6. Integration with External Services
This service may integrate with Slack, Google Workspace, Microsoft 365 (Entra ID / Microsoft Graph, including user lists and metadata on Teams, email, etc., but not message or email content), Zoom, SmartHR, freee HR (importing employee, attendance, and payroll data, and, if enabled by your admin, writing attendance records from this service to freee HR), freee Accounting (importing sales and cost by department, and, if employees connect their freee account, creating draft expense claims for approved travel, and submitting expenses based on user-defined rules such as IC card taps; for this, the connected freee account's email, user ID, rule details, and trigger dates are saved), various job boards, and other external services within the permissions you approve. Only one HR system, selected by you, is used as the source for the employee master data.
If a user logs in with a Microsoft work account, we only obtain the account's email and name, not email content, calendar events, or files.
For Slack integration, within your approved scope, we obtain post, reply, and edit times, character counts, reactions and their times, member online status, status displays (e.g., on leave), and scheduled notification-off times, using this for organizational activity and turnover analysis. To prevent missing data, our bot may auto-join public channels (never private channels unless invited).
Data obtained via integrations is used only to provide and improve this service, never for advertising or sale to third parties.
You can disconnect integrations at any time from this service's settings or the external service's admin panel.
Your administrators can issue a dedicated link (integration setup request link) to assign integration setup tasks to designated staff (such as IT personnel). The recipient can enter and confirm connection details without creating an account for this service. Employee data cannot be viewed from this link, and entered connection info is used only for the integration. The link has an expiration date and can be disabled by your administrator at any time.
If you use our browser extension for integration setup guidance, the extension displays setup steps on the external service's admin screen and assists with input and registration, with your confirmation. If the guide cannot display instructions correctly due to screen changes, only menu structure info (button/field names, not entered values, names, or emails) may be sent to our service to rebuild the guide and improve quality. Some of this info may be sent to external AI services for guide reconstruction.
The PC Attendance Tracking (PC Agent) feature allows your administrators to install our agent on employees' PCs. When installed, the agent collects and stores the following standard data: types and timestamps of PC usage changes (logon/logoff, screen lock/unlock, sleep/resume/shutdown, start/end of inactivity, and on Mac, start/end of periods without keyboard or mouse activity and elapsed seconds since last manual input, plus regular heartbeat checks to confirm the device is active), whether the microphone or camera is in use (to determine if a meeting or call is ongoing; no audio, video, or call participants are collected), overall CPU usage (not by application), device name, OS type, agent version, device serial number, and hardware identifier (used to match with your IT inventory and to identify reinstalled devices). To ensure proper operation, the agent also periodically collects the number of unsent records, recent transmission results, and available disk space. For secure device management, about once a month, it checks and displays to administrators whether OS version, disk encryption, firewall, antivirus and its definitions, screen lock, and OS auto-update are enabled (only enabled/disabled status). You can choose which items to collect; the above are standard. By default, the agent does not collect screen content, open applications or files, typed text, browsing history, or location data (inactivity is determined only by 'elapsed seconds since last input'). Administrators may optionally select additional items: 'names and usage time of foreground applications,' 'domains of websites viewed in browsers' (not full URLs or search terms), and 'titles of foreground windows.' To enable these, administrators must confirm that your work rules (or internal policies) specify them and that employees have been notified. Only devices with the additional items selected will have agents that collect them; unselected items are not collected. The additional data is displayed as daily summaries to the employee and to HR/administrators for labor management and work monitoring. Selected items are shown in onboarding and employee settings screens; if deselected, new data is not collected and the agent reverts to standard. Even with additional items, the agent does not collect screen content, typed text, audio/video, or location data. These records are used to suggest clock-in/out and break times (attendance is only recorded if approved by the employee), to display PC attendance summaries (including separating time spent with no manual input as 'programmatic activity'), and for turnover prediction/analysis (comparing current attendance and away times with the employee's past trends). Analysis is for reference only; final decisions on evaluation and treatment are made by your company. Administrators manage agent installation/removal; employees can check their device registration status in the service. Employees can always view the list of collected items and the agent program running on their company device in the employee settings screen. Employees can view their own daily attendance and history; HR/administrators can view records for all employees. When someone views another employee's attendance history, the viewer and time are logged.
The engineering analytics feature collects and stores development activity data—such as commit counts, pull requests (titles, descriptions, line changes, status), and issue progress—when you enable integration with GitHub, Linear, or similar tools. If your admin enables AI review of work, pull request diffs may be sent to an external generative AI service at review time to generate and save summaries and evaluations (score, notes); diffs themselves are not stored in our database. If you set up admin integrations with OpenAI, Anthropic, Cursor, or Manus, we collect and display AI usage per member or API key (token usage, request counts, agent task counts, credits spent; not conversation, prompts, or code content), and show usage costs per department/member. If per-member costs are unavailable, we estimate based on usage ratios. For AI tool invoices/receipts imported into contract/document management, we extract and display vendor, service, amount, and period as AI tool payments (prioritizing admin integration data to avoid duplication). If the provider or your CSV import includes breakdowns by product/feature, we save and display usage by category. If your admin enables AI usage telemetry and configures the target AI tool (Claude Code, Cowork, etc.), the tool sends per-member usage and status (interaction counts, user-issued commands, token usage by type, reported costs, time spent, task starts, AI suggestion adoption, code line changes, AI task counts and failures, provider error counts, AI-created commits/PRs, and email addresses; not conversation, prompts, code, or task content) directly to our service for storage and display. For AI review accuracy, basic employee info (tenure, position, etc.) may be used as context, and analytics may display estimated attendance from Slack and evaluation results alongside development metrics. In AI usage analytics, we compare trends for users vs. non-users, AI-involved vs. other work, and before/after AI adoption, based on usage and activity data. For before/after comparisons, the adoption date is determined from development records (AI tool signatures); users without sufficient records are excluded. These comparisons are shown only as aggregate values for groups of three or more, not individuals. We also display estimated time and cost savings based on admin-set assumptions (time saved per AI interaction, hourly labor cost); these are estimates, not actuals. Analytics results are viewable only by your admins and designated development insights staff. AI-generated evaluations and usage types are for reference only; final evaluation and decisions are made by you.
The offboarding feature allows your administrators, through explicit actions, to perform operations such as disabling accounts for departing employees, setting up email forwarding, transferring file ownership, and removing users from groups/shared mailboxes in integrated external services, within the approved permissions. These actions are performed only when individually instructed on screen, and the details, executor, date/time, and results are recorded.
The Single Sign-On (SSO) feature allows users to log in to connected external applications with their account from this service, by sending user information (email address, and, if set by the admin, name, employee number, department, position, etc.) to the application at login. The admin sets which items are sent for each application, and data is sent only when the user logs in. Login records (including success and failure) are saved and viewable by your admin. Users whose accounts are deactivated or offboarded can no longer log in via SSO.
