Productivity Input-Output and Price Test
Verify technology gains through value ratios and falling production costs
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 98%
The test begins by defining productivity strictly as a change in the relationship between valuable output and the labor or capital required to produce it. A process qualifies if the organization creates the same value with fewer inputs or more monetizable value with the same inputs. Faster workflows, prettier charts, or greater convenience do not qualify by themselves when staffing, revenue, and production costs remain unchanged. A second diagnostic examines prices: when technology genuinely removes input costs, competition should eventually transmit some savings through lower prices, higher margins, or both. The method therefore evaluates automation through unit economics rather than novelty. It also recognizes that productivity gains often arrive incrementally because adoption, infrastructure, regulation, and organizational redesign take years to mature.
Origin
Extracted from Masters in Business during a debate between Barry Ritholtz and Philipp Carlsson-Szlezak about Uber, AI, and service-sector productivity.
Core principles
- 01Convenience is not automatically productivity
- 02Productivity changes the ratio of inputs to valuable outputs
- 03More output counts only when it creates measurable value
- 04Technology-driven cost removal should create price pressure
- 05Adoption and infrastructure determine when gains appear
How to run it
- 1
Define valuable output
Specify the product or service being produced and how its value is measured through revenue, volume, quality-adjusted demand, or another economic outcome.
Pro tip Choose a unit that can be compared consistently before and after adoption.
Watch out Activity counts such as reports generated may not represent value if no one benefits or pays for them.
- 2
Count the inputs
Measure the labor, capital, software, energy, and time required to produce the output. Include the cost of the enabling technology itself.
Pro tip Use fully loaded costs rather than focusing only on hours saved.
Watch out Automation can shift costs into software or infrastructure without reducing total inputs.
- 3
Calculate the ratio change
Determine whether the same value is produced with fewer inputs or more value with the same inputs. Compare the result with a credible baseline.
Pro tip Track both output per worker and economic value per unit of total cost.
Watch out A better user experience alone does not prove that production became more efficient.
- 4
Look for economic transmission
Check whether the improvement appears in lower prices, higher margins, increased revenue, or reduced staffing needs. These effects validate that input savings or output gains are economically real.
Pro tip Allow for temporary margins before competition passes savings to customers.
Watch out Stable prices may reflect market power, so use margins and total costs as additional evidence.
- 5
Reassess after adoption matures
Repeat the measurement as workflows, infrastructure, and labor allocation adapt to the technology. Distinguish current results from credible long-term potential.
Pro tip Set staged reviews over several years for general-purpose technologies such as AI.
Watch out Do not declare an abrupt economy-wide transformation from early demonstrations.
In the wild
A research firm adopts AI tools that make charts and editing faster. It counts total software and labor costs, then checks whether the same team produces more reports that generate additional revenue or whether fewer staff hours produce the existing output. Prettier reports without more revenue or lower cost are logged as a quality benefit, not productivity growth.
→ Management can distinguish an enjoyable tool from an economically productive investment.
A ride-hailing app improves search, payment, and availability, but each ride still requires a vehicle and driver while customers often pay more than for a taxi. The service creates convenience and choice, yet the core production inputs remain. Driverless operation could later change the ratio by removing labor input.
→ Current convenience is separated from future transportation productivity.
Common mistakes
Equating speed with productivity
Completing a task faster matters economically only if saved resources are removed or redirected into additional valuable output.
Ignoring technology costs
Software, infrastructure, training, and oversight are production inputs. Omitting them exaggerates the apparent gain.
Counting unmonetized output
Producing more material is not productivity growth if it creates no additional value. Connect increased quantity or quality to revenue, cost, or another defined outcome.
Is it for you?
Best for
It is best for leaders evaluating automation, AI projects, software investments, and claims of technology-driven productivity.
Not ideal for
It is not ideal when outputs have important social or qualitative value that cannot yet be measured economically.
From the transcript
“It's less inputs per output.”
“When technology is truly productivity enhancing, you see that in falling prices, technology is deflationary.”
“It's got to be a change in the ratio of inputs to outputs on either side.”
From the episode
Economic Risks Versus Media Alarmism with Philipp Carlsson-Szlezak