CodeNSM
[ CODENSM ]PRIVATE BETAby Think North

Know which code actually earns its keep.

CodeNSM is drop-in observability that treats every function like an employee — tracking how often it works, how reliably, and how much it contributes to your product’s North-Star Metric. So you can see where tech debt hurts, what’s underused, and which code creates value — without wiring up dashboards by hand.

// one import · zero config · your NSM, mapped to your code
~/your-app · codensm
# 1. wrap any function — that's it from codensm import observe @observe(department="checkout", value_weight=9) def apply_coupon(cart, code): ... # 2. CodeNSM watches it run, in prod › apply_coupon calls=12,481 p95=38ms err=0.2% › contribution to NSM ......... █████████░ ▲ high # 3. and flags what doesn't earn its keep › legacy_pdf_export calls=3/mo underused — candidate to retire › retry_webhook err=6.1% tech-debt: RCA logged
[ 001 ]
The blind spot

You ship code for outcomes. You measure it like plumbing.

?

Value is invisible

You know latency and error rates. You don’t know which functions move the metric the business actually cares about.

Tech debt hides

The painful debt isn’t the ugliest code — it’s the code that’s both fragile and load-bearing. That’s hard to see by hand.

Dead weight lingers

Features no one uses still cost you maintenance, surface area, and risk. Underuse rarely shows up on a dashboard.

[ 002 ]
How it works

Treat every function like an employee.

Each function gets tracked the way you’d track a team member or a machine on a line — utilisation, reliability, and contribution — then rolled up into departments, KPIs, and one North-Star Metric.

1

Add the library

One import. Decorate the functions you care about — or let middleware auto-instrument requests.

2

It watches in prod

Calls, latency, errors, and self-healing RCA on failures — captured with near-zero overhead.

3

Map to your NSM

Give functions a value-weight and a department. CodeNSM rolls utilisation into OKRs, KPIs, and a North-Star.

4

Act on it

See top contributors, underused dead weight, and the tech debt that actually hurts. Decide what to invest in or retire.

[ 003 ]
What you get

OEE for your codebase.

utilisation

Per-function OEE

Usage, reliability and latency for every tracked function — like overall equipment effectiveness, but for code.

alignment

NSM & OKR/KPI mapping

Attach code to the outcomes it serves. Watch contribution roll up into departments and a single North-Star Metric.

tech debt

Debt-that-hurts heatmap

Find the fragile-and-load-bearing code — high error rate × high contribution — and fix what matters first.

cleanup

Underuse detection

Surface features and paths almost nobody hits, so you can retire dead weight with confidence.

resilience

Self-healing RCA

On failure, CodeNSM runs multi-level root-cause analysis, attempts safe remediation, and logs an incident.

roadmap

Engineering ROI SOON

Pull in Git history to attribute value to developers, sprints and PRs — see whose work and which sprints yield ROI.

1 line
to instrument a function
0 cfg
works out of the box
3-level
root-cause analysis on errors
1 NSM
every function rolls up to it
[ 004 ]

Get early access to CodeNSM.

We’re onboarding teams gradually. Tell us a little about your stack and we’ll reach out.

No spam. We email you when your slot opens.