Publyq Privacy Policy
1. About this Policy
This Privacy Policy explains how Sushilkumar S, trading as Publyq ("Publyq", "we", "us" or "our"), collects, uses, shares and protects personal information in connection with publyq.in, app.publyq.in and related features, APIs and services (the "Service"). We decide why and how this information is processed.
It covers two groups of people: people who use the Service or contact us, and people whose information appears in the public civic sources we use, such as complainants, other people mentioned in complaints, and public officials.
2. In short
- We build civic insights from publicly available grievance information and government responses.
- Complainants raise these complaints publicly on X. So that they can find and follow their own complaints, signed-in users can see the name a complainant posted under, and a summary of the complaints that complainant raised (a "complainant history"). Visitors without an account see neither. Anyone can ask to be shown as Anonymous, and we show a name only where there is evidence the complaint was public on X.
- We do not publish complainants' phone numbers, email addresses, postal addresses, social media handles or photographs.
- We remove personal details from complaint text where we can, and we remove records on request.
- We do not sell personal information and we do not show advertising.
- We use product analytics to see how Publyq is used and what breaks. It never records what you type, what you search for or people's names, and you can turn it off.
- You can contact us with questions or grievances at any time.
3. Information we collect
3.1 Information you give us
- Collaboration and contact forms: name, email address, phone number, what you do, and your message.
- Correction, removal and grievance requests: your contact details, the record concerned, and any information you provide to support your request.
- Accounts, where offered: name, email address and login details.
- Saved views, if you create them: a title, an optional description, the filters the view uses (which can include specific complaint numbers), whether it is private or global, and, for a global view, whether it shows your display name or "Anonymous".
- Text you type into search boxes.
3.2 Information collected automatically
- Technical information such as IP address, browser and device type, pages viewed, time of access, referring page and error logs.
- Information needed for security and fair use, such as request rates and blocked requests. If you use the Service without an account ("as a guest"), we limit requests per network address. We store only a keyed, one-way code derived from the address, never the address itself, and delete it after about an hour.
- Usage statistics, such as which features are used, whether by a guest or a signed-in user, and how quickly they respond. These statistics contain no IP address, email address, account identifier or complaint content.
- Product analytics, through PostHog: the pages and screens you open, the features you use (for example that a filter was changed, but not what it was changed to or what you searched for), errors the site runs into, your browser and device, the page that referred you, and an approximate location worked out from your IP address. If you are signed in, this is linked to your account's internal identifier, never to your name or email address. Search text, complaint numbers and complainant identifiers are removed from page addresses before anything is sent.
- Session recordings of visits to the app: how the page was used and what it showed, such as clicks, scrolling and the complaint records and statistics on screen. Everything you type, what you search for and people's names, including complainants' names and your own name and email address, are hidden in the recording before it leaves your browser.
- Preferences stored on your device, such as language, theme and filters (see section 15).
3.3 Publicly available civic and social media information
We collect information from public sources, including:
- complaint references that government bodies publish on their public social media accounts;
- the corresponding records on public government grievance portals, such as complaint number, date, category, department, location description, complaint text, status, and official responses and their dates;
- public social media posts through which residents raise grievances; and
- public information about public officials and elected representatives, such as name, role, area served, party and links to public profiles.
These sources can contain personal information, such as a complainant's name, contact details, address, social media handle or photographs, or personal details written into the complaint text. Information being publicly available does not mean it raises no privacy concerns, so we apply the safeguards described below.
3.4 Complainants' names, and what we do not keep
When we collect a record, we remove the complainant's phone numbers, email addresses, postal address, gender, property assessment number, social media handle and photographs, and we do not keep them.
Most complaints on Publyq were raised publicly: residents post them on X, and the Corporation registers them under the name the resident posted with. We show that name, so that a complainant can find all of their complaints in one place and follow what happened to them, and so that each record shows who raised it.
- Only names from public posts. Every complaint on Publyq was reached through X, but the channel the grievance portal records is not reliable: complaints made in public posts are often labelled as direct messages or left blank. So we show a name where the portal records the complaint as a public post on X, or where the complaint links to the public X post it was made in, whatever the channel label says. Where there is neither, the complaint is shown as Anonymous, and we do not store that name in our database.
- To signed-in users only. Names are shown to people signed in to an account, not to visitors exploring without one. A shared (global) saved view may follow one complainant: signed-in users who open it see that complainant, while visitors without an account see the view with the complainant locked, and never receive their name or any code that identifies them.
