How to build an AI assistant people can trust.
AI shopping assistant for online shops: product cards, order status. Available for WooCommerce, PrestaShop, Magento, Shopify, BigCommerce and Wix Stores.
How Kav is being built to answer deal, case and invoice questions from your CRM or ERP through common contracts. The contracts exist; vendor connectors are planned.
How Kav is designed to capture leads with a confirmation step and look up invoice or deal status only for the verified customer. Contracts built; connectors planned.
Yes, you can connect your own system or API to Kav, even financial analysis, an ERP or a CRM. We set it up together: connectors, modules and templates.
How Kav keeps customer data secure: database-level separation between customers, encrypted secrets, audited staff access and encrypted backups.
Illustrative examples of what Kav can connect to: appointments, order status and customer records. Patterns, not a list of finished integrations.
When Kav has no approved answer it says so instead of guessing. How it fails safe, where answers come from and how a person takes over.
Honest scope for banks: a public branch and ATM locator exists today; balance and statement questions would use the same identity-first mechanism.
How a clinic configures Kav: appointment and billing lookups from its own scheduling system, with health data held to the strictest policy, plus a worksheet.
How a university configures Kav: repeat semester questions answered from registrar data, eligibility decided by an editable rule table, plus a cost worksheet.
How an insurer configures Kav: coverage questions answered from the current policy handbook, wording reviewed by two people, and claims handed to a person.
How a delivery business would configure Kav: parcel status, delivery rescheduling and problem reports from its tracking system, plus a numbers worksheet.
How a law or accounting firm configures Kav: consultation intake captured with context, meeting changes confirmed before they run, plus a numbers worksheet.
How a property manager would configure Kav: request status, service-charge balances and visit changes, using mechanisms already proven on city requests.
How a retailer configures Kav: order status, branch pickup and problem reports answered from its order and inventory systems, plus a worksheet for your numbers.
How a carrier configures Kav: data-usage and plan questions answered from the billing system as a real table, with a plan change offered in the same reply.
How a utility configures Kav: bill-spike and usage questions answered from its own billing system, with a worksheet to find where your money moves.
A five-minute way to test whether Kav pays for itself for you: the published plan prices, one formula, and the inputs only you can supply.
Yes — Kav connects to your CRM and other systems through connectors, configured once and reused everywhere. Here is what that means in practice.
Yes — WhatsApp is live on Kav, alongside the website widget and Telegram. A new organization connects its own WhatsApp Business number; Meta approves that account.
Numbers come from your systems, answers only from approved content, and a confidence gate hands off instead of guessing. Here is each safeguard against invented answers.
Connecting a system to Kav is configuration, not a development project. How long it takes depends on your system — here is what actually decides it.
Kav's published prices: small-business plans from $39 a month, Business $325 and Organization $750, priced add-ons, and Enterprise by quote.
The widget and Studio are built to WCAG 2.2 AA and IS 5568, checked by an automated gate in CI — here is what that gate covers and what it can't prove.
Kav is cloud-neutral: it runs as containers with no dependency on one cloud vendor. Here is how it's deployed and what that means for where your data lives.
Live today: the website widget, Telegram and WhatsApp. Messenger, Instagram, SMS and voice are not live yet. Here is every channel's status.
Kav's own assistant answers in Hebrew, Arabic, Russian and English. The platform also supports Amharic; each organization switches on the languages it chooses.
Kav is for businesses and organizations whose customers ask questions their own systems can answer — companies across industries, and municipalities too.
From a message arriving to an answer on screen, the same fourteen checks run every time — never a guess the model made alone.
Kav isn't a municipal-only product. Its capability model and connectors fit any business whose customers ask questions Kav could answer from the business's own systems.
Asked about a sibling discount, Kav answers with a decision from a compiled rule table — and the reason behind it — never a guess the model made.
A long picklist stops being a dropdown and becomes a search box on its own — locked options shown with the reason why, never hidden.
When an answer is genuinely tabular, Kav renders a real HTML table a screen reader can navigate — never a wall of numbers in a sentence.
How a live municipal deployment answers a property-tax balance question end to end, sourced from the city's arnona system, never guessed.
Why some questions open a sign-in first and some answer instantly: the five assurance levels behind every capability.
A write never happens on the model's say-so. Kav shows exactly what it's about to do, and carries it out only once, only on your confirmation.
A crawled page becomes a draft, never a published answer. One named person submits it, a different person approves it, before a customer ever hears it.
One rule runs through every layer of Kav: when something can't be evaluated cleanly, it denies — never guesses yes.
A structured street-work permit, with a repeatable group of streets and a real attachment, filed for real from a conversation and accepted by the city's own system.
A real resident's balance, voucher list and a live payment link, reached from one conversation — verified end to end, including exactly where it stopped short.
A clinic's appointment question on WhatsApp is handled under the platform's strictest data class, floored there automatically, not by a rule someone remembered to add.
MCP names a real problem: how a model calls real systems. Here's the mechanism Kav actually runs to solve it — compiled fresh, never a tool server you deploy.
A resident portal doesn't need its own app to install — the same capability already answers on the website widget, Telegram and WhatsApp.
The same engine already answers for a utility, a clinic, a retailer and a law firm — each configured with new rows, never new code.
A store's customer asks where an order is. Identity federates to the store's own login; Kav never sees a password or takes an account number from the chat.
Every staff look at a resident's personal data is a permanent, purpose-carrying record — and the ledger itself refuses edits.
A blank message box is a bad first impression. Kav's menu editor lets an operator write the actual words that greet a customer, with no developer.
What happens when a customer asks to change something: confirmation first, then one durable, idempotent step — never a guess.
A capability has three separate faces, and only one of them is different for every organization that installs it.
Every knowledge answer a resident can be shown passes a two-person rule first: one person submits it, a different person approves it.
A utility customer asks why their bill jumped. Kav answers with the exact usage numbers from the billing system — never a figure the model made up.
When a bot hands a conversation to a person, here's exactly what arrives with it — so the customer never starts over.
Past four results, Kav's widget turns a card list into a swipeable rail on its own, with no auto-advance, so it can't fail the rule a carousel usually breaks.
Four guarantees a prompt bolted onto your website can't make — and why each one is a real constraint Kav is built around, not a feature to point at.
Point at your API, map what it returns, tell the AI when to use it, publish. The whole build is Studio configuration — nothing to deploy.
When a customer asks what they owe, the answer must be right to the cent. How Kav keeps the words a model writes apart from the numbers your own system supplies.
When every channel gets its own inbox, residents wait in the wrong one. Here's how Kav's agent desk queues conversations by department, not by the app they came from.
