Introduction
Learn where to start and what order to follow when using the evaluation version.
- Recommended routes for different evaluation goals (60 minutes / 90 minutes / your own files / DC integration)
- What to do first and what to avoid
- A checklist for organizing evaluation results
After reading this page, you can begin your evaluation without uncertainty.
Overview
This page explains what first-time users of the Agentiqs evaluation version should check and the order in which to try its capabilities.
Agentiqs is more than a tool for chatting with AI.
By combining AI agents, knowledge, workflows, file operations, and external tool integration, you can incorporate AI into work such as design, review, test design, meeting-minutes creation, and inquiry handling.
Immediately after obtaining the evaluation version, however, you may encounter the following questions.
- What should I try first?
- How do I create an AI agent or workflow?
- Which sample data should I use?
- How do I adapt it to my own work?
- How is it different from an ordinary AI chat?
- How can I quickly evaluate the value of integration with Next Design or Lightning Review?
This guide explains how to evaluate the value of Agentiqs in a short time using the evaluation version.
What you will learn on this page
This page explains the following.
- What to check first in the evaluation version
- How to proceed with the evaluation
- Points to check during evaluation
- Common evaluation pitfalls
- Pages to read next
Agentiqs works.For details, see Sample Data.
This evaluation guide explains how to deepen your understanding of Agentiqs using the sample data.
What to check in the evaluation version
Do not merely check whether AI can provide an answer. Confirm how Agentiqs can help with your work.
We especially recommend checking the following four points.
| Point to check | Details |
|---|---|
| Value of AI agents | Can you use AI tailored to a business objective? |
| Value of knowledge | Can it answer with reference to business documents such as manuals and specifications? |
| Value of workflows | Can it combine multiple processes and run them as standardized work? |
| Value of tool integration | Can it integrate with business tools such as Word, Excel, Next Design, and Lightning Review? |
When evaluating Agentiqs, it is important to look beyond the intelligence of AI by itself and determine whether AI can be incorporated into a business process.
What to prepare before evaluation
Prepare the following before beginning the evaluation.
| Preparation item | Details |
|---|---|
| Agentiqs | Follow the Setup Guide to install Agentiqs and make sure it can start. |
| AI provider settings | Follow AI Provider Settings to enable chats with AI. If your administrator has provided a configured environment, users do not need to configure an AI provider themselves. |
| Sample data | Download the sample data and place it in the specified folder. |
| Evaluation theme | Decide what you want to check. Examples: Can it answer with reference to an internal manual? Can it create minutes from meeting notes? Can it extract review findings from a design document? If you cannot decide on an evaluation theme yet, begin with the 60-Minute Quick Evaluation Route instead of forcing a decision. |
| Your own files | If possible, prepare one to three business documents you want to try. |
| External tools | If you want to try Next Design or Lightning Review, make sure you can start the target tool. |
Recommended evaluation environment
For your first evaluation, we recommend the following environment.
| Item | Recommendation |
|---|---|
| Input data | Sample data, or copies of your own files for verification |
| Number of files | 1-3 files |
| Document volume | Approximately 10-30 pages |
| Target work | Focus on one theme |
| Output destination | Verification folder, verification Excel file, or verification review |
| External tools | A sample or copied project, rather than production |
| Evaluator | Someone who understands the work and can assess whether the output is valid |
Always test with verification data before outputting directly to a production file or production project.
What to avoid at first
To ensure a smooth evaluation, avoid the following during your first attempt.
| What to avoid | Reason |
|---|---|
| Registering a large number of documents at once | It becomes difficult to isolate the cause if the result is poor. |
| Trying multiple business tasks at the same time | Evaluation perspectives become scattered, making the effect difficult to assess. |
| Outputting directly to a production file | Unexpected results may remain in the file. |
| Building a workflow from scratch immediately | It is faster to first understand the process using a sample. |
| Evaluating materials for which the correct answer is unknown | You cannot determine whether the AI output is valid. |
| Starting immediately with external tool integration | The many prerequisites make initial evaluations more prone to problems. |
First, confirm normal operation with the sample data.
Then replace it with a small theme close to your own work to make the evaluation easier.
Evaluation route details
Choose a route based on your evaluation objective.
| Evaluation objective | Recommended route | Overview | Link to details |
|---|---|---|---|
| Get an initial feel for the product | 60-Minute Quick Evaluation Route | Experience the two core values of knowledge use and deliverable generation in the shortest possible time. | ■ |
| Evaluate the overall value of Agentiqs | 90-Minute Standard Evaluation Route | Follow a timetable to experience knowledge use, deliverable generation, review support, and workflow automation. | ■ |
| Determine whether it can be used for your work | Evaluation Route Using Your Own Business Files | First verify operation with sample data, and then replace it with your own files. Replace only the input file, and adjust the question and perspectives while checking the output. | ■ |
| Integrate with Next Design or Lightning Review | DENSO CREATE Product Integration Evaluation Route | Experience retrieving design models, generating test cases, reviewing models, and automatically registering review findings. | ■ |
| Consider team adoption | Half-Day Evaluation Route | To complete a half-day evaluation with results you can explain internally, we recommend completing both of the following courses. ・90-Minute Standard Evaluation Route ・Evaluation Route Using Your Own Business Files | - |
60-Minute Quick Evaluation Route
If your time is limited, use the first part of the 90-minute course as a 60-minute short course to experience the basic value of Agentiqs quickly.
