What a Monday briefing actually says: one sample digest, annotated line by line

Most product marketing shows you a cropped screenshot and asks you to imagine the rest. I want to do the opposite. Below is a full Guidepost digest, every line of it, and then I take it apart in front of you: where each number came from, why it landed where it did on the list, and why one item tells you what to do while another only tells you to look.

One thing first, because it matters. This digest is for Northside Comfort, our sample shop. Northside is fictional: a made-up HVAC and plumbing business with a made-up owner, invented so I can show you real behavior without borrowing a real customer’s books. I could have dressed up a screenshot with a logo and a testimonial. I won’t do that. If the product were hollow, this post would be impossible to write, so I would rather write it than fake it.

Everything below traces to Northside’s sample data: about 230 completed jobs, the matching QuickBooks invoices, and a quarter of Google and Meta ad spend. Your version reads your own Jobber or Housecall Pro and QuickBooks accounts. The figures here are the sample’s.

The digest, in full

This is what lands at 7:00 AM Monday. Read it the way an owner would, top to bottom, before dispatch.

What needs attention · Mon, 7:00 AM

---

Worth acting on · $11,700 at stake You’re owed $11,700, and it’s aging fast. Three commercial accounts are 60 to 90+ days past due: Riverside Apartments ($6,200), Bella Vista Cafe ($3,400), and Greenpoint Dental ($2,100). The longer they sit, the less likely they get paid. Sending reminders is the obvious move; the list is one tap away. From: QuickBooks invoices · Jobber jobs

---

Worth acting on · revenue Last month’s revenue dropped, and it was all plumbing. Revenue held near $22k a month all year (about $13k HVAC, $9k plumbing). In May it fell off: plumbing came in around $1k while HVAC stayed close to normal. That lines up with Tina Alvarez, your plumbing lead, being out most of the month. Not a pricing or demand problem. Here are the plumbing jobs that didn’t happen. From: Jobber completed jobs · service line

---

Worth a look · marketing Meta is outspending Google but bringing back less. You spent more on Meta than Google last quarter, but Google brought in more booked work per dollar. Meta’s return has cratered while Google held. Worth reviewing those Meta campaigns before next month’s spend. We won’t pretend to know which way to move the budget. From: Meta & Google Ads spend · Jobber bookings

---

Worth a look · quality Dwight Okafor has 3 callbacks this month. Your newest HVAC hire has more redo visits than the rest of the crew. Worth seeing what those three jobs have in common (equipment, parts, the install checklist) before anything else. New-hire ramp, most likely. Not a pattern yet. From: Jobber jobs flagged as callbacks

---

Since last time: HVAC revenue is holding near its yearly average, and your overall close rate hasn’t moved. The dip really is the plumbing gap, not the whole shop. Nothing else needs you this week.

Four items and a footer. That is the whole thing. Now the three passes.

Pass one: where every number came from

The rule for the whole product is that no line appears without its source. Here is what that means in practice, item by item.

The $11,700 is not a Guidepost estimate. It is the sum of three open invoices in QuickBooks: Riverside Apartments at $6,200, Bella Vista Cafe at $3,400, Greenpoint Dental at $2,100. The aging (60 to 90+ days) is QuickBooks’ own aging field. The reason each invoice ties back to a real job is the Jobber link: every one of those three balances maps to a completed Jobber job, so you can see the work was done before you make the call. That join is the source line “QuickBooks invoices · Jobber jobs,” and it is checkable in about two minutes.

The revenue drop comes from Jobber completed jobs, split by service line. The $22k monthly average, the roughly $13k HVAC and $9k plumbing inside it, and May’s fall to about $1k of plumbing are all counted from job records, not modeled. The Tina Alvarez detail is the same data read one more way: her jobs on the schedule went to near zero for the month. The digest states the correlation and stops there.

The marketing line joins two sources that never talk to each other on their own: spend from Google Ads and Meta, booked jobs from Jobber. “More booked work per dollar” is spend divided into booked jobs, per channel, for the quarter. That is the whole calculation, and it is the reason the item can exist at all.

Trace any figure and you land on a row you can open yourself. That is the point of the source line. You are never asked to trust the number on Guidepost’s word.

Pass two: why they are in this order

The order is not chronological and it is not by dollar size alone. It is by what needs you soonest.

Overdue money sits first because it is money you have already earned and are about to lose, and because there is a clear next step. Ticket-size drift, close-rate wobble, a slow week: those are real, but they are slower bleeds and often noise. $11,700 aging past 60 days is neither slow nor noise. So it leads.

The plumbing revenue drop ranks second, above the two marketing and quality items, because it is the largest unexplained swing on the board and it now has an explanation worth acting on. It jumped the marketing line specifically because it went from a question to a cause: once the drop traced cleanly to one person being out, it stopped being “look into revenue” and became “here is what happened.”

Then the two “Worth a look” items, because they are early signals, not decisions. And here is the part most tools get wrong: the footer. “Nothing else needs you this week” is a feature, not filler. A briefing that flags four things and then tells you the other twenty metrics are fine is doing the harder job. Silence, stated plainly, is what lets you close the tab and go run the shop. A dashboard cannot tell you that nothing is wrong. It just shows you everything and leaves the ranking to you at 9pm.

Pass three: the one that tells you what to do, and the ones that don’t

This is the line I care about most, so I built it into the product as a rule, not a preference.

Look at the two “Worth acting on” items. Both earned a next step, because the data supports one. Overdue invoices: send reminders, then call at 60 days. That is the canonical data-backed move, and the digest says it outright. The plumbing gap: the cause is known, so the action (“here are the jobs that didn’t happen”) is concrete.

Now look at the marketing item. It has a real signal: Meta’s return per dollar fell while Google’s held. And it deliberately refuses to tell you to move the budget. The exact wording is “we won’t pretend to know which way to move the budget,” because last-click attribution on this data is partly guesswork, and shifting spend on one quarter of noisy numbers is how owners talk themselves into a mistake. The honest move is to review the Meta campaigns, not to reallocate on our say-so.

The callback item does the same. Three redos from your newest hire is worth a look, and the digest says the likely reason out loud: new-hire ramp, not a pattern yet. It does not tell you to write anyone up. A cluster this small is a question, not a verdict.

That split (act where the data earns it, refuse where acting would mean guessing) runs through every digest. An item that prescribes and an item that only points look similar on the page. They are not the same, and the difference is the whole reason to trust the ones that do prescribe.

What a quiet week looks like

Northside had a busy week, so the digest has four items. Most weeks are quieter. When nothing has moved enough to matter, the digest says so and shows you the baseline: revenue holding, receivables flat, close rate steady, no callback cluster. It does not manufacture a fifth item to justify the send. A short digest on a calm week is the product working, not failing. The weeks it has little to say are the weeks you most want proof it is still reading.

If you want the discipline behind reading these on a fixed rhythm, the 15-minute weekly review covers the habit, and the six numbers worth watching covers what to keep an eye on between digests.

See it yourself

You just read the annotated version. The plain digest lives at /sample-digest/, unannotated, the way it actually lands. Read that one cold and see whether it is the report you have been trying to build by hand on Sunday nights.

The version that matters is the one built on your numbers, not Northside’s. That starts with a 20-minute walkthrough: I connect your Jobber or Housecall Pro and QuickBooks with you, and we read your first digest together on the call. Bring the number that has been bugging you. We will trace it to the row it came from.