ROI Case File No.551: 'The Convenience One Person Found Never Became the Organization's Wisdom'
The Convenience One Person Found Never Became the Organization's Wisdom
Chapter 1: The Convenience Stopped at a Few Desks
"We want to use generative AI across the whole company. But we can't figure out how to spread it."
Soichi Takatori, CEO of TechBridge, spread out his documents as he spoke. His company had layered on paperless initiatives and DX promotion over the years. "We decided generative AI is next, but there are a lot of internal problems. The gap in IT literacy is wide, and how to use the systems has become person-dependent. Some employees use it well on their own, but it doesn't spread to the whole department."
"So there are people who can use it on their own," Claude asked.
"There are," Takatori answered. "But it stops with them. It doesn't reach the desk next to them. The MANA Studio we've adopted has usage caps, too, so it never accumulates as organizational knowledge. People feel that it's useful, yet it isn't becoming the company's strength."
"Are the usages individuals discover being shared?" I asked, to confirm.
"They aren't," Takatori said. "'This is how I use it' exists in scattered pieces. Someone tries it from zero, then someone else tries it from zero again. The same wheel gets reinvented over and over."
"We have to turn individual convenience into organizational wisdom," I replied. "Let's break it down with ECRS."
Chapter 2: ECRS Asks—Discard, Bundle, Rearrange, Simplify
"This case calls for ECRS."
Claude wrote "Eliminate, Combine, Rearrange, Simplify" on the whiteboard.
"ECRS—Eliminate, Combine, Rearrange, Simplify—is a framework for improving work from four angles," I explained. "The key is the order. First discard what isn't needed, then bundle what can be combined, then rearrange, and only at the end simplify. Keep the order, and you avoid the spinning of wheels where you make things efficient while leaving the waste in place. It's the tool that turns AI use that stalls at the individual into company-wide wisdom."
"Let's measure the current cost first," Gemini said, opening ROI Polygraph, and entered the data Takatori had provided.
"The monthly cost is in," Gemini read out. "Inefficiency hours from uneven generative-AI use caused by literacy gaps and person-dependence average 180 hours a month; at ¥4,200 an hour, that's ¥756,000 a month. Duplicated work and reinvented wheels from use staying at the individual level average ¥480,000 a month. Lost opportunity from being unable to accumulate knowledge under MANA Studio's usage caps averages ¥420,000 a month. Stalled non-users caused by the complexity of the onboarding process average ¥380,000 a month. The expected value of person-dependence risk from un-accumulated organizational knowledge averages ¥330,000 a month. The total is ¥2,366,000 a month—roughly ¥28.39 million a year."
Takatori stared at the figures. "I thought this couldn't be measured by a sense of usefulness alone. Once you include the cost of not spreading and not accumulating, I never imagined it would be this much."
"Then let's design it with ECRS," I continued.
[Eliminate—Discard the person-dependent ways of using it]
"First, we eliminate the scattered, person-dependent ways of using it," Claude said. "We replace the self-taught procedures each person memorized with standard manuals and training. What we discard isn't the individual ingenuity—it's the state of not being shared."
[Combine—Bundle each department's use into one]
"Next, we combine the ways each department uses it," Gemini continued. "We form a cross-functional team of representatives from each department, bring the scattered uses together, and compile them into case examples. Because we bundle, we don't build the wheel twice."
[Rearrange—Recompose the cap constraint by priority]
"After bundling, we rearrange the resources," I continued. "MANA Studio's usage cap doesn't work if you spread it thinly across everyone. We revisit the account-issuance criteria and assign them first to high-impact projects. A constraint becomes a weapon once you recompose it."
[Simplify—Make an entrance anyone can use]
"Finally, we simplify the onboarding process," Claude continued. "We arrange an entrance that even low-literacy people won't get lost in. We make the user interface easy to use and erase the step of the first step. Because simplifying comes last in the order, you don't end up just tidying the appearance while leaving the waste in place."
[Estimating the payback]
"Let's run the numbers with ROI Proposal Generator," Gemini proposed.
