Public pages

What anyone can open without signing in: the knowledge base, the support and estimate forms, surveys, lead forms, mailing-list pages, your own pages, the blog and the front page.

Most of Zenta is behind a sign-in. These pages are not: they are for your customers before they have an account, for visitors to your website, and for people who follow a link you sent them. Every one of them is switched off, or answers "not found", until the business turns on the feature behind it, and each is set up on a CRM admin screen that is linked below rather than described again here.

The addresses shown with … end in a long random code that the business's link or email carries. That code is the whole of the permission: nobody has to sign in, and a code that is wrong, or belongs to something switched off, simply answers "not found".

The bar at the top

The knowledge base, the forms and your own pages share a plain layout with a bar at the top: your name on the left (it leads to the knowledge base), and on the right the links set under Settings → Website menu & footer — the knowledge base, pages marked for the menu, your own links — and a Contact support button while the support form is open. Each link only appears while its module is on. The business can hide this bar altogether under Settings → Appearance → Sign-in page, for pages linked from a website with a menu of its own. The front page, pricing and the blog use the website's own header and footer instead.

The knowledge base

/knowledge-base opens with the heading the business chose (How can we help? as shipped) and a search box. Below it:

• The categories, each with up to six of its articles and, when it has more, All N articles.
• Watch instead — the video gallery, up to twelve videos. A video only loads when somebody presses play.
• Most read — the five articles read most often.

Typing in Search for an answer and pressing Enter replaces all of that with up to 25 matching articles, or Nothing matched. Try a different word, or ask us directly.

A category's page lists its articles, twenty to a page, each with its short description, and the category's videos under them.

An article's page shows the article, and beside it up to five Related articles from the same category and, while the support form is open, a Still stuck? card with a Contact support button. Every visit counts towards the article's views.

Was this useful? Unless the business has switched the question off, each article ends with Yes and No, and the share of readers who found it useful. Each reader has one vote: pressing the other button changes it, and pressing the same one again says You have already told us. A signed-in reader is recognised by their account and anybody else by their network and browser, and no address is stored against the vote. The answers feed the support report.

Contact support

/support lets anybody open a support ticket without an account. It links to the knowledge base first, in case the answer is already there.

From a stranger, the ticket is never filed against a customer on the strength of the address typed in: staff are offered the match and file it themselves. A contact who is signed in to the portal is different: their ticket goes straight to their company and appears under Support in their portal.

Press Send request. The form clears and says Thanks — your request is ticket #00012. We will be in touch by email. The ticket opens with the business's default status and a Medium priority. If the business acknowledges new tickets, an email arrives with the ticket's number.

A ticket from this form is never filed against a customer on the strength of the address typed into it — anybody can type anybody's address. It keeps the name and address it was sent with, and staff can attach it to the matching contact once they are satisfied it is them (Filing a public-form ticket against a contact). A customer with a portal sign-in should use Support in the portal instead, where the ticket is theirs from the start and they can follow it.

One address may send ten a minute. For a captcha as well, tick the support form under Settings → Integrations → reCAPTCHA.

Request an estimate

/request-estimate is the portal's estimate request for people who are not customers yet, with a link for those who are: Already a customer? Sign in to the portal instead. It asks for Your name and Your email (required), Company and Phone (optional), then the same questions as the portal: What would you like us to quote for?, Tell us about it, Budget, Wanted by, any questions the business added (The estimate request form), and up to five Attachments of at most 10 MB each.

Press Send request: the form clears and says Thank you - we have your request and will be in touch by email., and the Estimate request received email carries the reference. Each request opens a lead, or joins the open lead with the same email address, and appears in Sales → Estimate requests marked Public form. It has no customer yet: staff convert the lead, or pick the customer on the estimate form.

Surveys

A survey is answered from the link the business shares (the survey's own page gives it). The page shows the survey's title and description, a note when Answers to this survey are anonymous — nobody is recorded against them., and the questions: a line or a box of text, one choice, several choices, a rating from 1 to 5, or yes/no. Questions marked with a red star must be answered. Press Send my answers.

