Back to paper: why analog planning is beating dashboards

Maya Okonjo runs Fieldnote, a farm-logistics company with 22 employees and a routing engine that processes something like 40,000 delivery legs a week. Her planning system is a single ruled notebook, one page per week, three lines at the top. She has not opened the company’s executive dashboard since February. When asked whether it still exists, she said she assumed so, and that someone was probably paying for it.

This is not a Luddite position. Okonjo’s company is built on telemetry — every truck reports, every route is scored. The distinction she draws is between data the business needs and data she personally needs, and she is increasingly convinced those are different data, arriving at different speeds, for different jobs.

The dashboard measures, the page decides

The complaint founders make about dashboards is rarely that the numbers are wrong. It is that there are 40 of them, they refresh hourly, and none of them has an opinion. A dashboard will tell you churn moved from 3.1% to 3.4%. It will not tell you whether that is the thing to spend Tuesday on, because it does not know what else was on the table.

The Sherwood Index, which tracks tooling spend across roughly 1,200 private companies, found the median firm under 50 staff was paying for 6.4 analytics and planning products in 2025, up from 2.9 four years earlier. It also found that the median number of those products opened in a given week by the founder was one. The other 5.4 are, in effect, a subscription to the feeling of being informed.

Okonjo’s three lines are the week’s decisions — not tasks, decisions. Something with an answer that costs money either way. Last week’s page read: whether to keep the Marchford depot, whether to raise the co-op tier, whether to let Devi go. The satisfaction of the notebook, she says, is that a page has a bottom, and a screen doesn’t.

What survives the switch

Nobody serious is running finance on paper. The founders doing this have drawn a boundary, and it lands in roughly the same place each time:

• Anything a customer or an auditor touches stays in software. This is not negotiable and nobody pretends otherwise.
• Anything that must be true at 3am — inventory, cash, uptime — stays in software, with alerts.
• Weekly and quarterly decisions move to paper, because they are made once, by a person, at a desk.
• The paper is disposable. Nobody is building an archive. The notebook is a thinking surface, not a system of record — the moment it becomes a system of record it grows a schema and dies.

That last point is where most attempts fail. A founder buys the notebook, likes it, and within a month has invented a numbering scheme, a colour code and a migration plan. Priya Nair at Loom&Co described going through this cycle twice before concluding that the correct move was to throw the pages away on Friday. “If it mattered, it’s in the software or it’s in my head,” she said. “If it’s in neither, it didn’t matter.”

There is a cheaper explanation for all of this than the philosophical one, and it may be the true one. Dashboards are built for people who need to explain a company to someone else — investors, boards, regional managers. Founders of bootstrapped companies of 20 people don’t have that audience. They have themselves, on a Sunday evening, trying to decide three things. Nobody has ever needed a filter and a date-range picker for that.

Sherwood’s figure for the median annual analytics spend at those firms was $18,400. A notebook is about eleven dollars, and the thing about eleven dollars is that you never once open it out of guilt.