Why being data-driven is hard

Everyone says data-driven. Almost nobody is.

Everyone says data-driven. Almost nobody is. The honest part is why. Three reasons, and none of them is about software.

1. You have to say "I might be wrong" and still be able to move. Following the data means letting it contradict you. For most of us, identity is glued to position: the design we defended, the number we signed. Saying I was wrong, I need more information takes mental resources that a delivery week doesn't leave you. So we defend the position instead of checking it.

2. Information is expensive, and nobody pays you to look for it. We are paid to deliver. That is what the fee covers. Exploring, verifying, doing one more round of research: the work that would actually change the decision is rarely encouraged and almost never budgeted. So decisions get made with whatever was already on the desk.

3. Data-driven results compound, and compounding is slow. The payoff arrives late, after weeks of small steps that look like nothing, and many of us cannot hold out that long. The trader Richard Dennis said you could publish his rules in the newspaper and no one would follow them. He proved it: he taught his complete system to novices from a newspaper ad, and they made him a fortune. The system was public. The edge was staying with it on the weeks when nothing seemed to come of it.

Jobsite Automation Fills Labor Crisis — Construction firms are turning to jobsite automation not just to fill labor gaps, but to redefine how work gets done—from preserving expertise to rethinking precision and teamwork.

Closing Constructions Widening Workforce Experience Gap With AI (Construction Executive) GenZ estimators need AI tools that build judgment, not replace it; veteran knowledge transfer is the real bottleneck (per Construction Executive).

Adoption of jobsite robotics doubles in 2026: BuiltWorlds report (bdcnetwork) Accuracy gains, not labor cost, now drive adoption among contractors—reframe firm ROI models accordingly, per BuiltWorlds 2026 benchmarking.

Embracing Automation in Construction (ENR) Solar contractors prove full autonomy fails; pairing robots (panel placement, 2mm precision per ENR) with skilled humans maximizes both.

Upcoming events

14,600 open projects on the map 

That's this week's brief. If something here helped or something is missing, just hit reply. I read everything.

See you next week, Shicong

Reply

or to participate.