With account auto-provisioning, when your admin sets up a connection to an external app, employee info (email, name, employee ID) is sent to that app to create or deactivate accounts. Sync is based on the admin’s selected scope (all members or specific departments) and occurs via daily automatic sync, manual sync by admin, rehire registration, or termination processing. Access tokens for connections are stored encrypted. Records of account creation/deactivation (including failures) are kept for admin review.
The account automation feature uses rules set by your admin for each application to identify accounts with no activity for a set period, based on last usage data from integrated external services. Accounts without last usage data are excluded. For identified users, a Slack direct message or email is sent as advance notice. If a contact person is specified, only they receive the notice, with a list of affected users, application names, and days unused. If there is still no activity after the grace period, the account is suspended. For applications with account provisioning, suspension is also executed in the application. For others, your IT staff receives a list of accounts to suspend.
The usage review feature helps your admin review SaaS seat contracts by displaying lists of seats and last usage dates from integrated external services, along with employee status and per-app pricing. If last usage dates are not provided per seat, we use member lists (name, email, role, status) and sign-in records (sign-in times only, not source or actions) or usage records imported into this service. The display includes seats still linked to former employees and seats in external services not matched to any employee (account name, display name, email, last usage date). This feature only displays lists; it does not suspend or cancel accounts.
For GitHub, if you install our GitHub app and allow member management, we remove the affected user's GitHub account from your organization as a suspension process, using only the account name and membership status. We do not delete repository or commit content. Organization owners are not subject to automatic removal. If the user reapplies, an invitation to rejoin is sent.
When a user requests access to an application from the app portal, we save the applicant, target application, and any reason provided, and notify the designated approver (manager, IT, or specified user) via Slack direct message or email for approval. Requests matching admin-set criteria (department, position, etc.) are approved automatically. The applicant is notified of the result. All decisions, notifications, suspensions, and issuances (including failures) are recorded with details and viewable by your admin.
The approval workflow feature saves application details entered by users (title, type, amount, type-specific fields, body), approval routes, circulation recipients, comments and action history for returns/approvals/rejections. Attached files (quotes, contracts, etc.) are stored in a tenant-specific area and viewable only by related parties (applicant, approver, circulation, admin). When it's your turn to approve, or for returns, final approval, or reminders, you may receive notifications in the service and (if Slack integration is enabled) via Slack direct message. Proxy approvals are also recorded in the history.
The 1on1 feature saves meeting schedules (participants, date/time, duration), frequency, agenda, shared notes, private notes for each participant, action items, weekly check-ins (mood, progress, blockers, requests for help), and topics users want to discuss next. Weekly check-ins and discussion topics are visible to the 1on1 partner (usually the manager) and may appear in the next agenda. Shared notes are visible to both and accessible from the notes feature. Private notes are visible only to the author. If Google Workspace calendar integration is enabled, 1on1 events (title, date/time, participants) are imported and shown in the same 1on1 view. If a user enables this, a link to the 1on1 screen is written in the calendar event notes (no invitation email is sent; this can be stopped anytime). Some agenda items are auto-generated from available employee data (behavior signals, received recognition, survey trends, evaluation goals, carryover items). Items based on behavior/survey trends and private notes are visible only to the user (and HR admins), not to the partner. If a manager writes thoughts or messages to generate sample questions and talking points, the manager's input, agenda titles, previous topics, and weekly check-ins are sent to an external AI service. Generated questions/examples are visible only to the manager, not to the partner, and the input text is not saved. If a user records a 1on1 via the Chrome extension, the transcript and summary are saved as a private note and linked to the 1on1. Only the recorder can access this note; it is not shown to others. When a 1on1 or discussion topic is set manually, the partner is notified in the service (not for calendar-imported events, as invitations are already sent via calendar).
If you enable push notifications (browser or home screen app), we store the info needed to send notifications (browser-issued address and encryption key, environment type). This info is used only to deliver notifications to your device and is deleted when notifications are turned off or the subscription expires.
For the free assessment on our website, company names you enter may be checked with an external corporate information service (gBizINFO, operated by METI) to normalize to a corporate number. No assessment results or personal identifiers are included in the query.
If you use the feature to send Slack messages (such as survey reminders) in your own name, your Slack account's access token is securely stored with your consent (OAuth authorization) and used only for sending messages. You can disconnect this integration anytime from the settings, and the token will be deleted upon disconnection.
For AI assistant integration (external AI client connection), with your consent (OAuth authorization), you can connect external AI assistants (such as Claude MCP-compatible clients) to this service. The connected AI assistant can only view or send data (employee info, org chart, recognition, surveys, etc.) and submit feedback within your permissions; it cannot access data beyond your rights. Handling of data obtained by the AI assistant follows the provider's privacy policy. You can disconnect anytime from the settings, after which the AI assistant will no longer have access. Usage records (which features were used and when) are saved for monitoring.
The meeting scheduling feature allows you to connect only your own Google Calendar (without organization-wide Google Workspace integration) with your consent (OAuth authorization). We only access your calendar's free/busy info and create events (including web meeting URLs) as needed; we do not access Gmail content or inbox. Access tokens are securely stored and used only for these purposes. You can disconnect anytime from the scheduling screen, and tokens are deleted upon disconnection. If you start using the free scheduling tool, your consent to connect Google Calendar may create a dedicated workspace, collecting only your Google account email and name. The free tool screen is separate from your company's ChordOne workspace, and even if your Google account is a company member, the data is not shared. When a booking is made, changed, or canceled, notifications are sent to the creator and participants' registered emails (for the free tool, the connected Google account email). The personal/free tool screen may display ChordOne company feature info (in-house ads); no third-party ads are shown. When you create a booking link, a unique management link (hard to guess) is issued to the creator. Anyone with this link can change conditions, stop bookings, or view the booking list without registering for ChordOne. Keep the management link safe; it is different from the public booking link.
Use and transfer of information received from Google APIs by this service complies with the Google API Services User Data Policy (including Limited Use requirements). Specifically, data obtained from Google APIs (calendar availability, agent-connected Gmail info, etc.) is used only to provide and improve features above, not for ad delivery, third-party sales, or credit decisions. Except for legal requests, security investigations, or explicit user consent, humans do not view this data. This data is not used for general AI model training.
ChordOne's use and transfer to any other app of information received from Google APIs will adhere to the Google API Services User Data Policy, including the Limited Use requirements.
The applicant tracking (ATS) feature allows you to set up a dedicated Slack integration separate from general management. If enabled, notifications about interviewers being assigned, offer approvals/results, and overdue actions (such as exceeding stage deadlines) are sent to relevant staff via Slack direct message and posted to specified channels. Notifications include candidate name, department, job title, date/time, and responsible staff. You can disconnect this integration anytime from the settings, and the token will be deleted, stopping further notifications.
