What determines the cost of AI video analytics with SquintPRO

SquintPRO is quoted on three things:

• The site. Each site is quoted on its own, because cameras, networks and systems differ from one site to the next.
• The cameras in scope. Only the cameras connected to SquintPRO count. You can start with the cameras that cover your highest-risk areas and add others later.
• The use cases. A use case is one thing SquintPRO watches for, such as missing protective equipment, entry into a restricted zone or unsafe proximity to moving equipment. Each use case in scope is part of the price.

Because use cases are priced, it pays to start with the one or two that matter most on your site, prove them, and add others once the first ones have shown results.

SquintPRO can run on your premises or in the cloud. The choice changes what you provide and what we provide.

Retention decides how far back your team can search and review. In a cloud deployment you choose 3, 6 or 12 months. Pick the shortest period that covers your investigation and audit needs, and state it in your request so the quote is prepared for that option.

Detection records do not have to depend on keeping continuous footage. Event records can be kept for reporting while the continuous recording follows a shorter policy, which is a decision for your HSE and legal owners.

SquintPRO is designed to run on cameras you already own, so the camera estate is usually the largest unknown in a quote. A camera-suitability review sorts the cameras in scope into three groups:

• usable as they are;
• usable after adjustment, such as a change of angle, cleaning, or extra lighting;
• needing replacement for the specific detection, because of distance, resolution or position.

Only the third group adds camera hardware to the quote.

An alert is only useful if it reaches the right person in a tool they already watch. Interkey builds the integration for each site: into your video management system, your control room and alerting workflow, and your incident system. The more systems an alert has to reach, the more integration work the quote includes.

We recommend starting with a pilot: one site, a defined set of cameras, one or two detection cases, and success criteria agreed in writing before anything is switched on. A pilot's cost depends on whether processing runs on temporary on-site hardware, how deep the integration goes, and how much human review the measurement needs. Ask for the pilot to be priced separately from the rollout, so the decision to continue is made on results, not on a commitment signed in advance.