Estimate follow-up automation for auto repair shops
Mike sends a $9,400 transmission rebuild estimate at 11 AM Tuesday. Customer says they want to think about it. Mike's service writer makes one follow-up call Wednesday afternoon and leaves a voicemail. Then nothing. Thursday the customer takes the car to the dealership, gets a quote in 90 minutes, and signs the work order Friday. Mike never hears back. The dealership didn't win on price — they won because nobody from Mike's shop touched the customer between Tuesday afternoon and Saturday morning. Most independent auto repair shops follow up on 10-20% of unaccepted estimates. The other 80% rely on the customer remembering to come back.
Why estimates die between Tuesday and Friday in independent shops
Auto repair has the highest comparison-shopping intensity of any home services trade for tickets above $1,500. When a customer hears 'your transmission needs $9,400 in work,' the first thing they do is call two other shops. They are not necessarily looking for a cheaper price — they are looking for confidence that the diagnosis is real and the number is not inflated. The shop that handles the comparison-shop window well wins the work. The shop that delivers the estimate and goes silent loses, regardless of whether their number was the most competitive.
The decision window is short. Cox Automotive research found that 5-minute response converts at 25-32%; same lead at one hour converts at 3-5%; after 24 hours conversion drops below 1%. The follow-up math works the same way. A customer who has an estimate in hand and has not accepted within 24 hours is actively comparing. Touch them again at 24, 72, and 168 hours and you catch them in the comparison window. Don't touch them and they drift to whichever shop touched them last — usually the dealership service department, which has automated this exact workflow.
Why 'the service writer will call them back' is not a follow-up system
Most independent shops handle estimate follow-up the same way: the service writer makes one call the day after the estimate, leaves a voicemail if no answer, and waits. When the shop is busy (which is most days), that one call does not happen at all. When it does happen, the voicemail is not returned 60-70% of the time. The follow-up does not fail because the service writer is incompetent — it fails because manual follow-up is the lowest-priority task in a service writer's day. There is always a customer on the phone, a tech needing a part number, a walk-in needing attention. Estimates from yesterday lose every priority contest.
Generic CRM follow-up does not fix this either. Most shops that have a CRM use it for marketing campaigns, not transactional estimate follow-up. The estimate sits in Mitchell 1 (or Tekmetric, or Shop-Ware), and the CRM has no idea it exists. Manually entering each unaccepted estimate into the CRM to trigger follow-up takes 5-10 minutes per estimate, which means it doesn't happen. The CRM ends up sending birthday cards and monthly newsletters while the $9,400 transmission estimate dies in the system that holds it.
What works is automation that watches the shop management system, detects estimates that have been sitting unaccepted past 24 hours, and fires a structured SMS sequence to the customer at 24, 72, and 168 hours. Each message is short, specific to the vehicle and the work, and includes a one-tap link to view the estimate or call the shop. No service-writer effort required. The estimate follows up on itself, in the customer's text inbox, while they are in the comparison window. Recovery rate climbs from near-zero to 15-25% on unaccepted estimates — which is the single highest-leverage revenue lever in an independent auto repair shop.
The four-component estimate follow-up architecture
Estimate follow-up automation is not one workflow — it is four interconnected components. Build them in order. Component 1 (the data export) is the foundation that everything else depends on; without clean estimate data flowing out of Mitchell 1 or your shop management system, the SMS layer has nothing to act on.
Component 1: Estimate data export from shop management system
The foundation. Mitchell 1 Manager SE has REST API access on newer plan tiers — older deployments need middleware (typically a $50-$200/month service like Steer or a custom connector). Tekmetric, Shop-Ware, and AutoLeap have cleaner APIs. The automation needs to read: estimate ID, customer name and phone and email, vehicle (year/make/model), line items with totals, estimate timestamp, and current status (open/accepted/declined). If your shop management system does not expose this via API, a daily CSV export to a watched folder works as a fallback. Without this data, nothing downstream functions.