If you use the hiring management (ATS) feature, you can choose which selection activities (candidate email replies, new applications, document submissions via candidate pages, stage changes, mentions in internal notes, AI screening results) to receive notifications for, how to receive them (email or Slack), where to send them (your login email, another address, or a specified Slack channel), and the scope (only candidates you handle or all candidates) from your settings. These notifications include candidate names, job titles, email subjects and excerpts, and AI screening results, so please specify only contacts or channels authorized for business use. Notifications are sent only to users with access to the hiring management feature and can be stopped anytime in settings.
Additionally, your administrators can set, for each selection activity (including interview scheduling, evaluation submissions, candidate withdrawals, and applications from agencies or alumni), who receives notifications (relevant hiring managers, recruiters, interviewers, or those with specific ATS roles) and how (email, Slack DM, or posting to a designated Slack channel, which can be set per job). If roles are specified, notifications are sent only to those with access to the candidate. Since Slack channel posts can be viewed by all members, admins should select channels limited to authorized users. Admins can also specify who to mention in Slack posts (relevant users, specified users, Slack user groups, or all channel members). To allow user group selection, the Slack integration reads group names and IDs, but only saves the selected group ID. Users can stop company-wide notifications for themselves in their settings.
The recruitment management (ATS) feature may, based on your settings, provide functions to automatically send emails to candidates at scheduled times, and to send follow-up emails if there is no reply within a set period. These are only enabled if you activate them for each email and can be canceled before sending. When receiving an email from a candidate, or when opening a candidate with an unreplied email, the correspondence (subject, body), candidate name, job title, stage, and anonymized reply samples sent to other candidates at the same stage may be sent to an external AI service to determine if a reply is needed and to draft a response. Emails judged as not needing a reply are marked as such with a reason; if a reply is needed, a draft is created but not sent until confirmed or discarded by your staff (unless auto-send is enabled). For emails to candidates/agents, we may add a record-keeping email address (CC), so replies sent to 'reply all' are received and stored as part of the candidate's email history, with originals deleted within 30 days. Only ATS users can view these records, and the feature can be disabled anytime. We may also issue a dedicated email address for application intake; if added to job board notifications, emails received (subject, body, sender, attachments) are stored as candidate info, with originals deleted within 30 days. The address can be reset or disabled anytime. If you enable the AI screening agent, candidate documents and job requirements are matched by AI to generate reference info (recommendation, needs review, etc.), with summaries saved as internal notes and not shown to candidates. For this, past screening results and anonymized candidate summaries may be sent to the AI service. If AI and staff judgments repeatedly differ, adjustment proposals are generated and shown to your admin, applied only if approved or if auto-apply is enabled. Adjustments can be reverted anytime. If staff enter reasons for differing from AI, or rewrite AI-generated drafts, these are sent to the AI service to extract short rules (criteria, phrasing, company facts) for future use, excluding names, contacts, dates, and non-job-related attributes. Extracted rules are only used after repeated confirmation or admin approval, and can be reviewed, edited, or deleted by your admin, or auto-extraction stopped anytime. AI suggestions are for reference only; all hiring decisions are made by you.
With the applicant tracking (ATS) feature, you can set custom check items and tasks for each stage of a job or selection template, and save candidate-specific entries, task status, assignees, and completion records. For each item, you can choose to share it with all hiring team members who can view the candidate, or restrict it to users with special permissions. The current settings are displayed for applications that reach the relevant stage, and entered values and completion status are retained.
In the Applicant Tracking System (ATS) feature, if you select 'recording participant' as the method for recording interviews, a recording participant operated by us (displayed with the name you set) will join your Google Meet, Zoom, or Microsoft Teams interview as a guest via browser shortly before the start. For meetings that require host approval in waiting rooms or lobbies, the participant will only join if the host allows entry, and will record audio from the meeting (no video or screen recording; the recording participant does not speak or use a camera. On Zoom, a silent microphone is connected and muted to receive audio). Depending on your settings, a message about the recording will be posted in the meeting chat upon entry. Recording stops when the meeting ends, all participants leave, or the time limit is reached. The audio is sent to an external AI service for transcription and minutes creation, then saved in your managed storage and displayed as the interview record for the candidate. Transcripts are separated by speaker, and the system calculates the speaking time ratio between candidate and interviewer, as well as the number of interviewer questions, displaying these stats by interviewer to users with access to hiring reports. These stats are for reviewing interview conduct and are not used for candidate evaluation. When importing job postings from your careers page, our server reads your company site (under your contracted domain) or a specified public careers page, displays a list of jobs, and sends the selected job page content to an external generative AI service to create a draft. Drafts are not shown externally until you publish them. If you create a job posting from audio or files, the recorded audio is sent to an external AI service for transcription, and the text or transcription results are sent for draft creation. Recorded audio and original files are not saved. Applicable interviews are those registered in the ATS for Google Meet, Zoom, or Microsoft Teams, and, if enabled, calendar events in the interviewer's Google Calendar (only if Google Workspace calendar integration is enabled) where the candidate's registered email address is invited. To confirm this, we periodically read the interviewer's calendar for upcoming event titles, times, participant emails, and meeting URLs (including those in location or description), and only save the schedule (URL, title, time) for events matching a candidate. Other events are not saved. When using the recording participant, you must inform candidates in advance about the recording and obtain their consent.
For interview scheduling links in the recruitment management (ATS) feature, if Google Workspace calendar integration is enabled, the system checks interviewers' Google Calendar availability to display bookable times to candidates. If you set specific keywords for 'treat as available' or 'event name for reception,' the system reads event titles, times, and attendance status within the target period, and treats events with the specified keywords as available slots. If 'event name for reception' is set, only those times are shown as bookable. No event data is stored.
In the recruitment management (ATS) feature, resumes and other documents submitted by candidates via their dedicated page are used only for your recruitment process and securely stored in your managed area. If a candidate is hired and linked to your employee data, and their career/skills fields are empty, resume and work history content may be sent to an external AI service to extract work history, education, skills, certifications, and languages, which are then saved as employee data. Other documents, such as residence cards, are not sent. Saved career data can be edited or deleted by those with employee data management rights. If Slack integration for recruitment is enabled, interviewers may receive Slack notifications to enter evaluations after interviews, and their input (recommendation, comments) may be saved as selection records. You can set individual data retention periods for each candidate, and admins can delete expired candidate data (including applications, interviews, evaluations, and emails) after confirmation. Deletion is manual and cannot be undone.
