or 2, the later touches are doing little and the sequence can be shortened. If replies cluster after touch 4, the examples and options are what move people, and that is a hint to put more of both into the proposal itself. No tutorial can promise what your numbers will show, which is why you measure.
How follow-up sequences backfire
- Sending after a phone reply. The household rang the office on Tuesday and still got the day-5 nudge. The fix is an inbox and call-log check before every manual send, and a stop that fires automatically in software.
- Doubling up. Software reminders and manual touches land on the same day. Map the days once and write them on the sheet.
- Texting the wrong person. A joint enquiry can leave you with a partner's number, or a number from a contact form that belongs to a relative. Check who asked for the quote.
- Invented urgency. The AI will write "slots are filling" unprompted. If it is not true it breaks trust, and if it is true you should say how many.
- Identical wording for every lead. Add one real detail from the site visit, such as the awkward corner or the question the household asked, and the message reads like a person wrote it.
- Chasing too soon. A text an hour after the quote lands looks like pressure. Day 1 is early enough.
- Ignoring a stop request. One STOP reply that you miss turns a lead into a complaint.
For comparison, a sequence for a different trade with a long decision cycle is in the tutorial on an HVAC replacement follow-up sequence, which spaces its touches differently for a decision made under pressure. The remodelling version differs because the household is usually weighing months of disruption as well as money, which is why its messages answer questions about timing and payment before they ask for anything.
Questions about following up remodelling quotes
What if the household says yes on the phone but never signs?
Treat a spoken yes as the start of a new, shorter sequence. Send the paperwork the same day with a one-line summary of what was agreed, follow up once after three days, then phone. Do not order materials or hold a start date on a spoken yes alone, and log the date of each step so the gap stays visible.
Can a follow-up mention the date the quote expires?
Yes, once, and only the validity date already printed on the quote. A line such as 'The prices on your quote hold until [date]' in touch 4 or 5 gives a real reason to reply. Never invent a new deadline in a follow-up, and if supplier prices genuinely change, say what changed and by how much.
Who should the messages come from: the owner, the estimator or the office?
From the person the household met at the site visit, because a reply then goes to someone they recognise. If the office sends on that person's behalf, sign with their name only if they will answer replies themselves within a working day. Otherwise sign as the office and say who will call.
Further reads
- Window Replacement Leads: Reply Fast and Book the Survey With AI — Speed matters before the quote: reply to new enquiries fast.
- Following Up Unapproved Repair Estimates With AI — Following up unapproved repair estimates, a shorter-cycle sibling.
- How Much Does It Cost to Automate Lead Follow-Up With AI? — What automating lead follow-up costs, so you can budget it.
- Kitchen Remodel Estimate Template: Line Items Written With ChatGPT — Write the kitchen estimate that the sequence then follows up.
- Remodelling AI Estimator: Test It Against Your Own Past Jobs — Test an AI estimator against your own past jobs first.
- Can AI Draft a Remodelling Contract Template in Plain English? — What happens after yes: a plain-English remodelling contract.
- Getting More Reviews After Every Job With AI Follow-Ups — Ask for a review once the job is finished and paid.
- AI Lead Follow-Up for a Tint Shop: Quote Texts and Reminders — A three-text quote sequence for a window tint shop, when to send each one, when to stop, and the reminders that protect the fitting slot.
- Deck Builder Leads: What They Cost and How to Reply Fast With AI — Work out the most a deck lead can cost you, compare Thumbtack, Bark, Angi and your own site by cost per booked visit, and reply in minutes with AI drafts.
- Foundation Repair Leads: Screen Crack Photos With AI Before the Visit — Six intake questions, a four-photo request, an AI summary that describes but never diagnoses, an engineer-referral rule, reply scripts and the lead-cost sums.
- Loft Conversion Leads: Screen Roof Type, Head Height and Budget With AI — Five screening questions, a scoring rubric, AI prompts and a follow-up cadence that cut wasted survey visits for a loft conversion firm.
- Remodelling Company Website: Project Pages Written With AI From Photos — A before, scope, challenge and result template, a photo shot list, client permission wording and alt text, with one bathroom from 14 photos to a live page.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: Jobber help centre article on Automations (quote follow-ups), read 1 November 2026; Twilio documentation on SMS character limits and message segments, read 1 November 2026.