- Only the name. We show the name as the portal records it, without the channel label the portal adds. Where the portal's copy of a name is unreadable or missing but we know which X account posted the complaint, we show the name that account's other complaints were registered under. Otherwise we show none.
- Grouping by complainant. To show one person's complaints together, we treat the complaints posted from one X account as coming from one complainant, whichever wards they concern. Two people who share a name stay separate. Where we cannot tell which account posted a complaint (for example, because the post was deleted), or the name the Corporation recorded is not the one that account usually uses, we treat complaints with the same name in the same ward as one complainant. If no X account we know of uses that name at all (for example, because the account's posts were all deleted), we treat complaints with that name as one complainant across every ward. Where a grouping rests on the name alone across wards, or more than one X account has used the name in a ward, we say that the complaints may come from more than one person. Tell us if a grouping is wrong and we will correct it.
- How we check groupings. To tell whether a name belongs to one person or several, we look up which public X account posted each complaint, using X's public post pages. We keep the account's identifier and handle only in our internal working files, which only the Publyq team can access. The database the Service reads from holds only a one-way code made from the account's numeric identifier, never the handle. We never show the identifier, the handle or the code on the Service, and never use them to contact anyone. They are used only to group complaints by complainant and to count how many different people report the same problem in the same place.
- Complainant history. A signed-in user can see one complainant's complaints together, with their name, their ward (or how many wards) and how many of those complaints are open, awaiting a reply, resolved or closed. It is worked out from those published complaints each time it is opened. We do not store it, and it adds nothing that the complaints themselves do not show. We deliberately do not break a person's complaints down by subject, department, place or time. To find a complainant, a user types a name, optionally with a ward number, and sees at most five matches, each with its ward or number of wards. A name shorter than three letters must be typed in full. Each account, and each network address, can look up only a limited number of different complainants and run a limited number of name searches each day. A complainant history is never available to visitors without an account. A shared saved view that follows a complainant must not name them in its title or description, and we refuse one that does.
- Signing in does not prove who you are. An account does not show that you are any particular complainant, so being signed in never gives access to more about any complainant, including yourself. To see, correct or remove your own records, write to us (see section 13).
- Anonymous on request. Anyone can ask to be shown as Anonymous, for all of their complaints or for a single one (see section 13). Once we act on a request it applies everywhere on the Service straight away, to all of that complainant's complaints in every ward, including complaints added later. Where some complaints under the same name cannot be linked to an X account (for example, because the posts were deleted), we hide those too, as they may be the same person. Signed-in users also see an "Ask to be anonymous" link beside each complainant.
We keep a limited technical record of each collection, such as the complaint reference, the source link, the time of collection and a digital fingerprint of the source page, so that we can show where information came from.
3.5 Information we generate
From the information above we create derived information, including ward, zone and constituency mapping, standard categories, translations, groupings of similar complaints, labels such as category, severity or urgency, statistics, trends, and views of complaints by area and by the officials and representatives responsible for that area.
4. How we clean, classify, enrich and analyse data
- Collect: we retrieve records only through public pages and public references.
- Minimise: we remove contact details and other direct identifiers, show complainants' names only as described in section 3.4, and redact personal details such as phone numbers, email addresses, house numbers and personal names from complaint text, using automated tools and review. No method is perfect, so we also act on removal requests.
- Standardise: we clean formatting and map categories and departments to standard labels.
- Locate: where a record does not state its ward, we infer it from the location description using official boundaries.
- Translate and group: we translate Tamil text into English and group similar complaints.
- Analyse: we produce labels, statistics and trends, mostly at the level of wards, zones, departments and constituencies.
Information we infer is marked where practical. Status reflects what the source showed on the date of collection.
5. AI and automated analysis
- We use an AI model to translate complaint text, and rule-based methods to classify, group and label records. For example, a complaint is labelled a safety risk from the Corporation's own complaint category and from what the complaint describes (such as an open manhole, an injury or a fall), and as a repeat when the same problem is reported again at the same place. Each label shows why it was given.
- We do not use these tools to make decisions about individuals, to profile complainants, or to assess individual public officials. Officials are shown by the area they serve. Grouping complaints by complainant, and the complainant history (section 3.4), are used only to show a person's published complaints together and how the Corporation handled them. They never score, rate, rank or predict anything about the complainant, never describe their character, reliability or motives, and are never used to make decisions about them or to contact them.
- We translate complaint text with IndicTrans2, an open source model from AI4Bharat that we run on our own systems. Complaint text is not sent to an outside AI service.
- AI generated output can be wrong. We label it as machine generated.
6. How we use information
- To operate, maintain and secure the Service.
- To publish civic insights in the public interest.