The applicant tracking (ATS) feature allows you to issue a login-based candidate portal where candidates can log in with their email and password. We send a guidance email (including a password setup link) to the candidate on your behalf, and passwords are stored in a non-recoverable format (we do not retain the actual password). Candidates can use this portal to check their application status, schedule or reschedule interviews, view feedback and company materials you publish, submit documents, and send questions to recruiters. If you publish an offer (job terms) on the portal, candidates can review the details and respond with acceptance or decline. You are notified of their response, which is saved in the selection record. Interviewer names and workspace profile images may be shown on the portal. Messages exchanged between candidates and recruiters on the portal are saved as part of your selection record and shown to your recruiters. Feedback visible to candidates is limited to content you create, review, and publish for them; internal interview evaluations are never shown as-is. To ensure smooth communication, the last access date to the portal and the first time an offer is viewed may be recorded and shown to your recruiters (and may trigger notifications if an offer is not opened for a certain period). You can disable access to the portal at any time. Candidate login accounts (email and name) are managed separately from your environment to allow future use across multiple companies’ portals.
If you enable candidate experience surveys in the applicant tracking (ATS) feature, a survey invitation may be automatically sent to candidates after their selection process ends, using your company name. You can prepare multiple survey templates and select which to send per job. Survey responses are optional and are shown to you as anonymous aggregates for process improvement, not for evaluating individual candidates. We may also provide a feature to use AI to organize interview notes or transcripts and generate draft evaluations; these drafts are for reference and must be reviewed and edited by your team. All hiring decisions remain with you. Weekly summaries of recruitment activity (new applications, interviews, offers, hires, etc.) may also be sent to your recruiters by email.
In the hiring management (ATS) feature, you can issue accounts for a dedicated page (recruiting agent portal) to representatives of recruiting agencies you work with, allowing them to view jobs and recommend candidates. When you register a representative's name and email, or import them from another system and send an invitation, we send an email with a login address, password setup link, and your custom message (e.g., notice of ATS change). The representative sets their own password; we do not store passwords. If you resend the invitation, the previous password link becomes invalid. Records of sent invitations (recipient, date, message) are saved in your hiring management dashboard, and you can delete or deactivate accounts at any time. Representatives register by opening the invitation link and setting their name and password; email confirmation is completed by receiving the invitation, and no code entry is required. When recommending candidates, responses to your job-specific application form (including attachments) are provided to you and saved as candidate information. You can choose whether all agency representatives or only the recommender can view each candidate and their status (default: all representatives). If set to recommender-only, candidate status is not shared in multi-agent CRM workspaces. Interview schedules (date, location, online link) are shown on the portal, and representatives are notified of changes or job closures. If you reject a candidate, we may use your input (reason, interview evaluation) to draft feedback for the agency using AI. This draft is shown only if your team reviews and chooses to share it. Your internal notes (reasons, evaluations, memos) are not shared with agencies. If enabled in settings, aggregated trends from selection results (number of applications, pass rates, common current jobs, highly rated evaluation points) are displayed for each job. Only aggregate data is shown; no personally identifiable information is included, and data is not shown for jobs with too few applications or categories with only one entry.
When your employees open ChordOne Sourcing from the hiring management (ATS) feature, we create an internal agent account in your workspace using their name and email. Candidates recommended by these employees are saved as candidate information (application source: internal sourcing). Internal agent accounts do not receive status update emails intended for agencies. If the employee's role is removed or their account is deactivated, the internal agent account is deactivated the next time ChordOne Sourcing is opened.
If you host a briefing for agencies using the hiring management (ATS) feature, we send an invitation and a link to a dedicated response page for each selected agency representative (or the company contact if no representative is registered). Their attendance responses (yes/maybe/no), comments, and response times are saved in your ATS dashboard and used only for scheduling. Responses are not shown to other agencies. If a date is set, you can create a calendar event in your Google Calendar (if Google Workspace integration is enabled), adding invited representatives as participants and sending invites from Google. Participant emails are not visible to each other. Without calendar integration, an event file is attached to each representative's email. If the briefing is canceled, the event is deleted and representatives are notified.
Those who register via the agent portal can use the candidate CRM without an invitation from a hiring company, by registering their name and email. Recruiter users ("agents") can create candidate CRMs (personal, company, or team workspaces). Company workspaces are created per email domain (excluding free email), and agents logging in with the same domain join automatically. Candidate data and communication records from personal workspaces (except those created as personal) are moved to the company workspace and become accessible to its members (duplicates are merged). If a removed agent requests to rejoin, the admin is notified and must approve. Agents can move candidate data from personal to company/team workspaces, making it accessible to workspace members (duplicates are merged). The CRM stores candidate info (name, LinkedIn/profile URL, current employer/title, skills, experience, contact, preferences, tags, notes) and communication records (scout sent/replied dates, interview notes, referrals, selection status). Opening a profile in the Chrome extension does not save candidate info; only anonymized identifiers and view counts/times are saved for matching and analytics within the workspace. If "Save to CRM" is clicked or a message is sent, public profile data (name, URL, title, employer, location, headline, education, experience, skills) is saved to the CRM. If a profile photo is public, its URL is saved and shown. If contact info is public, it is saved only when "Save to CRM" is clicked, and only if the CRM's email field is empty. Agents can upload, change, or delete candidate photos (from uploads or interview recordings; only the selected still image is sent/saved, not the recording). Photos are not used for identification. Agents can enter recruiter, university, age, gender (including "prefer not to say"), and save resumes/portfolios (up to 20MB) or portfolio URLs. Files can only be opened by CRM users and are deleted from our server when removed. Age/gender are not used for analytics, selection, or advertising. Agents can add custom fields and stages, and enter values/notes per candidate. The timeline records changes (field values, document/photo updates, stage changes) but not document content. This is not shared with hiring companies or other agents and can be deleted anytime. In the legacy workspace CRM, only name/URL/experience are saved when "Save to CRM" is clicked, and only for saved candidates are scout send/reply dates recorded. For scouts, only anonymized identifiers, send/reply dates, employer, and title are saved; company-level stats are anonymized and not shared with employers. Scout message content is not saved. When searching by text, the CRM finds similar candidates and sends the query to an external AI service to structure it, then fetches and displays public candidate info from search results. Public web data is saved only if "Save to CRM" is clicked. If interview recordings are saved, audio is transcribed and summarized by AI, and preferences (salary, timing) are extracted and filled only in empty fields. If using the CRM's phone feature (Amazon Connect), calls are recorded and transcribed, and summaries are generated by AI (not used for AI training), with recordings, transcripts, and summaries saved in the candidate's "Phone" record. For incoming calls, only if the number matches a single CRM candidate is it recorded; otherwise, it is not. Only the last 4 digits of phone numbers are saved. Agents can enter jobs under consideration and records of calls, interviews, emails, and notes. When recommending a candidate via the agent portal, only the agent's own recommendations are reflected in the CRM, including company/job name and selection stage/result. Recommendations outside the CRM are imported only if the email/name matches a CRM candidate. Company name is used to check public data (e.g., social insurance headcount) and display trends in the CRM, using only company name, not personal info. Agents can save job postings by URL or CSV; for URLs, our server reads and extracts job details by AI. Imported jobs are visible only to the agent and not shared with companies or third parties. Agents can set automation rules (e.g., on reply, stage change, job closure, inactivity), which create tasks, update candidate