- Initial cost: Standard manual development, company-wide training, cross-functional team launch, account-criteria redesign, and UI simplification customization—¥5,300,000 total
- Monthly cost: Operations and ongoing update fees combined—¥220,000 a month
- Monthly savings: Inefficiency eliminated by standardized use = ¥560,000 a month (assuming a 70% reduction); duplicated work reduced = ¥380,000 a month; effect of knowledge accumulation and priority reallocation = ¥360,000 a month; expanded use from simplified onboarding = ¥280,000 a month; ¥1,580,000 a month total
- Net monthly savings: ¥1,580,000 − ¥220,000 = ¥1,360,000 a month
- Payback period: ¥5,300,000 ÷ ¥1,360,000 = about 3.9 months
"Payback in just under four months," Gemini summarized. "What works is keeping the order—not adding convenient features right away, but discarding, bundling, and rearranging before simplifying. Because we eliminate person-dependence and combine the uses before rearranging the cap, the resources don't get spread thin. Because it follows the order, the investment doesn't whiff."
Takatori said, checking the figures, "I'd been thinking that handing out more convenient tools would spread it. The order was backwards. Unless you discard and bundle first, handing it out just scatters it."
"ECRS is the tool that turns individual convenience into organizational wisdom," I replied.
Chapter 3: A Rollout Plan That Discards and Bundles Before Handing Out
"Let me lay out the approach," I said, standing at the whiteboard.
"Month 1—inventory the current usage and design the standard manuals and training; the Eliminate phase. Month 2—form the cross-functional team and combine the use cases. Month 3—revisit the account-issuance criteria and allocate by priority; Rearrange. Months 4–5—UI simplification customization and company-wide deployment. Month 6—pilot operation and effect verification. Month 7 onward—settling in the accumulation of organizational knowledge and expanding the scope of use."
"Will it really spread to the whole department?" Takatori asked, to confirm.
"It will spread," Claude replied. "It stops at the individual because the way of using it isn't shared and the convenience is closed inside each person's hands. Eliminate removes the self-taught styles, Combine bundles the cases, Rearrange concentrates resources where they work, and Simplify builds an entrance anyone can step into. Because you erase the causes of not-spreading in order, it reaches the whole company through a mechanism, not a command."
Takatori said, taking notes, "Before adding tools, discard and bundle. I can see the order now."
Chapter 4: The Day Individual Convenience Became the Company's Strength
Nine months later, a report arrived from Takatori.
The person-dependent ways of using it lined up after the standard manuals and training were introduced. "The self-taught, scattered operations became something anyone could use with the same procedure. The ingenuity that never reached the next desk got onto a foundation for being shared," Takatori had written.
The walls between departments grew lower, too. The cross-functional team brought each department's cases together and spread them sideways. "'This is how I use it' became a company casebook. The waste of re-testing from zero went down," the report said.
The biggest change appeared in how knowledge accumulated. Uses that had vanished in individual hands began to accumulate in the organization. "The sense of usefulness had ended with each person. Once we used Rearrange to direct the cap to important projects and built a mechanism to keep the cases, it began to accumulate as company assets," Takatori had written.
The literacy gap closed, too. The simplified interface drew in employees who had never touched it. "People who'd avoided it because it looked hard began using it once the entrance got simple. The gap shrank on its own," the report said.
As a secondary effect, the way they pursued improvement changed. The order of discarding and bundling before simplifying became a phrase around the office. "Before adding a feature, discard the unneeded state. Think about whether it can be bundled. This order began to work on improvements beyond AI, too," Takatori had written.
At the end of his report, he wrote: "I thought introducing generative AI was about how many convenient features we could hand out. But the real problem was that the convenience one person found never became the organization's wisdom. The moment we used ECRS to discard, bundle, and rearrange, and simplified last, the convenience spread across the company. Before handing out, discarding and bundling came first."
The day a company where the convenience one person found had stopped at a few desks became a company that could turn it into the organization's wisdom, the generative-AI rollout had changed from a command to hand out tools into a design that spreads it through Eliminate, Combine, Rearrange, and Simplify, the report noted.
"Generative-AI rollout requests usually come in the form of 'how do we spread the convenient features.' But there's something to ask before spreading. In the organization right now, is there a state to discard and a use to bundle? What ECRS asks is the order of Eliminate, Combine, Rearrange, Simplify. Discard the self-taught, bundle the cases, recompose the resources, and simplify the entrance last. The day a company where the convenience one person found had stopped at a few desks turned it into the organization's wisdom, what changed wasn't the generative-AI tool but the very perspective that turns individual convenience into organizational wisdom."
Related Files
Tools Used
- ROI Polygraph — Visualizing the hours of uneven use, duplicated work, and the cost of being unable to accumulate knowledge
- ROI Proposal Generator — Payback simulation for a company-wide generative-AI rollout starting from Eliminate, Combine, Rearrange, and Simplify