Component 2: Workflow engine for sequence logic
Make.com is the dominant choice for shop owners building this themselves ($10-$30/month at the volume a 6-8 bay shop produces). n8n is the self-hostable alternative with lower long-term cost at scale but requires technical setup. The engine watches the estimate data feed, identifies estimates that hit the 24-hour-unaccepted threshold, and triggers the SMS sequence. Conditional logic handles the edge cases: estimate gets accepted between message 1 and message 2 → sequence stops. Customer replies STOP → all future messages cancelled. Estimate gets manually marked declined → sequence stops. Without conditional logic, customers get follow-up texts on jobs they have already accepted, which destroys trust.
Component 3: SMS delivery with 10DLC compliance
Twilio is the developer-friendly default ($0.0083/SMS) and the foundation under most independent shop SMS stacks. OpenPhone is the turnkey alternative if you do not have technical staff ($19-$33/user/month). Federal 10DLC SMS registration is mandatory for business texting and takes 2-4 weeks to approve — start the registration before the build. Without 10DLC, carriers will block your messages, often silently, and your follow-up automation will silently fail. Message templates need to be specific (vehicle, work, total) without being pushy. The shops that get the SMS tone right see 15-25% estimate recovery; the shops that send aggressive sales texts get blocked or generate complaints.
Component 4: Service-writer escalation on engagement
When a customer responds to the SMS sequence (any reply, not just acceptance), the automation pings the service writer's queue. The follow-up SMS handles the cold-touch problem; the service writer handles the warm conversation that follows. Service writer sees a notification: 'Customer Jane Doe replied to her 2017 Camry transmission estimate follow-up — she asked about financing options.' Without the escalation layer, customer replies sit in a shared inbox that nobody monitors, and warm leads go cold within 4-8 hours. The SMS automation gets you the conversation; the service writer closes it.
What estimate follow-up automation is worth
Numbers below are for a typical 6-8 bay independent shop ($1.5M-$2.5M annual revenue) sending 40-60 estimates per month at $1,500-$5,000 average ticket. The math scales linearly above and below this size, but the ratios hold. The highest-leverage shops are those with high comparison-shopped ticket volume — transmission, head gasket, AC compressor, and suspension work.
ROI ranges based on Cox Automotive lead response research, Harvard Business Review and MIT lead response time studies, DAS Technology dealership response data (NADA 2025), and aggregated independent shop operator interviews verified May 2026. Specific lift varies by current estimate-acceptance baseline (shops with very high baseline acceptance see smaller lifts), ticket mix (high comparison-shopped ticket shops see larger gains), and SMS message quality. Shops with average baselines and tight execution land in the middle of the ranges shown.
Four implementation gotchas
Estimate follow-up automation deployments fail for predictable reasons. These four show up most often in independent auto repair shops.
Skipping 10DLC registration before launch
Federal SMS compliance rules (the 10DLC framework) require businesses to register their phone number and use cases with carriers before sending automated text messages. Skipping this gets you carrier-blocked, often silently — your messages never reach customers and you never get an error. Twilio, OpenPhone, and most SMS platforms handle 10DLC registration in onboarding, but it takes 2-4 weeks to approve. Start the registration the day you start scoping the build. The shops that try to ship the automation in week three discover the messages are not actually delivering until customers do not respond.
Sending the same follow-up message to every customer
Generic SMS messages — 'Hi, this is Reyes Auto, just following up on your estimate' — convert at half the rate of specific messages. The follow-up has to reference the vehicle and the work: 'Hi Jane, following up on the 2017 Camry transmission estimate from Tuesday. Happy to answer any questions or walk you through the timing — just reply or call.' Specificity signals that a human cares about the customer's specific situation. Generic templates signal that the customer is one of 200 in a marketing blast. The message-template work is not optional; it is most of the conversion-rate difference.