status, and display notifications, with execution logs saved. Automation uses only CRM data and does not contact candidates or third parties. Notifications are shown only to the agent. If "Ask AI" is used, the question and relevant CRM data (title, skills, summary, records, job criteria) are sent to an external AI service for each query, not saved in our database or used for AI training. If "Fill with AI" is used, the field prompt and candidate info (excluding contact) are sent to the AI service, and the result is saved in the CRM (not used for AI training). For filtering by text, the query and workspace field/stage names/options are sent to the AI service to generate filter conditions (no candidate data sent, not used for AI training). If daily summary emails are enabled, agents receive a morning summary of replies, tasks, selection updates, and candidate list activity (including names) to their registered email. No emails are sent during set absences; a summary is sent after. Receipt can be stopped anytime. Candidate lists in activity reports are shown only to workspace agents, not in shared links to companies. The CRM table shows the latest contact date and saves "next action date" and "next action" per candidate. Agents can export the table as CSV (up to 5,000 candidates at once); we log the agent, workspace, date, count, and scope, but not file content. In team workspaces, admins can view these logs. File management is the agent's or agency's responsibility. Duplicate candidates (same email or name/company) are shown for review and merged only if confirmed by the agent; records and files are moved and unused files deleted. Saved views can be set as "private" by the creator, hiding them from other agents (candidate data access is unchanged). Candidates with "next action date" today or overdue are notified only to the responsible agent. When importing candidates from CSV, matches by email, URL, or name/company are shown for review, and the agent chooses to merge, add, or skip per row. When merging, values are shown side-by-side and empty fields are filled. In team workspaces, admins can restrict editing/deleting to responsible agents and CSV export to admins. CRM data is visible only to workspace agents, not shared with companies or third parties. Agents can edit/delete saved candidate data anytime (subject to team workspace settings). To delete a workspace, contact support. If using the same email for ChordOne, the agent's icon shows their ChordOne profile photo (including employee directory photo), which can be changed by upload. Icons are shown to the agent and workspace members, and deleted from our server if removed.
Agents can save, per client company, the company name, contact person's name/department/title/email/phone, contract terms (fees, refund policies, etc.), and communication records in the Candidate CRM. Only workspace agents can view this information. Notwithstanding the above, if an agent creates and sends a recommendation letter to a client, only the candidate info included in the letter is provided to the selected client contact. For drafting, the candidate's CRM info (title, experience, skills, notes, meeting summaries, etc.) and job requirements are sent to an external AI service (not used for AI training). Agents can create anonymous recommendation letters (hiding names, companies, schools). Letters are sent as unguessable shared links (with expiration, revocable by the agent) from the agent's linked email account. We record and notify the agent of link opens and last access time. Agents can also set up follow-up emails to candidates at set intervals. Follow-ups are sent from the agent's linked email and stop automatically if the candidate replies, schedules a meeting, or requests to unsubscribe. To check replies, sent email threads are read. If a meeting is booked via a scheduling page, we save the booker's name, email, topic (optional), and time, and send a calendar invite from the agent's email. If the booker's email matches a CRM candidate, the meeting is added to their record. Booker info is used only for scheduling and agent contact.
Agents can generate formatted resumes from CRM data for printing or PDF (with options for anonymization or omitting desired salary). Agents can also create candidate lists for multiple candidates and share them with selected clients via secure, expiring links (revocable anytime). Only in this case is candidate info shared with the client. We record link access times and client reactions ("Interested," "Pass," and comments) and notify the agent. Resumes, recommendations, and candidate lists display the company name and logo set by the workspace owner/admin.
Internal recruiters using the agent portal (with an internal recruiter role in ChordOne ATS) can transfer CRM candidates to the ATS as applicants for their jobs. With user confirmation, candidate name, contact, experience, skills, notes, and saved documents (resume, work history, etc.) are registered in the ATS and made viewable to internal recruiters for that job. Internal recruiters' CRMs also show names and photos of current employees who share the same company (previous employer) or university as the candidate, along with the common company/university name. For this, ChordOne employee work/education history is checked in real time but not saved in the CRM. This is not shown to users without an internal recruiter role or to external agents.
With the candidate agent portal, agents can connect their mailbox (Microsoft Outlook, or Gmail if their company's Google Workspace admin has authorized our app; personal Gmail accounts are not supported) to the portal inbox with their consent (OAuth authorization). If a personal Gmail account is authorized, we immediately revoke access and do not obtain or store any mail data. For company recruiters using the agent portal, if their company has enabled ChordOne's Google Workspace integration and the admin has granted domain-wide mail read/send access, we use this authorization instead of individual consent to connect the company mailbox. In this case, we do not store access tokens, retrieving them as needed. Admins can revoke access via the Google Admin Console. If Gmail is connected via individual consent, we only request permission to send mail, not to read mailbox contents. To record sent/received emails, a dedicated logging address is BCC'd, and candidate replies are recorded via Gmail auto-forwarding to this address. While connected to Outlook or company mail, we read sender, recipient, subject, date, and the first 200 characters of recent emails, and only save these details for exchanges with candidates registered in the candidate ledger. Other emails are not saved. Email bodies are read from the mailbox only when the agent opens a conversation and are not stored in our database. When an agent replies from the inbox or sends mail to up to 50 candidates from the CRM, the subject and body entered by the agent are sent from their address to the candidate(s). For CRM mail, each candidate receives a separate email, with personalized fields (name, company) replaced, and the subject and date are recorded in the candidate's timeline (the body is not stored). Agents can mark candidates as 'Do not email,' and such candidates are excluded from bulk mail. If enabled, an unsubscribe notice is added to bulk emails, and if a candidate replies with 'unsubscribe,' the system automatically marks them as 'Do not email,' records this in the timeline, and notifies the agent (only the candidate's name is saved in the notification, not email content). Agents can also set a follow-up task for replies, which is automatically cleared upon receiving a response. Agents can schedule emails up to 30 days in advance; scheduled email bodies are encrypted until sent or canceled, then deleted. If a candidate is marked 'Do not email' at send time, the email is not sent. Outgoing emails (replies, CRM, sequences) include a small image for open tracking and links routed through our server, unless disabled by the workspace owner/admin. We record the date/time and count of image loads and link clicks per email for agent display/notification (we do not store IP addresses or device info). Open tracking may not work if the email client blocks images, and may be triggered by automatic image loading. While connected to Outlook or company mail, we also read delivery error notifications (sender, subject, excerpt) and mark the relevant send as undelivered, stopping further sequence emails to that candidate (the notification email itself is not saved). Access tokens are encrypted and used only for these purposes. Connections can be removed at any time from the inbox, deleting stored tokens and inbox lists (mailbox emails are not deleted). Alternatively, agents can record exchanges by CC'ing their dedicated logging address or forwarding candidate emails to it; we temporarily store received emails for up to 30 days, and only save the same details as above for exchanges with registered candidates. Other emails are not saved. Google confirmation codes for Gmail auto-forwarding are saved for agent display. If a resume or CV (PDF or text, up to 3 per email) is attached to an email received at this address, the content is sent to an external generative AI service to extract name, work history, skills, etc. (not used for AI model training), and the candidate is added to the agent's CRM (merging with existing candidates and filling empty fields as needed), with the file saved as a candidate document and a notification sent. Email data is not shared with employers, third parties, or other agents, and is not used for advertising or AI model training. The Chrome extension reads notification emails from job boards (e.g., 'You have a reply from XX') displayed in the agent's browser Gmail or Outlook, and adds them as replies to saved candidates. The extension does not send email bodies or names to us, only a non-reversible identifier from the board name and profile URL, and the reply date/time.