The thank-you page shows the survey's own message, or Your answers have been recorded. Instead of the questions, the page may say why it cannot be answered:

• This survey has closed., or that it opens or closed on a date the business set;
• This survey is for our customers. Sign in to the client portal to answer it. (or …for our staff), with a Sign in link, when the survey is not open to anyone;
• You have already answered this one. Thank you. for a signed-in person who has.

A survey still in draft answers "not found", so a leaked link shows nothing before it is published.

Lead forms

A lead form is a page of questions the business built (Web-to-lead forms): its name as the title, its introduction, and exactly the questions it was given, in the form's own language whatever the visitor's browser asks for. The button's words and colours are the form's too. Sending it opens a lead and tells the business; the visitor then sees the form's thank-you message (Thank you — we have your details and will be in touch. unless it was changed), or is taken to the address the form names.

A hidden field catches bots, one address may send five a minute, and reCAPTCHA is asked for when the form requires it or the business ticked lead forms under Settings → Integrations. A form that is switched off answers "not found".

Mailing lists: signing up, confirming, leaving

A list that lets people sign themselves up has a page of its own: the list's name and description, Email (required), First name, Last name, a Mobile number when the list collects one, and a box to tick saying I would like to hear from …, without which the form is refused (Please tick the box to say you would like to hear from us.). Press Subscribe. The same form can be put on the business's own website; either way one address may send ten a minute.

The thank-you page says the same whatever happened — new, already on the list, or once unsubscribed — so nobody can use it to find out whether an address is on the list. On a list that asks new sign-ups to confirm, it says a link is on its way.

The confirmation email's link opens Confirm your subscription, and nothing happens until the person presses Yes, subscribe me; the page then says You are subscribed. A button rather than the link itself, because mail scanners open every link in a message and would otherwise confirm addresses nobody meant to confirm.

The unsubscribe link in every campaign opens Unsubscribe: Stop all marketing to …?, an optional reason (Too many messages, No longer relevant to me, I never signed up, Something else, or Rather not say), and Unsubscribe me. The page then says You have been unsubscribed: no more marketing by email, text or WhatsApp, while messages about an account, such as invoices, carry on. The same address takes the one-click unsubscribe a mail client sends from its own button. Confirming and unsubscribing keep working even if the business switches Marketing off, because a link in mail already sent must go on working.

Your own pages

/p/ followed by the page's address shows a page the business wrote under Settings → Pages — delivery terms, a privacy policy, about us — while it is published and its audience is Public website. A page for the portal or for staff is never shown here; its address answers "not found".

The front page, pricing and the blog

While the website is published (Settings → Website), the root address shows the front page the business built from sections, or the ready-made page until it has built one, with the three latest blog posts at the foot unless that is switched off. While it is not published, the root address goes straight to the sign-in page (or, for somebody already signed in, to their own home page). Staff who can see Settings can look at an unpublished front page by adding ?preview to the address.

/pricing lists the platform's public plans. While companies may sign themselves up, each plan has Start free trial (or Get started for a plan with no trial); otherwise Contact us, which opens the support form. It exists only on the platform's own site, the one that sells the product: a customer workspace's website has no plans to sell, so there it answers "not found". The plans themselves are set in Plans.

/blog lists the published posts, ten to a page, newest first, each with its date, title and excerpt. A post dated in the future appears on that date; a draft never does. The blog answers "not found" while the website is not published.

A post shows its title, date and author, its cover picture if it has one, the post itself, and More writing: the three most recent other posts.

robots.txt and sitemap.xml

Both are written by the application as they are asked for, from the settings under Search engines.

• /robots.txt always answers. It always disallows /admin, /portal, /tenant/admin, /api, /install and /uploads; while indexing is off it disallows everything; it adds the business's extra rules; and while indexing and the sitemap are both on it names the sitemap.
• /sitemap.xml answers "not found" unless indexing and the sitemap are both on. It lists the front page (and pricing, on the platform's own site), the blog and its published posts while the website is published; the knowledge base, its categories and its articles while the knowledge base is public and its module is on; and every published public page: at most 2,000 of each.

Addresses on this page

For reference and for anyone scripting against the panel. Anyone can open these without signing in. Each is refused, or shows nothing, while the feature behind it is switched off.