The quote-turnaround playbook: photo to estimate

Quote turnaround drops when the request arrives complete, the price comes from a written matrix instead of a person's head, every request follows one path with a time limit on each step, and follow-up runs without anyone remembering. This playbook does those four things in order, measured before and after, with software only at the end.

Why does turnaround matter more than people think?

Because the customer is not waiting for your quote. They are collecting quotes, and they decide when they have enough of them. The quote that arrives first sets the price the others are compared against, and the quote that arrives last often arrives after the decision. Speed is not a courtesy; it is position in the customer's order of consideration.

The reason quotes are slow is rarely the pricing. It is the waiting: waiting for the estimator to drive out, waiting for a photo the customer was never asked for, waiting for the one person who knows the rate for that material, waiting for someone to notice the request in the third inbox. Every one of those waits is a step in a path nobody wrote down. Write the path down and the waits become visible, and visible waits get fixed.

One opinion to argue with: for standard jobs, a consistent price sent today beats a perfect price sent Thursday. The matrix is what makes the fast price consistent, and consistency is what accuracy means once the same job stops being priced three different ways by three different people.

How do I measure turnaround today?

From the system, for every quote sent in the last quarter. Two timestamps per quote: when the request arrived, from the call log, the form inbox or the CRM, and when the quote was sent, from the quoting tool or the sent folder. The gap between them is that quote's turnaround. Sort all the gaps. The middle one is the number you carry; the longest ones are the ones to read closely, because each is a story about a wait.

Write the middle gap on line 1 of the worksheet with the date, and the three longest on line 2 with what each was waiting on. If your systems cannot give you the request timestamp, that is the first finding: the intake has no clock, and a path with no clock cannot be measured or promised. Add the clock, which is usually a field or a queue, and measure for a month before anything else.

This number is also your baseline. Every promise a vendor makes about quote speed is a promise against this line; if there is no line, there is nothing to hold anyone to. The measured turnaround for a business we have worked with is Details to follow, and yours is the only one that matters.

What should the photo checklist ask for?

Whatever your best estimator looks at before they price. Sit with them and one job type at a time, ask: what do you need to see? The answer is a list: the whole area from a distance so the scale is clear, close-ups of the damage or the surface, the access route in from the street, the panel or the meter, a tape measure in one frame, the material or the finish they want to match. Write that list per job type on line 3.

Then turn the list into the questions your intake asks. On the phone, the person asks for photos by text and reads the list. On a web form, the list is the upload prompt, one line per photo. By text, the request itself is a message that says what to send. A tool that helps is one your field team may already have; CompanyCam is one that keeps photos with the job. The checklist matters more than the tool.

The rule that makes this work: a request that arrives complete against the checklist can be priced without a visit, and a request that arrives incomplete triggers one ask, immediately, for exactly what is missing. Not a phone tag. One message, listing the gaps, sent inside the hour.

How do I write the pricing matrix down?

In a table, with your best estimator, over a few sessions. Columns are the things that change the price: the unit, the rate per unit, the minimum charge, the multipliers for access, height, distance and rush, the material tiers, the seasonal adjustments if you have them. Rows are the job types. Under the table, the exceptions: the cases where the estimator says "it depends," each with what it depends on.

Two things happen when you do this. The estimator finds rules they did not know they had, and two estimators find they disagree. Both are the point. The owner settles the disagreements, the table records the settled rule with the date on line 4, and from then on the same job gets the same price whoever quotes it.

If a rule cannot be written, it goes to the exceptions, and the exceptions go to a person. Over a few months the exceptions repeat, and each one that repeats becomes a row. The matrix is never finished; it is agreed, dated, and revised. Keep it where the office can read it, in the price book of your scheduling software if it has one, or in a spreadsheet with one owner.

What does the one path look like?

A request arrives through any door: phone, form, text, email, a marketplace. It lands in one queue with a timestamp. Whoever owns the queue that day checks it against the checklist. Complete: price it from the matrix, build the branded quote, send it through the channel the customer used, and log the sent timestamp. Incomplete: send the one ask, and the request waits in a separate column so it is visible. Needs a visit: schedule the visit at intake, with the quote pre-built from what is already known, so the visit confirms rather than starts.

Every step has a time limit, written on line 5, and one person owns the queue each day. The path is short enough to fit on one page and it is the same for every door. If a step is regularly late, that step is the next thing to fix; the measurement in the last section tells you which one.

The quote itself goes out from the system the office already uses, Jobber or ServiceTitan or Housecall Pro, so the customer sees the same branded document and the same accept button whether a person or a system built it. The job is created there from the accepted quote. That is the seam where software fits, and it fits only because the path was written first.

How does follow-up run on its own?

Write the rule on line 6: which day after sending the first follow-up goes, through which channel, how many messages, what each one says, and what stops it. The messages are short and plain: is there anything you need from us to decide; here is the quote again; we hold this price until a date. An accepted or declined quote stops the sequence the same minute, and the check for that runs before every send, not just when the answer arrives.

Then let a system send it. The point is not the messages; it is that no open quote depends on anyone remembering. The messaging can run through your scheduling software's own follow-up if it has one, or through a texting provider; Twilio is one. The follow-up rule you start with will be wrong in some way, and the way to find out is to track which step wins the job: the share of accepted quotes that were accepted after message one, after message two, after none. The rule that comes out of that tracking is Link coming soon, and it will not be the one you guessed.

Where does this go wrong?

The measurement is taken from memory and the number is what the owner hoped. The photo checklist is written but the intake never asks for it, so requests arrive as incomplete as before. The matrix is written and then one estimator keeps pricing from their head because the table "does not cover this one," and the exceptions never make it back into the table. Three doors feed three inboxes and the queue is a fiction. The follow-up is set up and then switched off the first time a customer says "stop emailing me," instead of the rule being changed.

And the one that costs most: software is bought first, to automate a path that was never written down, and it faithfully automates the confusion. Every item in this playbook before the last step is paper and decisions. Do them in order.

What do I measure after?

The same two timestamps, the same way, a month after the new path is running, on line 7. Compare the middle gap and the three longest. Then look at which step the longest ones waited on; that step is the next fix. Repeat monthly. When the number stops moving, the path is as fast as the people can make it, and the remaining gap is what software is for: pricing from the matrix without a person, sending the quote the minute the request is complete, and running the follow-up. Bring line 1 and line 7 to a working session and we will say whether the remaining gap is worth automating, and what the written commitment for it would say.