In the Notes feature, if you set a database list to “Receive by email every morning,” you will receive up to 10 matching row titles by email and in-service notification each morning. You can stop this anytime from the list menu.
To standardize company names in candidate records in the recruitment management (ATS) feature, company names are matched against the National Tax Agency's corporate number data (our copy) during candidate registration/editing or upon your request, and updated to the official name and number. We may also query external corporate info services (gBizINFO, run by METI) upon your request. Only company names are used for matching; no candidate personal info is included. The original name is retained in the candidate record.
In the agent portal, when registering a client company, we check the entered company name against gBizINFO to prevent duplicate registrations and save the selected company’s corporate number. Only the company name is used for this check; no candidate or contact personal info is included.
If you enable the organization report feature, aggregate results based on linked data (overall organization score, communication activity and positivity, interdepartmental collaboration, recognition activity, turnover risk, headcount changes, period comparisons, key changes and recommended actions, and culture trend analysis from activity data such as communication and calendars) may be regularly delivered by Slack direct message or email to HR and management contacts you specify. Reports contain only aggregated data by organization or department. You can change recipients, frequency, or stop delivery anytime from the organization report screen.
The organization report feature displays estimated turnover costs as organization-level aggregates. Calculations are based on the number of departures in a period, number of at-risk employees, admin-set assumptions (hourly labor cost, handover time, replacement hiring cost, ramp-up period, etc.), and media/agency costs registered in ATS. These are estimates only and do not guarantee actual or future costs. No individual employee cost is calculated or shown.
In culture trend analysis, Slack channel structure data (ratio of public/private channels, proportion of public conversations, channel size, etc.) may be analyzed as an indicator of organizational transparency. This analysis is handled as statistics aggregated by channel or organization; individual message content is never distributed or disclosed.
For communication analysis, frequently used words and their counts may be aggregated from Slack messages to identify trending topics within the organization. To refine the results, only the aggregated words and counts (not message content or speaker) may be sent to external AI services (such as OpenAI) to remove generic terms, merge synonyms, and classify topics.
In the talent management skill registration feature, to assist employees in registering their skills, the service may periodically extract potential work skills from their posts in public Slack channels (excluding DMs and private channels). The message content is sent to external generative AI services (Anthropic, OpenAI, etc.) only for skill extraction (not for AI training). Suggested skills are shown only as proposals on the employee’s profile and are not registered unless approved by the employee or your admin.
For company-wide feedback (feature requests, bug reports), submitted titles, content, attachments, and replies are used to improve the service. If you use AI help or the AI post assistant, your questions, conversations, and optionally attached screenshots are sent to external AI services (Anthropic, OpenAI, etc.) only to generate answers or drafts, not for AI training. For AI help, if 'Attach current screen' is enabled, a screenshot is automatically taken and sent to the AI service only for generating an answer, not stored on our server and deleted after use. You can remove the screenshot before sending. When responding to feedback, AI-generated fixes may be created; for this, we may obtain screen data (types/counts/lengths of settings, not content or names) and send it to the AI service (not for AI training). We notify you of progress (in progress, fixed, released, etc.) via in-service notifications and, for key updates, by email. For major changes, we may provide a temporary preview URL (managed by us, shared only via our reply) for you to check before release. The preview uses your normal login and real data; the URL is deleted after confirmation. To verify fixes, our system may open your view of the screen in a short (20-minute) read-only session, obtain the display and screenshot, and use it only to re-run AI fixes if needed (may be sent to the AI service, not for training). For ambiguous or new features, we may show you the post-fix screenshot for confirmation before release (not shown to others). For count discrepancies, we may perform read-only queries to our database for verification. If AI help content is deemed a feature request or bug, it is recorded as such, visible only to you and our staff, and can be deleted by you. When your feedback is resolved, ChordOne points are added to your balance as a thank you.
To share progress on collaborative improvements, anonymous aggregate statistics (such as the number of reports, resolved cases, deployments to production, and days to resolution) may be shown to users of the service. No individual titles, content, or company names are included.
When you open the feature request/bug report form, a screenshot of the current screen is automatically taken and attached as a candidate image to help us identify the context. This image is stored only in your device (browser) until you submit the form. You can review or delete it before sending; if deleted or not submitted, it is discarded and never sent to us. If attached and sent, it is handled like any other optional attachment. To better understand your issue, the service also temporarily records the addresses of screens opened in the last 5 minutes (excluding search terms), types/recipients/results of failed communications, and error details, and sends them only if you submit a report. Input text and sent content are not recorded. If not submitted, this data is not sent and is deleted when you close the screen. Submitted data is used only for handling your report (including AI-generated fixes, which may be sent to external AI services, not for AI training).
In the "Questions & Improvements" section of the Recruitment Agent Portal, agent-submitted questions, requests, and replies are handled in the same way as AI Help and company-wide feedback. By default, requests are private—visible only to the submitting agent and our staff, not to hiring companies or other agents. If "Publish to board" is selected, the subject, content, status, and our replies are shown to other agents (ChordOne Sourcing users) in the Questions & Improvements (board) menu, where others can comment or upvote. Even if published, names, companies, emails, and attached screenshots are not shown to others. Agent Portal reports are managed separately from ChordOne main (HR/recruiting/ops) feedback and are not shown to main users. If "Include current screen" is enabled, a screenshot is attached to the request (can be disabled). Status updates and confirmation requests are sent via the Agent Portal "Messages" and to the agent's registered email.
