// METRIC // SKU_COVERAGE

SKU Coverage: How we measure humanoid robots against the human hand

SKU coverage is the percentage of a warehouse's total SKU universe that a humanoid robot can reliably pick and place with its end effector. The human hand operates at 100% by definition. In our live production today, humanoid robots operate at approximately 5%.

Paul Baker, CFO and co-ownerMay 20268 min read
Talk to Our Operations Team
OPERATING_SCALE
Kits / year
30M+ASSEMBLED
Items / kit
10–15AVERAGE
Pick-place ops / year
~500MPRODUCT OF
SKU types / year
100K+FACILITY
Robot SKU coverage
~5%MAY 2026
// DEFINITION

What SKU coverage measures

SKU coverage
The percentage of a warehouse's total SKU universe that a humanoid robot or cobot can reliably pick and place with its end effector.
Human baseline
100%. A trained operator can pick essentially any item in the warehouse — band-aids, hammers, lip balm, body wash, fishing lures, surgical kit components — without retraining.
Today's humanoid robots
Approximately 5%. Across a facility universe of 100,000+ SKU types per year, the robots in our live production can reliably pick about 5 of every 100.
// OPERATING_SCALE

Why SKU coverage matters at scale

Productiv assembles more than 30 million kits a year. The average kit contains 10 to 15 items. That works out to roughly 500 million pick-and-place operations annually — across a facility universe of more than 100,000 SKU types.

The product universe is wide

Our kitting lines run through:

  • Band-aids and first-aid kit components
  • Hammers and hand tools
  • Lip balm tubes and cosmetics
  • Body wash and personal care bottles
  • Fishing lures and tackle
  • Surgical kit components

Every shape, weight, packaging type, and surface friction variation in that list is a separate validation problem for a robot.

Why coverage is the constraint

Pick accuracy measures whether a robot picks correctly when it can pick at all. SKU coverage measures how many of those 100,000+ items the robot can reliably attempt in the first place.

A robot at 99% pick accuracy on 5% of SKUs is a specialty device. A robot at 99% pick accuracy on 25% of SKUs is line coverage.

If your SKU coverage is too low, we can't keep the robot busy even for a shift.

// THRESHOLDS

What the numbers mean

Five thresholds describe the SKU-coverage curve from where we are today to full parity with the human hand. The economics of deploying a humanoid robot change fundamentally between each step.

100%

Trained human worker

The human hand achieves SKU coverage of 100% by definition. Any item in the warehouse, a person can pick.

BASELINE
~5%

Humanoid robots in our live production today

5 of every 100 SKU types in our facility have been validated against an end effector. Today's operating reality (May 2026).

WHERE WE ARE
25%

Robots become meaningful line coverage

At 25% coverage robotics stops being a specialty insert and starts taking the repeatable middle of high-mix kitting work.

NEXT THRESHOLD
40%

Deployment economics shift

At 40% coverage the cost-per-unit math on robot-led lines starts crossing fully human lines at meaningful volume.

MEANINGFUL
100%

Robots match the human hand

Full parity. A long way off — and not the realistic threshold for deployment decisions in the next several years.

LONG WAY OFF
// PARTNERS

What raises a robot's SKU coverage

Three levers move the number: the breadth of validated end effectors, partner engineering iteration on real production data, and process design that sequences SKUs into the robot's validated subset. Below is the deployment status of every robotics partner we currently run in live production — described factually, not as endorsement.

Tutor Intelligence's Cassie palletizer

~3 months in live production

Tutor Intelligence's Cassie palletizer has been running in our Dallas facility for approximately three months as of May 2026. Doing real work on real product, not a controlled pilot.

Avatar Robotics

Wheel-based, teleoperated → autonomy roadmap

Two Avatar humanoid units have been in live production since November 2025. Wheel-based locomotion. Currently teleoperated with a roadmap to increasing autonomy.

Blue Sky Robotics

13 cobots in live production since August 2025

13 Blue Sky Robotics cobots have been running on our kitting lines since August 2025. Our first robotics deployment partner and the foundation of our hybrid line model.

A note on form factor: warehouse floors are flat, hard, and continuous. Wheels are faster, more energy-efficient, and more reliable than bipedal locomotion for our workflows. We can't think of any workflows in our facility that require legs to complete.

Lighting, packaging variability, and end-effector breadth drive coverage. Walking around the warehouse does not.

// FIRST_DEFINED_PUBLICLY

Coined in Forbes, May 5, 2026

Paul Baker introduced the SKU coverage metric in conversation with Forbes contributor John Koetsier. The framing on this page is the operator-grade definition behind that quote.

Read the Forbes article
// FAQ

Frequently asked questions

What is SKU coverage in warehouse robotics?

+

SKU coverage is the percentage of a facility's total SKU universe that a robot can reliably pick and place with its end effector. The human hand achieves 100% by definition. In live humanoid robot production at Productiv today, SKU coverage sits at approximately 5% — meaning 5 of every 100 product types in the warehouse can be picked by the robot.

Who coined the term SKU coverage?

+

Paul Baker, CFO and co-owner of Productiv, introduced the term in a conversation with John Koetsier of Forbes published May 5, 2026. The metric reflects how Productiv evaluates humanoid robotics partners against operational reality rather than vendor benchmarks.

What's the difference between SKU coverage and pick accuracy?

+

Pick accuracy measures whether a robot completes a single pick correctly when it can pick the item at all. SKU coverage measures how many different items the robot can reliably attempt in the first place. A robot can have 99% pick accuracy on a narrow product set and still have only 5% SKU coverage.

What is Productiv's current SKU coverage rate?

+

As of May 2026, our humanoid robots in live production sit at approximately 5% SKU coverage across a facility universe of more than 100,000 SKU types per year. This is the same operational number we've previously published as 'SKU validation rate' — the public-facing term is SKU coverage.

What SKU coverage do humanoid robots need to replace warehouse workers?

+

Realistically, robots need 25–40% SKU coverage to take meaningful line coverage on high-mix kitting work, and roughly 100% to fully replace human pickers. At 5% coverage today, robots augment humans on the repeatable middle of the work; humans handle the long tail. That mix is unlikely to invert in the near term.

How do you measure SKU coverage?

+

We test each end effector — gripper, claw, suction cup variants — against the actual product on a live conveyor. A SKU is counted as covered when the robot can pick it reliably across normal production conditions: lighting, orientation variability, packaging deformation. The number is the share of validated SKUs over the facility's full SKU universe.

Why is SKU coverage harder than it sounds?

+

Each SKU has a unique shape, weight distribution, packaging format, and surface friction. Validating one SKU means testing multiple end effectors against the actual product, sometimes under varying lighting and orientation conditions. A facility with more than 100,000 SKU types has a long validation backlog, and the human hand remains extraordinarily capable by comparison.

Why does Productiv prefer wheels over legs for humanoid robots?

+

Warehouse floors are flat, hard, and continuous. Wheels are faster, more energy-efficient, and more reliable than bipedal locomotion for the workflows we run — kitting, pick-and-pack, labeling, display building. We can't think of any workflows in our facility that require legs to complete.

// SEE_IT_IN_PRODUCTION

Want to see what 5% SKU coverage looks like?

We run cobots and humanoid robots on live kitting lines in Dallas. If you're evaluating warehouse robotics, we'll walk you through the SKU validation work in person.

Talk to Our Operations Team

Direct access to Paul Baker and Doug Legan. No gatekeepers.