Case Study · Atlas Supply

Stock counts disappeared.

Built a real-time inventory pipeline that eliminated weekly manual stock counts entirely.

IndustryDistribution
Engagement8 weeks
StackPython · PostgreSQL · Metabase
Result−100% manual stocktakes

The brief

Atlas ran weekly manual stocktakes across 3 warehouses. A team of 4 spent every Friday afternoon counting. Discrepancies were found Monday, resolved Tuesday. The data was always 5 days stale.

What we built

A real-time inventory pipeline that reconciles movements (inbound, outbound, returns) against expected stock levels continuously. Discrepancies surface in Metabase within minutes, not days.

Results

  • Eliminated weekly manual stocktakes entirely
  • Real-time discrepancy alerts (minutes, not days)
  • 16 hours/week freed across the warehouse team
  • 99.2% accuracy sustained over 6 months
Book a call

Fifteen minutes to figure out if we're a fit.

Bring the messiest thing you're dealing with. If we can't help, we'll tell you — and usually recommend someone who can.

  • No sales deck, no follow-up sequence
  • Walk away with a written plan of attack
  • Free, zero-obligation, 15 minutes max
Reply within
24 hours · usually 2
01 · Scope02 · Time03 · Confirm

What do you need help with?

How would you like to work?
Sprint (4 weeks, fixed)
Retainer (monthly)
Custom / not sure yet
What's hurting?
Automate manual work
Unify data / reporting
Build an internal tool
Connect systems
Process audit
Something else