In culture surveys, employee responses (choices, scores, free text) and response times are saved. If the creator enables 'allow multiple responses per person,' multiple responses from the same employee are saved. For named surveys, the creator and authorized viewers can see each response and submission time. For anonymous surveys, responses are saved and displayed as aggregates with no identifying information.
In the Culture (Recognition) feature, messages and points sent by employees to colleagues are saved. Depending on your settings, posts may be made to a specified Slack channel or direct messages sent to recipients via Slack. If weekly point reset reminders are enabled, employees with remaining points receive a Slack DM with their remaining points (including any custom message), sent weekly at a set time. Reminders and messages can be changed anytime in settings. If an admin uses points for incentives, records of the date, period, points used per employee, points before use, admin, and reason/timing are saved and shown to the admin and affected employees. If notification is selected when saving or using points, affected employees are notified of the name, date, period, points, and admin note via in-service notification, Slack DM (if enabled), and email. Records of received recognition and points remain after use. If “AI Gratitude Suggest” is enabled, the service analyzes Slack channel activity (reply/mention counts, posts with many reactions, celebratory events, registered birthdays) to suggest colleagues to thank, shown only to the user. To generate message drafts, the system may send the candidate’s name, interaction counts, excerpts of relevant posts, recent completed tasks (if dev tool integration is enabled), and recent public recognition posts to external AI services (Anthropic, OpenAI, etc.), not for AI training. “Internal channels” means public channels and large private channels (30+ members); DMs and small private channels are excluded. Suggestions are shown only to the user and can be disabled anytime in settings.
On the employee home, announcements and polls (including quick questions) posted by users in the same company are saved and shown in the home timeline. Posts limited to a department are shown only to its members. Based on existing data in the employee directory and leave features, new hires, birthdays, and people on leave this week may be displayed on the home. If a photo is included in a new hire announcement, it is saved as their profile photo.
In the culture mini-app (group message board), messages (author name and content) written by logged-in employees are saved and, at the creator's direction, may be delivered to the recipient via Slack direct message. By default, only invited employees can view or write messages, but the creator can make it visible to all employees. When inviting members or requesting messages, the creator can send Slack DMs (including custom request text) to selected employees. When delivering (publishing), the creator can optionally enter the recipient's personal email, which is used only to send a viewing link and is not stored. After delivery, anyone with the special link (QR code) can view the board without logging in. The link is hard to guess, and pre-delivery boards are not shown. The creator can upload images for the background or cover photo, which are used for display (including via shared links) and deleted if the creator removes them or deletes the board. Admins can add company-wide background images for use and can delete them. The creator can delete the board, which also deletes all messages.
If the creator enables a 'participation link' for the group message board, anyone with the link (including external users) can write their name (free input) and a message without logging in. For company boards using ChordOne, users with accounts are encouraged to log in; if logged in, their directory name and photo are used for display (otherwise, name is free input). Only the entered name, message, chosen icon (emoji or uploaded image), and any attached photo are collected and used for display. If 'surprise protection' is on, message content and photos from other participants are hidden until delivery (names are shown); if off, all messages are visible before delivery. Uploaded icons and photos are deleted if the message or board is deleted. If 'AI draft' is used, only the selected relationship, optional keywords, title, and recipient name are sent to an external AI service for draft generation; input is not stored or used for AI training. The participation link is hard to guess and can be disabled anytime; messages via this link can be deleted. Participants can edit/delete their message before delivery using a personal edit link (saved in their browser). If an email is entered, it is used only to send the edit link and is not stored. Free boards can be created without an account; only the title, recipient name, and any uploaded images are stored. The creator receives a management link (hard to guess) for editing, deleting, or delivering the board. If the creator enters emails for invitees, only those addresses are used to send invitation links (including custom text) and are not stored. For delivery, the creator can enter the recipient's email to send a viewing link (used only for sending, not stored). The creator can delete the board from the management link, which deletes all messages and images. Boards with no messages for 30 days or judged as test entries may be auto-deleted (including messages and images). Paid ChordOne boards are not auto-deleted. Optionally, you can register with a Google account for identity verification; only your Google email and name are collected (no calendar access). If the creator links the board to their account while logged in, it becomes manageable from their in-app board list (no new data collected).
If you use the anonymous pulse survey tool for free, you can create surveys without registering an account. Only the survey title and questions are collected and stored. You receive a management link (hard to guess) for viewing results, closing responses, or deleting the survey. Respondents can answer via the link without logging in; only their selected rating (5-point scale) and any free comments are collected. No names, emails, or identifiers are collected or stored. Results are shown only as anonymous aggregates. The creator can delete the survey and all responses at any time from the management link.
For the Culture feature's mini-app (meeting slot booking), external guests can book available slots via a reservation link without logging in. To calculate availability, the system checks the calendars of specified participants (Google Calendar busy/free status only; event titles and details are not accessed). If specified, additional non-workspace calendars may be checked via the connected Google account (again, only busy/free status; these calendars are not auto-added to invites). Guest-entered names, emails, co-attendees' emails, and messages are used for confirming, changing, or canceling bookings, adding events to participant calendars (including sending invites), and generating web meeting URLs (Google Meet/Zoom) as set. Depending on settings, a specified email (e.g., shared department address) may be added to calendar invites. Guests and co-attendees may receive confirmation and reminder emails (with date, URL, and change/cancel links) as set. Change/cancel links may also appear in the calendar event description. Permissions and deadlines for changes/cancellations follow link settings; notifications are sent to relevant members by email, Slack DM, Slack channel post, or specified email as set. Slack channel posts include guest name, date/time, and assignee only (no guest email/message). Notifications to specified emails include guest name, email, and message. Paid plan customers can display images/videos (uploaded files or YouTube/Vimeo) on the booking page; uploaded files are stored in your tenant's area and shown to guests. YouTube/Vimeo videos are loaded by the guest's browser (subject to their privacy policies). Paid plan booking pages do not show our ads. Booking links are generated in a secure, hard-to-guess format and can be closed by the creator/admin anytime. Guest-facing conditions (group structure, priorities) are not shown.
In the culture mini-app (date poll), employees designated as organizers can set event date options, and selected participants can log in to respond (yes/maybe/no). The poll title, date options, participants, each participant’s responses, comments, and response times are saved and used to show organizers and participants who chose which dates. When organizers create a poll, add participants, or finalize a date, Slack DMs may be sent to participants. Organizers can delete polls, which also deletes all participant responses.
