Five reports every custom home builder should read weekly.
Pipeline aging, job margin, ad ROI, change orders, and satisfaction. The five reports every custom builder needs, and why nobody has all of them.
A custom home builder does not fail from doing bad work. Most builders I know are excellent at the work. They fail from not knowing which part of the business is bleeding. A leak in one column of the ledger, a job going 15% over on a trade allowance, a lead sitting in the pipeline for 40 days that should have closed at 21, an ad channel quietly stopping , takes months to notice by feel. Weeks by report.
Every serious business runs on a weekly rhythm of five or six numbers. Custom home builders typically do not, not because they do not care about numbers but because the numbers live in five different systems and nobody has consolidated them.
Here is the list, one report at a time, and what it actually tells you when it lands on your desk Monday morning.
1. Pipeline aging by stage
Rows: every lead currently in the pipeline. Columns: the stage each lead is in, the date they entered the current stage, days since last contact.
The signal: leads that have been in the same stage for more than twice your typical time-in-stage.
Custom home leads sit for months. That is normal. What is not normal is a specific lead sitting in “proposal sent” for 40 days when your average is 12. Those are the deals where a competitor is quietly winning, or the homeowner has decided against building this year and no one has told you. The report says which conversations to have this week.
What kills this report today: leads live in an inbox, or a whiteboard, or a spreadsheet, or a real CRM that nobody runs the aging query on.
2. Job margin by trade
Rows: every open job. Columns: budgeted trade cost, actual to date, variance percent, projected finish variance.
The signal: any trade currently more than 8–10% over budget on a job past 40% complete.
Trade overruns compound. A framing overrun on Monday is a schedule overrun by Friday and a punch-list overrun in Q3. The builder who catches an 8% overrun at 40% complete can renegotiate, subcontract differently, or eat it deliberately. The builder who catches it at closeout has already eaten it.
What kills this report today: budgets live in QuickBooks (or a spreadsheet); actual costs land there weekly at best; variance gets calculated once a month by the office manager, if at all.
3. Ad spend by lead source
Rows: every paid channel the firm runs. Columns: spend, leads captured, cost per lead, cost per booked discovery visit, revenue attributed.
The signal: a channel where cost per booked discovery visit has drifted 20% or more over the last four weeks.
This is the report Angi and Houzz Pro will never write for you. It requires the ad platform (Google Ads or Meta Ads) to know which forms fired, and the CRM to know which of those leads booked. Most builders read the ad platform’s own report and stop there. That report is the map, not the territory. It counts clicks and lead-form fills. It does not know whether the lead was worth the click.
What kills this report today: the ad platform and the CRM are separate systems. Nobody wires the conversion event.
4. Change order log for the quarter
Rows: every change order signed this quarter, plus every one in flight. Columns: job, homeowner, requested date, signed date, dollar amount, category (owner request / design change / site condition / trade error).
The signal: category share drifting. If “trade error” is climbing quarter over quarter, that is a specific person or a specific process leaking.
Change orders are the truest picture of a firm’s build discipline. Every custom home has them. What matters is who requests them and why. A homeowner-request-heavy month means the intake conversation is under-scoped. A trade-error-heavy month means field discipline is slipping. The report says which meeting to have.
What kills this report today: change orders are logged in a subfolder of a job folder, one PDF at a time. Nobody aggregates.
5. Post-completion satisfaction and referral
Rows: every job completed in the last four quarters. Columns: completion date, walkthrough smooth (yes/no/deferred), punch list finished within 30 days, homeowner still responding to a followup email, referrals attributed.
The signal: a completed job that has not generated a referral within twelve months.
A custom home builder’s next twelve months are built on the last twelve. This report is the leading indicator of the referral pipeline that most builders realize too late is empty. It is also the only report that says whether the closeout process, the punch list, the warranty visits, the six-month check-in, is working, or is happening in principle only.
What kills this report today: nothing. It just does not get written. There is no system where “job completed” and “referral received” live in the same row.
Why nobody has all five
Because the five reports live in five different systems.
- Report 1 needs the CRM to be real and used.
- Report 2 needs the budget system and the accounting system tied to the job.
- Report 3 needs the ad platform and the CRM to be wired together.
- Report 4 needs the change-order process to be logged in a database, not a folder.
- Report 5 needs post-completion contact tracked as data, not remembered as anecdote.
A builder running Buildertrend for jobs, Google Sheets for leads, an agency for ads, QuickBooks for books, and Gmail for follow-ups will never have all five reports on the same desk. Which is why builders do not have them.
The one-platform argument, briefly
The reason Yglu ships as four modules on one login is that these five reports (and the twenty like them) fall out for free when the marketing, the leads, the job, and the finances share a database. Every module you take off the platform is one report you lose the ability to run without stitching.
If you already have four vendors and cannot get any of the five reports above on demand, the fastest fix is not another vendor. It is moving one system at a time onto a platform that can talk to itself. That is what the Yglu bundle exists to do.
What to do Monday morning
Pick one report from the list. The one you know is going to hurt the most to look at.
Run it, however painfully, from whatever spreadsheet stack you have today. Print it. Bring it to Monday’s meeting.
Do that for four weeks. That is the report. That is the meeting. That is the leading indicator you did not have on your desk.
The software choice comes later. The report comes first.
Written by Kol Dorney, founder of Yglu. Field Notes drops when there is something worth putting on the record, not on a publishing schedule.