Four Months to Trust the Numbers Again.

The Setup

The same real estate coaching company, six months after we started. By this point we'd fixed duplicates, fixed the pipeline logic, and fixed the status chaos. The CEO wanted the full executive dashboard — not pieces, not partial views, the whole revenue engine in one place.

The Problem

He'd tried three times before us.

The first was a freelance analyst who built a Looker Studio dashboard off manual exports. It was accurate the day it shipped. Two weeks later, the data stopped updating because nobody was running the exports. Four weeks later, the dashboard showed numbers that were three weeks old and a month-over-month comparison that was a year off.

The second was an in-house hire — a data person the COO brought on. She was good, but she owned the entire stack alone. When she left eight months later, three Tableau dashboards went dark within a quarter because nobody knew where the queries lived or how to update them.

The third was a BI consultancy. They delivered a beautiful dashboard, a 40-page documentation package, and a handover meeting. Six months later, the dashboard was technically still running but the ops team had stopped looking at it — because the company had launched two new offers and the dashboard didn't know about them. It was showing accurate answers to questions that no longer mattered.

What We Found

All three previous attempts had failed on the same thing: ownership. Each had delivered a snapshot of truth at a moment in time and then handed it over to someone who didn't have the time, skills, or authority to maintain it. The dashboards degraded not because they were built wrong, but because nothing on the business side was staying still and nothing on the data side was keeping up.

The CEO didn't need another deliverable. He needed a team that was on-call for his data the same way his accountant was on-call for his books.

What We Built

Five dashboards, over four months, built on the source-to-cash pipeline we'd architected in the earlier engagements.

Setter Performance — bookings, show rates, no-show recovery, per-source.

Closer Performance — close rate, revenue per appointment, offer mix, time-on-call, commission-to-date.

Executive Overview — the CEO's morning view. One scroll. Lead volume, cash collected, CAC by channel, gross margin after refunds.

Marketing Attribution — true blended ROAS by channel, net of refunds and chargebacks. This was the one that moved the most money.

Finance — cash-collected reconciliation, refund tracking, accrual view, commission liability.

Every dashboard is live. None of them require exports. When the business launches a new offer, we update the model before the dashboards break, because we're watching.

The After

Revenue grew 32% period-over-period in the two quarters following full rollout. Average order value moved from $413 to $3,510 as the team doubled down on the channels the attribution dashboard showed were truly profitable. Blended ROAS on $23,845/month in tracked spend settled at 1.33x — a number the CEO could finally make capital-allocation decisions against.

The difference this time wasn't the dashboards. It was the retainer. When the client added a new payment processor, we built the integration in a week. When they restructured commission tiers, we updated the rules before the next payroll cycle. When the marketing team wanted a new cohort view, we shipped it in three days.

Ongoing client time: about three hours a month.

Closing

This client has been with us for over three years. The dashboards from the original four-month build are still the dashboards they open every morning — because we've kept them current. The difference between the fourth attempt and the three before it wasn't the code. It was that we never left.