Your Reviews Are Great. Your Lead Response Time Isn't.
Your Reviews Are Great. Your Lead Response Time Isn't.
Month two of Reeves Electric, I had an NPS of 4. Not 40. Not 14. Four out of 100.
The reviews weren't the problem. Early reviews from former Lonestar customers were fine — 4.9, genuine, earned. The problem was that when someone saw those reviews and called, something broke on the other end of that phone. I had no idea what, because I wasn't listening.
That's the trap most shops are sitting in. You've got 4.8 stars and 180 reviews and you're still not as busy as you should be, so you assume the answer is more reviews. Get to 200. Get to 4.9. Run the NiceJob drip harder. Meanwhile, 40% of the people who already trusted you enough to call are hitting voicemail or a two-hour text gap and booking somebody else before dinner.
The reviews are doing their job. Your intake isn't.
The Reputation Trap: Why Five Stars Isn't the Same as a Full Schedule
Reviews do one thing well. They clear the credibility bar. A customer who finds you on Google, sees 4.8 stars, reads a few recent ones, and decides to call — that customer has already made a trust decision without talking to you. That moment matters. But the moment they pick up the phone to call you, the reviews are done working.
Everything after that is operations.
Reviews cannot answer the phone. They cannot respond to the web form at 7pm. They cannot give a customer a booking window in 90 seconds the way a trained CSR can. Most shops don't know this is happening because they're not recording calls. They're guessing.
I was guessing in month two. I had seven years at Lonestar, two branded vans, a real website. What I hadn't built was an intake process — because I hadn't planned for the fact that I was now the person answering the 9:30pm call from the customer whose breaker kept tripping. When I finally started answering every call myself and taking notes on what was actually breaking, I found customers who called once, got voicemail, and didn't leave a message. Callbacks I was doing at 6pm when the customer had already moved on. Calls I answered where I fumbled the scope question and quoted a range so wide the customer didn't trust it.
None of that showed up in my star rating. The reviews were still coming in at 4.9. The business was underperforming.
What the Data Actually Looks Like If You're Tracking It
Most contractors can tell you their star rating within a decimal point. Ask them their call-to-book rate and the room goes quiet.
I know guys who will tell you confidently that most of their work comes from Google. They're not wrong — they're just guessing. They've seen the Google calls come in. They feel like it's most of them. That feeling is based on 80 or 100 jobs, which isn't nothing, but it isn't data either.
In my shop, I can tell you which channel produced which revenue — not which channel produced the most calls, but which one produced completed, invoiced jobs. That specificity changes decisions. When my LSA rep tells me to raise the budget, I have a number to check that against instead of a feeling.
CallRail is how I made that shift. You assign tracking numbers to every marketing channel — Google ads, LSA, your truck wraps, your direct mail piece — and every call gets recorded, transcribed, and attributed to the source. It runs roughly $50 to $150 a month.
On top of that, I built an Airtable table that connects job source to completed invoice. The fields are simple: call date, source channel, booked or not, job number if booked, invoiced amount. Takes my dispatcher five minutes to update after close of each job. It's not complicated. But it means my marketing decisions are coming from completed revenue, not call volume.
The attribution data tells me where leads come from. The recordings tell me what happens to them when they arrive. Both matter.
The People You're Losing After They've Already Decided to Trust You
There's a specific failure mode I want to name. Not the cold shopper who finds you on Google and calls three shops. That person is already comparison shopping on price and the margin on that job reflects it.
I mean the person who got your name from a neighbor, looked at your reviews, looked at your website, and picked up the phone. That person was sold before the call. When they hit voicemail, or get a text back two hours later, or get a callback at 5:30pm when they're making dinner — you didn't lose a lead. You burned earned trust. That's a different and more expensive failure.
LSA leads close at a lower rate because the customer is shopping. Referral and review-driven leads close higher because trust is already established. Which means when you fumble a referral on response time, you're losing a high-probability booking, not a long shot. The cost per missed call is not uniform across channels.
When I spent those first 90 days answering every call myself, I wasn't losing people on price. I wasn't losing them on reviews. I was losing them in the 20-minute window between when they called and when I called back. Some had already called someone else. A few left irritated voicemails that told me, very directly, what they thought of my response time.
The fix wasn't better reviews. It was a five-minute callback standard and a real intake script.
Contrarian Take: Your Review Count Is Probably Fine
NiceJob and Podium are fine tools. I run one. You should too. But neither one answers a call faster. Neither one follows up a web form in under five minutes. You can have a perfect review-request sequence and still have a broken intake workflow swallowing 30% of your inbound before those leads ever book.
If you have 150-plus reviews at a 4.7 or better, your next 15 reviews will move your booking rate by less than setting a response time SLA and holding your team to it.
The attribution problem is related. Most contractor marketing budgets run on feel — money going toward ads, review campaigns, and truck wraps with no clear read on what each channel is producing in booked revenue. If you can't tell me what your review-driven calls converted at last quarter, you're spending on faith.
