How to select tasks and candidates for implementing Robo Claw in local governments.
Evaluate target tasks based on the impact on residents' rights and interests, whether AI will make administrative judgments, whether personal or sensitive information is handled, whether interpretation of laws or ordinances is involved, whether there is external transmission, and whether approvals can be retained. It is recommended to prioritize standardized tasks that span multiple departments and are mainly focused on reading or information organization.
When municipalities choose the tasks for which to introduce Robo Claw, please prioritize tasks that meet the following criteria: ① The impact on residents' rights and interests is small, or human verification can be assumed; ② AI does not make administrative decisions itself; ③ Management of personal and sensitive information is organized; ④ Does not involve legal or ordinance interpretation, or interpretation is done by humans; ⑤ There is no external transmission, or it can be made subject to approval; ⑥ Decision-making and approval can remain in existing processes; ⑦ Escalation to relevant departments or professionals is possible; ⑧ Stopping, correcting, or rolling back is possible. Tasks that meet these criteria will be easier to verify for effectiveness in the pilot phase.
Who This Is For
Who This Is For
This is intended for those considering which tasks to start with when introducing Robo Claw, such as personnel in the digital promotion department, information system managers, and resident service desk managers.
What You'll Decide
What You'll Decide in This Step
In this Discover STEP, we will identify candidate tasks for Robo Claw, prioritize them, and decide on the first task to undertake. Detailed design of requirements, permissions, and approvals will be handled in the next Refine STEP.
Fit Check
Suitable/Not Suitable Tasks
Suitable Tasks
- Tasks with high occurrence frequency and standardized procedures, such as first-level classification of in-house inquiries or searching rules and guidelines.
- Tasks in which staff can easily check and modify preliminary judgments, such as first-level classification of resident inquiries or suggesting FAQ candidates.
- Tasks that require cross-checking and integrating information from multiple departments.
- Tasks that do not involve writing resident information, or can have the approver's approval before updates.
- Tasks that do not involve the final judgment of administrative actions or benefit eligibility, limited to information organization or draft preparation.
Unsuitable tasks
- Tasks that occur very infrequently and for which automation benefits are difficult to anticipate
- Tasks that require immediate confirmation without approval for administrative actions, permits, license issuance, benefit eligibility, or reviews.
- Tasks that attempt to make decisions directly affecting residents' rights without human confirmation
- Tasks that include a large amount of sensitive personal information in input data and whose handling is unorganized
Industry Challenges
Challenges unique to local governments
Many local governments face the following challenges when selecting target tasks.
Too many candidate tasks to narrow down
There are many candidates, such as internal inquiries, regulation searches, resident inquiries, and meeting minutes organization, making it difficult to decide where to start.
Unable to formulate hypotheses of effectiveness
The actual workload and man-hours of each department are not visualized, making it difficult to estimate the introduction effect in advance.
Uncertainty about connectivity with resident information systems and core systems
Whether existing systems support API integration can only be determined by the information systems department.
Uncertainty in handling tasks related to administrative decisions
There is no criteria for deciding whether tasks related to benefits, examinations, or administrative sanctions can be candidates.
Method
Implementation Steps
1. Identify candidate operations
Inventory tasks that occur across departments, such as first classification of internal inquiries, regulation searches, resident inquiry classification, and meeting minutes summarization.
2. Check workload and frequency
Confirm the occurrence frequency of each operation (constant, daily, weekly, etc.) and the labor required to handle it.
3. Confirm the impact on residents' rights and interests
Confirm whether the target tasks involve decisions that affect residents' rights and interests, such as administrative sanctions, eligibility for benefits, or examinations.
4. Confirm the handling of personal information and sensitive information
Confirm whether the target business handles personal or sensitive information such as names, addresses, and tax, welfare, or health information.
5. Confirm the presence of legal or ordinance interpretations
Check whether the target business involves interpretation of laws or ordinances, and if it does, whether it can be designed so that humans make the final judgment.
6. Confirm whether external transmission or connection to resident information systems is possible
Verify with the information systems department the location of the data required by the target business, whether external transmission occurs, and whether connection to resident information or core systems is possible.
7. Confirm approval and escalation routes
Check whether the outputs of the target business can be integrated into the existing approval process and whether there are escalation routes to the relevant departments or specialists.
8. Determine priorities
List the evaluation results and determine the first department and business to start with.
Evaluation Table
Candidate Task Evaluation Sheet
The following is an example of an evaluation table. Please replace it with your department's candidate business. The closer a business is to ◎, the higher its priority for Discover.
| Candidate tasks | Frequency | Formality | Impact on residents’ rights and interests | Handling of personal information | Post |
|---|---|---|---|---|---|
| Primary classification of internal inquiries | Daily | Expensive | None | Low | None |
| Search of regulations, guidelines, and manuals | As needed | Expensive | None | None | None |
| Primary classification of resident inquiries | Daily | Expensive | Low (assuming human verification) | Moderate | None |
| Support for checking missing items in application documents | As needed | Moderate | High (acceptance determined by staff) | Expensive | None |
Data & Systems
Data and Systems Used
In the Discover STEP, the status of the following data and systems is checked to evaluate candidate tasks.
Human-in-the-loop
Where Human Approval Is Required
At the Discover STEP stage, implementation is not performed, but during the evaluation of candidate tasks, the following judgments are organized based on the assumption that they will always be made by humans.
- Final decision on whether the target task can be selected as an automation candidate
- Judgment on whether to include tasks related to administrative dispositions, eligibility for benefits, and examinations as candidates
- Judgment on whether to include tasks handling personal information and sensitive information as candidates
Measurement
KPI Candidates
In the Discover STEP, the following are recorded as hypotheses for each candidate task and used as verification material in subsequent steps.
Current work effort
Estimation of the human workload currently required for the candidate task
Effect hypothesis
Hypothesis of the workload and time savings expected if automated
Implementation priority score
Priority calculated based on frequency, standardization, impact on residents' rights, etc.
Pitfalls
Common Pitfalls
Select only tasks where the effects are easily visible
If you select only the prominent tasks without considering their frequency or the burden of handling exceptions, it becomes difficult to verify the effectiveness of the pilot.
Postpone confirmation of connection to resident information systems
If you decide on candidate tasks first and then check connectivity, rework may occur in later steps.
Underestimate the impact on residents' rights
If you choose tasks without evaluating the magnitude of their impact, you may need to redesign the approval flow in later processes.
FAQ
Frequently Asked Questions
How many tasks are appropriate to start with first?
In most cases, it is recommended to start with one department and one task. Advancing multiple tasks simultaneously complicates authority design and Pilot evaluation.
Should tasks involving residents be excluded from candidates?
It is not necessarily required to exclude them. By automating information organization and draft creation while having the final decision made by the responsible department or approver, they can be treated as candidates.
What should I do if I do not know whether connection to resident information systems is possible?
Coordinate with the information systems department and confirm whether the target system's API or data linkage is possible. Individual consultations are also available.
Who should evaluate the candidate tasks?
It is recommended that the on-site responsible department and the information systems department conduct this jointly. When handling personal or sensitive information, it is also recommended to include the person in charge of personal information protection in the evaluation.
Continue
Next Step
Once the target operations are decided, the next step is to concretize the requirements, authorities, and approval design for those operations.
Shall we organize the selection of target operations together?
Based on the current workload, frequency, and the configuration of resident information systems, you can formally consult us via LP regarding the application candidates and priorities for Robo Claw.