> For the complete documentation index, see [llms.txt](https://docs.archisketch.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.archisketch.com/guide/intro-en/update-note/2026/202609/20260903.md).
# \[2026.09.03] Added consultation·customer statistics dashboard and modular panel presets·AI formula generation
### (New) Added consultation/customer statistics dashboard
We added a statistics dashboard to the Consultation Management, All Consultation Management, and Customer Management pages. Instead of counting the items shown on screen yourself, it pulls the aggregated values directly from the server.
* Consultation page: total consultations, in progress, purchase completion rate, average lead time, and trends for consultation count, revenue, average quote amount, and quote count. Open the detailed statistics to see status distribution, conversion rate, consultations by assignee, and area-range distribution as well.
* Customer management: customers within the period, new customers, returning customers, and traffic source distribution. Detailed statistics include customers with many consultations and status and area distributions.
When expanded, only the summary is shown; click 'View detailed statistics' to open the rest. You can choose the query period from the last 30 days, 12 weeks, 12 months, or this year, and the statistics and the list below always use the same period.
### Bug and issue improvements
* Customer registration flow simplified — you no longer need to create a consultation first to register a customer, and a mobile entry point has been added
## 2. Parametric Editor
### (New) Modular panel structure presets provided
We added panel presets based on the Construction Principle to the modular editor. There are seven types: top panel, bottom panel, side panel, fixed shelf, movable shelf, back panel, and door, and it automatically calculates panel size and position according to Onset and Inset relationships.
If you change the furniture's W/D/H, the panel structure follows the formulas accordingly. Depending on the situation, when creating a new modular unit you can choose whether to start from a blank state or from the default cabinet template.
### (Improvement) Enhanced modular panel editor editing features
You can switch between 2D and 3D views, adjust the editing area width, set properties by template, and add offsets to panel placement. Panels can also reference variables by name, just like other components.
We also fixed issues encountered during editing. We corrected the right panel positioning error, the preset application error, and layout issues with panel animations and the edit modal.
### (New) Added natural-language-based AI formula generation
We added an AI input to the formula editor that turns the calculation you describe in words into a formula.
The generated formula is verified once more against the variables and syntax currently available. Formulas that reference nonexistent variables or have invalid syntax will not be inserted as-is. AI requests go through the server rather than the browser, so the API key is not exposed to the client. Even in environments without AI settings, you can check the behavior with a fallback response.
### Bug and issue improvements
* We added a variable list and real-time preview to the formula editor
* In the panel-style editor, when you widen the right panel, the input fields inside it widen as well
## 3. Credit system applied to AI furniture replacement & AI finish replacement
A credit system has been newly introduced for the 'AI furniture replacement' and 'AI finish replacement' features. Below are the credit deduction criteria and important notes for each step of the process.
### Image upload (initial setup)
* Free benefit: On the first day of each month, Basic (free) and Enterprise plan users receive two free upload passes per month. (Standard, Plus, and Team plans are not eligible for the free passes.)
* Credit deduction: After the free passes are used up (or for plans that do not provide them), 0.2 credits are deducted per upload.
* Important: When you upload an image, a setup process runs in which AI automatically detects the 3D space and mask area. If you close the window or leave during setup, refunding credits that have already been deducted may be difficult, so please keep the page open until setup is complete.
### Image rendering
* Credit deduction: For both AI furniture/finish replacement, 0.1 credits are deducted per rendered image.
### 3D extraction from another image (AI furniture replacement only)
* Credit deduction: 0.3 credits are deducted per generation.
* Editor integration: The extracted 3D model is saved in your personal library, and you can import it into the Archisketch main editor and place it freely in your project.
* Important notes:
* If you leave during the extraction process, deducted credits may not be returned.
* It may take some time for the extracted 3D model to be fully registered in your personal library.
* Due to the nature of AI generation, extraction quality may vary depending on the condition of the original image you upload (resolution, lighting, angle, etc.), and there may be shape differences from the actual product.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.
Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:
```
GET https://docs.archisketch.com/guide/intro-en/update-note/2026/202609/20260903.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