- To show signed-in users who raised a complaint, the other complaints raised by the same person, and how many of them are open or resolved (the complainant history), as described in section 3.4.
- To respond to your messages, collaboration enquiries, grievances, and correction or removal requests.
- To prevent abuse, enforce our Terms of Service and apply usage limits.
- To improve the accuracy and usefulness of the Service.
- To understand how the Service is used and to find and fix what does not work, through product analytics and session recordings (see section 3.2).
- To comply with law and legal process, and to establish, exercise or defend legal claims.
We do not sell personal information, use it for advertising, or build marketing profiles of individuals.
7. Legal basis and Indian privacy law
- We handle personal information in line with the Information Technology Act, 2000 and the rules made under it, including the Information Technology (Reasonable Security Practices and Procedures and Sensitive Personal Data or Information) Rules, 2011, where they apply. We do not intend to collect sensitive personal data or information as defined in those rules.
- The Digital Personal Data Protection Act, 2023 and the Digital Personal Data Protection Rules, 2025 are being brought into force in phases. Some provisions are already in force. Others, including most of those on notice, consent, the duties of organisations that handle personal data and the rights of individuals, take effect on later dates notified by the Government of India. We are applying this framework's principles now: we collect only what we need, use it only for the purposes in this Policy, keep it accurate and secure, keep it only as long as we need it, and act on requests from the people it concerns. We will update this Policy as further provisions take effect.
- Information you give us: we process it with your consent, for the purpose for which you gave it. You may withdraw consent at any time. Withdrawal does not affect processing that took place before it.
- Publicly available civic information: we process it for civic transparency, research, statistical and public interest purposes. Indian law treats some publicly available personal data, and some research and statistical processing, differently from other processing. Whether and how those provisions apply depends on the source and nature of the information and on when the provisions take effect. The complainants' names we show, and the complainant histories built from them, are limited to names complainants themselves made public, by posting their complaints on X under that name. The Act does not apply to personal data that the person it relates to has made publicly available. Even so, we treat these names as personal information. We show them only to signed-in users, with the limits in section 3.4, and we stop showing them as soon as someone asks. In every case we apply the minimisation, redaction and removal practices in this Policy.
- Legal obligations: we may process information where the law, a court order or a lawful direction requires it.
8. Sharing information
- Publicly on the Service: Publyq generated insights and redacted records. Complainants' names and complainant histories, only to signed-in users and only as described in section 3.4. We never publish complainants' contact details or social media accounts.
- Saved views you share: a global saved view, with its title, description and filters, can be seen by anyone, including visitors without an account (a filter for one complainant is shown only to signed-in users), together with your display name unless you chose to share it anonymously. When you share anonymously, other people see "Anonymous" and we store no name with the view, but we can still link it to your account to handle abuse or legal requests. A private view is visible only to you.
- Service providers: companies that help us run the Service, namely Vercel (website hosting), Neon (database), Supabase (accounts, sign-in and answers to our collaboration form), PostHog (product analytics, session recordings and error reports, stored in the European Union) and Google (Sign in with Google, email through Google Workspace, and web fonts). They receive only what they need to provide their service.
- Legal requirements: government agencies, courts or others where required by applicable law or valid legal process, or where reasonably necessary to protect the safety or rights of any person.
- Successor entity: if the Service is transferred to a registered entity or another operator, information may be transferred to it, subject to this Policy.
9. Processing outside India
Some of our service providers may store or process information outside India. We choose providers that maintain reasonable security safeguards, and we will comply with any restrictions on transfers that the Government of India notifies under applicable law.
10. How long we keep information
| Information | How long we keep it |
|---|---|
| Collaboration and contact form submissions | 12 months after our last contact with you |
| Account information | Until you close your account, then up to 30 days |
| Saved views | Until you delete them or close your account, then up to 30 days. A shared link stops working as soon as the view is made private or deleted |
| Request counters used for fair use limits (keyed codes, never the IP address, email or account id) | About 1 hour for per-minute limits, and about 2 days for daily allowances |
| Record of which complainants an account looked up, used only to count against the daily limit (a keyed one-way code for each account and complainant pair, never the name or the account) | About 2 days |
| Complainant histories | Not stored. They are worked out each time they are opened, and the response is marked so that browsers and networks do not keep a copy |
| X account identifiers and handles behind complaints (internal working files only, see section 3.4) | As long as we hold the complaint they relate to |
| Usage statistics (no personal information) | 90 days |
| Product analytics and error reports | 12 months |
| Session recordings | 30 days |
| Grievances and correction or removal requests | 3 years, keeping only what we need to show how the request was handled |
| Technical and security logs | 90 days, or up to 12 months where needed to investigate a security issue |
| Redacted copies of source pages | 12 months from collection |
| Complainants' names | As long as we show the complaint they belong to (see published record level data below), or until the person asks to be shown as Anonymous |
| Requests to be shown as Anonymous | As long as we hold that person's complaints, so the name stays hidden when the data is updated. We keep only what we need to apply the request: a one-way code made from the name and ward or from the X account that posted the complaints, or the complaint reference |
| Collection records (reference, source link, time, fingerprint) | For as long as we hold information derived from that record, and as needed for legal purposes |
| Published record level data | 24 months from the complaint date, after which it is kept only in aggregated form |
| Aggregated statistics with no personal information | Without a fixed time limit |
We may keep information for longer where the law requires it or where it is needed for an ongoing legal matter.
