You've read the five accountabilities and the way the work runs. Now you need the part a founder can actually use: a single page that tells you, at a glance, whether your engineering spend is producing what you paid for.
This isn't a wall of charts, and it isn't a tool only an engineer can read. It's ten numbers, grouped the way the book is grouped, two per accountability. Read top to bottom and you've taken the temperature of the whole operation in about ninety seconds.
The point of the scorecard isn't to turn you into an engineering manager. It's the opposite. You shouldn't have to sit in a standup to know whether things are on track. The numbers do that for you, and they do it in business language: how fast, how predictable, how often it breaks, how much of the work mattered.
A note before the metrics. Four of these come straight from DORA, the industry-standard framework for measuring software delivery (deployment frequency, lead time for changes, change-failure rate, and time to restore). The other six are the operational numbers a non-technical owner needs to see the rest of the picture. Where a benchmark exists I'll name it. Where one doesn't, I'll tell you what "good" looks like in plain terms rather than invent a figure.
This is Chapter 3, made measurable. Two numbers tell you whether the right things are shipping on a cadence you can plan around.
1. Lead time for changes. The clock from "we decided to do this" to "it's live for customers." Short lead time means a request becomes working software in days, not quarters. Long lead time is the stop-start pattern you felt in Chapter 3, where a decision sits for weeks before anyone touches it. This is one of the four DORA metrics, so you can compare yourself against a recognised standard rather than your own gut.
2. Delivery predictability. Of the work you committed to this cycle, how much actually shipped? Plan ten tasks, deliver eight, and your predictability is 80%. The absolute speed matters less than whether the number you're told is the number you get. A team that reliably delivers eight of ten is more useful to a business than one that promises twelve and lands five. Predictability is what lets you make commitments to your own customers without crossing your fingers.
Chapter 4, on the dashboard. Throughput you can count, and the key-person risk the title was supposed to remove.
3. Cycle time per task. Lead time measures the whole journey; cycle time measures the active part, from the moment someone starts a task to the moment it's done. It's your read on flow. When cycle time creeps up, work is queuing somewhere, or tasks are too big to move cleanly. Small, scoped tasks keep this number low and stable, which is exactly why the work is shaped the way Chapter 9 describes.
4. Bus-factor / key-person concentration. How many people could keep a given system running if one person were out for a fortnight? A bus factor of one is the quietest risk in your company. Everything works right up until the day it doesn't, and then nobody else can fix it. You measure this by asking, per critical area, how many engineers have shipped to it recently. One name everywhere is a flashing light, however fast that person is.
Chapter 5. Nothing stays up forever. These two tell you whether failure is rare and recovery is fast.
5. Change-failure rate. Of everything you ship, what fraction breaks something and needs a fix, a rollback, or a patch? A DORA metric, and a blunt honesty check. A team shipping fast with a high failure rate isn't fast, it's accruing a bill. A low rate means the speed is real and customers aren't paying for it in outages.
6. Time to restore (MTTR). When something does break, how long until service is back? The fourth DORA metric, and the one your customers feel most directly. A short restore time means a bad deploy is a blip; a long one means a bad deploy is a bad week. Notice that reliability isn't about never failing. It's about failing rarely and recovering quickly, which is something you can actually measure and hold someone to.
Download the full PDF for free?