In the Culture mini app (Document Sharing), employees can upload PDF files that anyone with the share link can view without logging in. To track document access, we record which pages were viewed, viewing time, number of views, referrer, browser details, a hashed value of the IP address, and link identifiers (recipient name and notes) set by the creator. Document creators and admins can review this data for each file. Depending on the document settings, viewers may be asked to enter their name and email (and optionally company name) to access certain pages. This information is shared with the document provider (your company) and may be used for access management or follow-up. The viewing page can display your company logo, and share links may use your company’s custom domain. Share links are generated in a format that is hard to guess, and creators or admins can stop or delete sharing at any time.
The document sharing feature can be used from the service site without account registration. PDFs uploaded without registration are stored on our servers, and the uploader receives a management link to review access analytics, change settings, or delete the file. The access data recorded (pages viewed, viewing time, number of views, referrer, browser info, hashed IP address, and any name/email entered by viewers) is the same as when logged in. Files uploaded without registration are not linked to an account, so only those with the management link can manage them. We may delete files that violate laws or have not been accessed for a long period.
In the Culture mini app (Virtual Office), employees can log in and enter a 2D office view. Presence information (location in the office, mic on/off, screen sharing status, entry time) is temporarily stored and shown to other members in the same company. This data is deleted when users leave (including when the last person exits) or after a period of inactivity, and is not kept as activity logs. Audio and shared screens are exchanged directly between users’ browsers and are not stored, recorded, or relayed through our servers. Only the information needed to establish the connection is temporarily relayed and deleted immediately after. Microphone access permission is required to use audio, but entry is possible without it.
In the Culture mini app (Quiz), anyone with the participation link (QR code) can join and answer quizzes created by the host. Logged-in employees participate with their real name; if the host allows anonymous participation, users can join with just a nickname without logging in. We save the nickname (or name), selected answers, and time taken to answer. Rankings and answer distributions (how many chose each option) based on correct answers and total points (set by the host) are shown to participants and the host. If group competition is enabled, group names and affiliations are also saved, and group rankings are displayed. Images attached by the host for questions or answers are stored on our servers and shown to those with the link (answer images are shown after the correct answer is revealed). Participation links are hard to guess, and correct answers are not revealed before the deadline. If a participant selects a language other than Japanese, quiz titles, questions, and options may be auto-translated by AI, and this content is sent to external AI services (like OpenAI) for translation (names and answers are not sent). Hosts can delete quizzes and attached images; deleting a quiz also deletes all responses and images.
In the Culture mini app (Icebreaker), anyone with the participation link (QR code) can join sessions created by the organizer. Logged-in employees participate with their real name; if the organizer allows anonymous participation, users can join with just a nickname without logging in. We save the nickname (or name), answers to prompts, and votes for "whose answer it is." Answers are first shown anonymously, then after the reveal, participant names, vote distributions, and rankings based on correct guesses are shown to participants and the organizer. Participation links are hard to guess, and answer authors are not revealed before the announcement. Organizers can delete sessions, which also deletes all responses and votes.
In the Culture feature’s mini-app (event check-in), we use your issued API key to receive event participant email addresses or IC card numbers (employee ID, transit card, etc.) and check-in times from your reception system (e.g., card reader), recording them in an event table. If the email or card number matches a registered employee, it is linked to the employee/department; otherwise, it is recorded as “external/unregistered” (unregistered cards are labeled as such). Duplicate check-ins in a short period are recorded only once. Our reception app (runs on your PC) sends only card number and check-in time; it does not read card history or balance. Failed check-ins and logs are saved locally. Any other items sent (e.g., reception number, venue) are also recorded. If your system sends a new event name, a table is auto-created and the API key user (or admin if none) is set as creator. Records are saved as a Notes table, viewable/editable by the creator and shared employees. The creator can stop reception or delete rows/tables anytime.
For the time clock feature (card clock-in), we use your issued API key to receive the IC card number (employee ID, transit card, etc.) and timestamp from your card reader. The card number is matched to cards assigned to employees as IT assets (or, if specified, to asset numbers, employee numbers, or email addresses), and the first tap of the day is recorded as clock-in, the last as clock-out. The card number used for matching is saved with the attendance record as evidence. Unregistered card numbers are not recorded. Card usage history or balance is not received.
For missed clock-in applications, we store the applicant, the person submitting (self or proxy), date, clock-in/out times, work location (office or remote), optional reason, and approval/rejection result. Approved applications are reflected in attendance records and clock-in lists. If Slack notifications are enabled, application arrivals are posted to a selected Slack channel or sent as direct messages to admins/HR, and approval/rejection results are sent via direct message to the applicant and subject. Notifications include the subject's name, date, and work location, but not the reason text.
In the Culture feature's mini-app (employee aggregation), a table is created in Notes for all current employees matching the creator's selected criteria (all, by department, employment type, or individual inclusion/exclusion), with each row showing employee name, department, and email. Rows are automatically added or updated for new hires, transfers, or departures (answered rows remain). For added fields, employees log in and answer their row, with responses and timestamps saved and used for aggregation (count, sum, average, etc.). Answers can be viewed/edited by the creator and those shared via Notes. Viewing scope is set by the creator: only own answers, own plus aggregates, or all responses for the group. Based on creator actions or settings (reminder interval/time), Slack DMs and ChordOne notifications may be sent to remind unanswered employees. Existing tables in Notes can also be linked to employees for collecting responses in the same way. Linking and response collection can be stopped in table settings, and tables can be deleted from Notes.
The personality assessment feature saves the results of a 4-group, 16-type analysis and the answers based on a self-reported 5-point questionnaire completed by employees or candidates (or entered by an admin). The results are intended to support mutual understanding, placement, and development, not to be used alone for hiring or evaluation decisions. Organization-level summaries (talent portfolios) are for role design and understanding, not for individual ranking or evaluation.
The company culture assessment feature saves the results of a self-reported 5-point questionnaire about "how you see your company culture" and the resulting type classification (based on decision-making, results and evaluation, risk, and communication axes). The answers are used to visualize what your company values and to identify gaps between individual and group perceptions (shown only as anonymous summaries when enough people have responded). Individual answers are never disclosed. This feature is intended to support understanding and dialogue, not for individual evaluation or HR decisions.
For surveys and internal note databases, if your admin sets a destination URL (such as an external system, spreadsheet, or automation tool), responses or row changes (creation, update, deletion) will be automatically sent to that URL. For surveys, this includes answers and employee IDs (unless the survey is anonymous). The destination is managed by your company, not by us. Admins can delete these settings at any time; after deletion, no further data is sent. Survey requests and reminders can be sent via Slack or email (to registered employee addresses) by admin action, and we record the recipient, date, and delivery status.
The development culture assessment feature saves the results of a self-reported 5-point questionnaire and the resulting type classification (based on purpose, decision-making, speed, collaboration, and AI utilization). The answers are used to visualize development team culture and AI usage, and to support team understanding and dialogue. This feature is for organizational understanding only, not for individual evaluation or HR decisions.