Timetable
| Time | Activity | Purpose |
|---|---|---|
| 0-10 minutes | Prepare for the evaluation and confirm the goal | Decide what to focus on |
| 10-30 minutes | Experience knowledge use | Check answers that refer to business documents |
| 30-50 minutes | Experience meeting-minutes generation | Generate a deliverable from input data |
| 50-60 minutes | Review the evaluation points | Organize ideas for applying it to your work |
Samples to try
| Order | Sample | Purpose |
|---|---|---|
| 1 | Search Across Large Document Sets | Experience knowledge use |
| 2 | Automatically Generate Meeting Minutes | Experience generating a deliverable from file input |
For example questions and points to check, follow the later sections "10-25 Minutes: Experience Knowledge Use" and "25-40 Minutes: Experience Meeting-Minutes Generation" at a 60-minute pace.
What to check in 60 minutes
| Perspective | What to check |
|---|---|
| Clarity | Can you understand the operational flow? |
| Answer quality | Is the answer based on the input data? |
| Business applicability | Does it appear possible to replace the sample with your documents or meeting notes? |
| Difference from AI chat | Is it using business data rather than merely holding a conversation? |
If you have more time, continue to the 90-Minute Standard Evaluation Route.
Evaluation points
Points to check
Use the following perspectives during your evaluation.
| Evaluation perspective | What to check |
|---|---|
| Usability | Can the evaluator run it without getting lost? |
| Use of input data | Can AI work with documents, models, review perspectives, and other inputs? |
| Output quality | Is the output detailed enough for business use? |
| Evidence | Are answers and findings based on the input data? |
| Reusability | Can it be used repeatedly for the same work? |
| Ease of correction | Is the format easy for people to review and correct? |
| Ease of replacement | Is it easy to substitute your own files or perspectives? |
| Workflow automation | Could it be run as standardized work? |
| Tool integration | Could deliverables be reflected in an external tool? |
| Business impact | Does it appear likely to save time, improve quality, or prevent omissions? |
How to compare it with ordinary AI chat
When evaluating Agentiqs, also check how it differs from ordinary AI chat. Comparison perspectives are summarized in Experience the Value of Agentiqs.
Common pitfalls
The following are common evaluation pitfalls.
| Pitfall | Common cause |
|---|---|
| Not knowing what to try first | There are several samples, making it hard to choose. (Start with the 60-Minute Quick Evaluation Route.) |
| Receiving a generic answer | Knowledge or an input file has not been specified. |
For more detailed pitfalls and solutions, see Experience the Value of Agentiqs, Try It with Your Own Business Files, or Experience DENSO CREATE Product Integration, depending on your route.
How to organize evaluation results
Organizing your results after the evaluation makes them easier to use in internal explanations and adoption planning. Evaluation-notes templates are available in Experience the Value of Agentiqs, Try It with Your Own Business Files, and Experience DENSO CREATE Product Integration, depending on your route.
Evaluation checklist
Use the following checklist during your evaluation.
[ ] Able to start Agentiqs
[ ] AI provider settings are complete
[ ] Able to chat with AI
[ ] Able to place the sample data
[ ] Able to import the sample settings
[ ] Ran the knowledge utilization sample
[ ] Ran the meeting minutes generation sample
[ ] Ran the design document review sample
[ ] Checked the output results
[ ] Chose a sample close to your own business
[ ] Understood how to replace it with your own files
[ ] Can explain the difference from a typical AI chat
[ ] Decided the next business theme to try
Common evaluation themes
Specific examples of evaluation themes and their corresponding samples are summarized in Experience the Value of Agentiqs and Try It with Your Own Business Files.
Notes
- First verify operation with the sample data in the evaluation version.
- When trying your own files, use copies made for verification.
- When handling confidential information, personal information, or customer information, follow your internal rules and security policies.
- Treat AI output as a draft for the person responsible to review, rather than as a final deliverable.
- Before outputting or registering content in an external tool, check the destination and update target.
- Always verify with test data before outputting directly to production files, projects, or reviews.
Next steps after the evaluation
After confirming value with the evaluation version, consider the following next steps.
| Next step | Details |
|---|---|
| Choose a business theme | Decide which work to apply Agentiqs to. |
| Organize knowledge | Organize manuals, design standards, FAQs, and similar information. |
| Create an AI agent | Create an AI agent for the business objective. |
| Create a workflow | Automate standardized work. |
| Establish usage rules | Decide responsibility for reviewing AI output, the scope of use, and the handling of confidential information. |
| Try it with a team | Use it with a small group and identify improvements. |
| Incorporate it into a business process | Incorporate it into reviews, inquiry handling, test design, or another process. |
Pages to read next
Continue to the following page based on your objective.
| Objective | Next page |
|---|---|
| Review how to begin using the evaluation version | Evaluation Version Start Guide |
| Review the basic operations | Quick Start |
| Review the sample data | Sample Data List |
| Experience the value of Agentiqs in 90 minutes | Experience the Value of Agentiqs |
| Try it with your own business files | Try It with Your Own Business Files |
| Use knowledge | Use Knowledge |
| Create an AI agent | Create an AI Agent |
| Create a review-specialized AI agent | Create a Review-Specialized AI Agent |
| Create a workflow | Create a Workflow |
| Create a workflow for comprehensive reviews | Create a Workflow for Comprehensive Reviews |
| Review the tool list | Tool List |