I'm not saying stop getting reviews. But intake is where I'd put the next hour of attention, every time. It's not a product you can buy. It's something you have to build and hold.
The Fix Is Boring: Response Time, Scripts, and Recordings
My inbound SLA during business hours is five minutes. Call comes in, someone picks up or calls back within five minutes. Not "as soon as we can." Five minutes, tracked, and reviewed when we miss it. That standard, held consistently, did more for our booking rate than any single marketing change I've made since launch.
The script matters too. When I rebuilt intake in month three, I designed a five-question pre-quote phone screen. I started with EV charger calls because that's where scope variation was most dangerous — the difference between a panel-ready half-day install and a full service upgrade can be $8,000 and the customer doesn't know it. But the logic applies across call types. Before you roll a truck: property type, approximate age of the electrical system, panel location relative to the work, what the customer has already tried, and whether they own the home. Five questions, 20 minutes of call time, and you know what you're walking into. No one's surprised when the tech arrives. The quote is accurate. The job gets sold on scope, not on the customer's assumption.
The recordings are where improvement actually lives, though. The script is what you intend to do. The recordings tell you what you're doing. Those aren't the same thing.
Two booked calls, two unbooked calls, once a week, with your dispatcher or office manager. Tag three things on each: what created trust, what created friction, what question wasn't answered that should have been. You'll see patterns inside three weeks. Quality drift — where intake slips without anyone noticing — gets caught early instead of after it shows up in the booking rate.
Forty-five minutes a week. That's the requirement.
Three Things You Can Do This Week
Pull your missed call report. Whatever phone system you're running has this data. CallRail shows it clearly by channel. Your carrier can probably pull it too. What you're looking for: how many inbound calls in the last 30 days went to voicemail, and how many got a callback within 30 minutes. If you don't have that number, that's the first problem — before any marketing spend, before any review campaign, before anything else.
Listen to four calls before Friday. Two that booked, two that didn't. Listen for the moment the energy shifted — where the customer got hesitant, where the question they were actually asking didn't get answered, where the hold stretched too long. Write down one specific thing you'd change on each call. Don't build a spreadsheet yet. Just listen and write one thing.
Set a response SLA and tell your team what it is. Pick a number. Mine is five minutes during business hours. One-truck shop, maybe 15 minutes with a real voicemail script that tells the customer exactly when to expect a callback. The specific number matters less than the fact that there is one, everyone knows it, and someone is checking whether you're hitting it.
FAQ
I have a 4.9-star average with 200-plus reviews. Why am I still slow in the off-season?
Seasonality is real. Demand drops regardless of your rating. But the more useful question is whether your booking rate on the calls you're getting is where it should be. A slow season with a 70% booking rate is a different problem from a slow season with a 42% booking rate. Pull your call volume and booked jobs for the last two or three off-seasons and find the actual number. If you're closing well on the calls coming in, the issue is demand. If you're not, fix intake now — the busy season will pay for it.
What's an acceptable call response time for a residential service shop?
Under five minutes on business-hours inbound is what I run. After hours, the standard shifts — a real voicemail that tells the customer when they'll hear back, followed by a first-thing-in-the-morning callback, is workable. A generic voicemail and a same-day callback at 4pm is not. The shops with the best booking rates treat response time the way a restaurant treats table turn time. It's an operational metric, not a preference.
Is it worth hiring a CSR just to answer phones, or can I use an answering service?
Answering services are better than nothing. They take messages. They won't qualify scope, they won't handle an upset customer well, and they don't know your pricing or your service area. At five trucks, I'd hire a part-time CSR before I'd rely on an answering service. The CSR costs more per hour. Per call, it's not close. Use an answering service as after-hours overflow while you build toward real coverage — not as a permanent intake layer.
My tech is the one answering calls in the field. How do I fix that?
You fix it around him, not through him. A tech in a crawlspace should not be your intake layer. Forward the main line to someone who can actually take the call — a CSR, an answering service for overflow, a call queue with a hold message that sets expectations. Every call your tech handles from the field is a job running slower and an intake interaction that's worse than it should be. It's a structure problem.
I tried CallRail for two months and didn't see a difference. What am I probably doing wrong?
The tracking numbers are not the product. If you set up CallRail, assigned numbers, looked at the attribution dashboard, and called it done — you used maybe 20% of what it offers. The part that moves booking rates is the weekly call review. Sit down with whoever handles your phones, pick four calls, listen together, identify one specific thing to change. Do that for six weeks before you evaluate the tool. CallRail makes the listening possible. The listening is what changes the number.
Should I be responding to Google reviews?
Yes. Keep your expectations calibrated on what it does, though. Responding to a positive review is a minor signal that you're paying attention. Responding to a negative review matters more — prospective customers read the bad ones specifically to see how you handled it. A calm, factual response to a 2-star beats ignoring it. But no amount of review response activity makes up for slow callbacks. Given the choice between 20 minutes a week on review responses and 20 minutes a week on call recordings, I take the recordings.
Enjoyed this article?
Get articles like this in your inbox every Monday. Free, no spam.