11. Security safeguards
We use reasonable security practices suited to the information we hold, including:
- encryption of data in transit;
- access limited to people who need it;
- separation of collected records from published data, showing complainants' names and complainant histories only to signed-in users and never sending them to visitors without an account;
- for complainants: opaque codes in place of any identifier that could be worked out from a name, daily limits on how many complainants an account can look up or search for, no caching of complainant histories, and requests to be shown as Anonymous applied straight away;
- encrypted backups of collected records;
- redaction before information is sent to external processors;
- logging, monitoring and rate limiting; and
- review of the service providers we use.
No system is completely secure. If a security incident affects your personal information, we will take reasonable steps to limit its impact and will notify you and the relevant authorities, including CERT-In, where required by law.
12. Your rights and choices
Whether or not a specific law requires it, you can ask us to:
- tell you what personal information you have given us and how we use it;
- correct or update information you have given us;
- delete information you have given us, including your saved views, or withdraw your consent;
- redact or remove a record that contains your personal information or concerns you; and
- review a concern about how we handle personal information.
You can also turn off product analytics for your browser at any time (see section 15).
Once the relevant provisions of the Digital Personal Data Protection Act, 2023 are in force, you will also have the rights that Act provides. These include the right to nominate a person to exercise your rights in case of death or incapacity, and the right to complain to the Data Protection Board of India after using our grievance process.
To find the records that concern you, it helps to give us a complaint reference or link; we can also search by the name you posted under and your ward. Being signed in to an account does not show that a record is yours, so we may ask for reasonable information to verify your request, such as a post from the X account the complaint was made from, and we will ask only for what we need. If we cannot meet a request, for example because the law requires us to keep certain information, we will explain why.
13. Correction and removal requests
If you are a complainant, a person mentioned in a complaint, or a public official, and you want information corrected, redacted or removed:
- write to auth@publyq.in, including the link or complaint reference and what you would like us to do. To be shown as Anonymous on all of your complaints, give the name you posted under and your ward, or any one of your complaint references;
- we acknowledge requests within 72 hours;
- we aim to show a complainant as Anonymous, or to redact or remove a person's own personal information, within 7 days; and
- we aim to decide other requests within 15 days.
Removal from Publyq does not remove information from the original government portal or social media platform. Please contact them directly to change the original.
14. Children
The Service is not directed at children under 18, and we do not knowingly collect personal information from children through our forms or accounts. If you believe a child has given us personal information, please contact us and we will delete it. If a civic record mentions a child, we treat it as a priority for redaction and remove it on request. We do not track children or target advertising to them.
15. Cookies, local storage and analytics
- Essential: we use cookies or local storage to remember preferences such as language, theme and filters, to keep you signed in or remember that you chose to continue as a guest, and for security and fair use.
- Analytics: we use PostHog to understand how the Service is used (see section 3.2). It keeps a random identifier in a cookie and in local storage, shared by publyq.in and app.publyq.in, so that one visit across both counts once. It does not run if your browser sends a Do Not Track or Global Privacy Control signal, and you can turn it off for this browser below.
- We do not use advertising cookies.
- You can control cookies and local storage through your browser settings. Blocking essential storage may affect how the Service works.
16. Third party links
The Service links to government portals, social media platforms and other websites. Their own privacy policies apply when you visit them.
17. Changes to this Policy
We may update this Policy as the Service or the law changes, including as provisions of the Digital Personal Data Protection Act, 2023 take effect. We will post the updated Policy on this page with a new "Last updated" date. If a change is significant, we will also give notice on the Service and, if you have an account, by email.
18. Contact and grievances
- auth@publyq.in
We acknowledge grievances within 72 hours and aim to resolve them within one month.