No stop condition when the estimate is accepted offline
Customer calls the shop directly and accepts the estimate by phone, but the service writer forgets to mark the estimate accepted in Mitchell 1. The automation keeps firing follow-up texts on a job that is already booked. Customer gets annoyed and complains. The fix: any acceptance-related action — phone acceptance, deposit collection, parts ordering — has to flip the estimate status in Mitchell 1. If the service writer does not reliably do this, the automation needs additional triggers (parts order created = stop sequence) to catch the edge cases.
Treating estimate follow-up as a marketing channel
Some shops route the follow-up SMS through their marketing CRM and let it bleed into promotional content — 'Don't forget about your transmission estimate, and BTW we are running 10% off brake jobs this month.' This kills response rate. Estimate follow-up is transactional, not promotional. The customer is making a $9,400 decision, not browsing for deals. The message should be focused, brief, and helpful. Mixing in promo content tells the customer this is just another marketing list and trains them to ignore the next message.
Questions auto repair shop owners ask before building this
Five questions independent shop owners ask most when considering estimate follow-up automation for the first time.
We are on Mitchell 1 Manager SE on a legacy plan with no API access. Can we still do this?
Yes, but expect 1-2 extra weeks of build time and a $50-$200/month middleware cost. Services like Steer extract Mitchell 1 estimate data via screen-scraping or CSV exports and feed it to the automation layer. The alternative is upgrading to a Mitchell 1 plan with REST API access, which usually costs less than the middleware long-term but requires a plan migration. If you are on Tekmetric, Shop-Ware, or AutoLeap, API access comes standard and the integration is cleaner.
What if customers respond and we do not catch it fast enough?
Build the service-writer escalation layer in Component 4 with redundancy. Replies should ping the service writer via two channels: their SMS-receiving phone number AND a Slack channel (or Microsoft Teams, or email) that the office monitors during business hours. Two-channel notification reduces missed-reply rate from 30-40% (single channel) to under 5% (dual channel). Set the expectation that responses get touched within one hour during business hours. Customers replying to the follow-up are in the warmest part of the funnel — letting them go cold loses the work the automation just opened up.
How many follow-up messages is too many? Will customers feel spammed?
Three is the sweet spot. 24 hours after estimate, 72 hours, and 7 days. The shops that send 5-7 messages over two weeks see opt-out rates climb sharply and Google review complaints. The shops that send one and stop do not recover most of the estimates. Three messages — short, specific, helpful — is the band where recovery rate is high and complaints stay low. Each message also has to give the customer an easy out (reply STOP, call to decline, link to view estimate one more time). The escape hatch matters; customers who feel trapped on a list become hostile, customers who can leave easily mostly do not.
Will this work on older customers who do not text?
Mostly yes, with a caveat. For customers over 65, response rates on SMS run about half of what they are for under-65 customers — but they are still 5-8x the response rate on voicemail follow-up. The customers who genuinely do not text are about 8-12% of an average shop's customer base, and they will either ignore the messages (no harm done) or call the shop directly (which routes them to the service writer normally). The downside scenario — older customer ignores SMS and feels alienated — is rare and worth the upside on the 88-92% who do engage with SMS. For shops with very heavy older-customer mix (over 40%), pair the SMS sequence with a backup human call at the 72-hour mark.
How quickly can we get this live? We tried a shop management tool in 2020 that took six months and never worked.
Realistic timeline is 4-6 weeks from scoping to live, assuming Mitchell 1 API access (or middleware in place) and 10DLC registration started day one. The shops that miss this timeline almost always miss it because the service writer was not involved in scoping — they get blindsided by the new system and quietly work around it. Bring the service writer into the first scoping meeting. Show them how their workload changes (fewer manual calls, more warm leads in queue). The 2020 shop management tool failure was almost certainly a tech-adoption problem at the bay. Estimate follow-up automation runs in the background and does not touch tech workflow at all — different failure surface, much faster path to working state.
Continue reading
Find out what's actually right for your business
Industry pages get you most of the way. The real question is whether the workflow you'd build on this stack is genuinely the highest-leverage thing your business should be automating right now. The audit looks at your operations and shows you what to fix first, in plain language, without selling you anything.
No credit card. No follow-up call unless you ask.