FAQs

Getting started

Do I need to know how to code? No. Productised is built for coaches and consultants, not developers. If you can describe what you want your product to do, you can build it.

How long does it take to build a product? Most products take between 10 and 30 minutes to build using the Create with AI flow. More complex products with custom Powers and a designed Output Page may take longer, but there is no coding involved at any stage.

What is the difference between an AI product and a chatbot? A chatbot is an open-ended conversation with no defined goal. An AI product is a structured experience — it collects specific information, evaluates it using your decision model, and delivers a personalised outcome. It has a beginning, a middle, and an end. Every person goes through it once and receives something tangible.

Products and features

What is a Power? Powers are add-ons you connect to your AI Product on the canvas. They extend what the AI can do — Research pulls in live web data, Knowledge grounds responses in your own documents, Visual Insights adds charts and visualisations, and Connections bridges your product to your CRM and other tools. See What are Powers? for a full overview.

What is an Outcome? An Outcome is what your audience receives at the end of the product experience — a Report, Document, Offer, or Redirect. See What are Outcomes? for a full overview.

Can I customise how the product looks? Yes. In the AI Product node's Design tab, you can set colours, fonts, background effects, and layout. For the output (Report, Document pages), you can use the built-in page builder to fully customise the design.

Can I embed the product on my website? Yes. From Product Settings → Deployment, you'll find your embed code and your public product URL. Both are available as soon as your product is published.

Can I use my Lovable, WordPress, Webflow, or other website? Yes — no embed code required. Copy your product's public URL from Product Settings → Deployment, then create a button or link on your site that points to it (open in a new tab). When a visitor clicks, they land on your Productised product and the AI begins the conversation. If you want the chat to start immediately without a welcome screen, set the Welcome Screen mode to Direct Start in the Canvas design panel.

Data and privacy

Where is response data stored? All response data is stored securely and is accessible only to you in your Productised workspace. You can view, export, and manage all responses from your product's analytics dashboard.

Can I export my data? Yes. From the Analytics section of any product, you can export leads, all responses, or raw JSON. See Exporting data for details.

Billing and plans

What are credits? Credits are consumed when your products are used — each AI conversation, generated report, document, or offer uses a small number of credits. Your plan includes a monthly credit allowance. See Credits for a breakdown of what uses credits.

Can I use my own AI API key? Yes. On Pro plans and above, you can connect your own Anthropic API key under Product Settings → Secrets. This reduces credit consumption for core AI runs.

AI reliability and failures

What happens if the AI fails while someone is using my product? Their answers are saved regardless. As soon as a visitor starts your product a response is recorded, so if the AI can't respond, everything they've told you up to that point — including their contact details — is kept and appears in your Analytics. A failed generation never costs you the lead. The visitor simply sees a friendly "the AI service is temporarily unavailable, please try again in a moment" message, never a technical error.

Will I know if my product stops working? Yes. If AI generation fails for a visitor, you receive a notification telling you which product failed and why in plain terms — for example, that the AI provider rejected your API key, or reported that the account is out of credits. To avoid flooding you, you'll get at most one alert per product per hour, even if many visitors are affected.

How can I tell which submissions were affected? Each response records a generation status you can see in Analytics: Completed (everything generated normally), Partial (the conversation worked but some output couldn't be generated), or Failed (the AI couldn't respond, but the visitor's answers were still saved). After a bad spell — say an expired key over a weekend — you can find the affected leads and follow up manually.

I use my own API key. What happens if it runs out of credits or gets revoked? Your product doesn't silently break. Visitors' answers are still captured, you're alerted within the hour that your key was rejected or reported a billing problem, and the visitor sees a polite "try again shortly" message. To fix it, top up or replace your key in Product Settings → Secrets — new sessions pick up the working key immediately, with no need to republish. Productised never quietly switches to platform credits when your own key fails: your key means your billing, so we alert you instead of moving the cost.

Does a brief provider outage break my product? Usually not. Temporary problems — rate limits, provider hiccups, network blips — are retried automatically before anything is treated as a failure, and most resolve on the first retry without the visitor noticing. Problems that retrying can't fix, such as an invalid key or an out-of-credits account, fail fast, save the lead, and alert you — because waiting wouldn't help.

If I pick a specific Claude model with my own key, is that the model that runs? Yes. The model you select on the AI Product node is the model that runs on your key. If you don't select one, a sensible default is used.

Do visitors ever see technical error details? No. Visitors only see a friendly, generic message. Technical details stay in your owner notification and our internal logs — they're never shown to your audience.