Documentation

Getting started with ShieldView and getting the most out of the AI that powers it

Getting Started

Every agency gets guided, hands-on onboarding — no manuals required

1. Onboarding Session

We set up your agency's secure instance and walk your team through the platform live.

2. Role-Based Training

Investigators, supervisors, and admins each learn the workflows that matter to them. See the Training Center.

3. Ongoing Support

Questions any time? Email support@shieldviewai.com or contact support.

Getting the Best Results from ShieldView AI

The AI works best when you talk to it like a sharp new partner — clear direction, real context, and a quick review of its work

Be Specific About What You Need

Vague requests get vague results. Tell the AI exactly what you're looking for and what form you want the answer in.

Instead of: "Summarize this."
Try: "Summarize the text messages between these two numbers from March 1–15, focusing on any mention of meeting locations."

Give the AI Context

The more case context the AI has, the sharper its analysis. Mention what kind of case it is, who the key parties are, and what you already know.

Example: "This is a fraud investigation. The suspect used the alias 'DK' in earlier chats. Flag any messages where 'DK' discusses wire transfers or gift cards."

Break Big Jobs into Steps

For large evidence sets, work in stages: first a broad triage pass, then targeted follow-up questions on what it finds. Each follow-up sharpens the result.

Example: Start with "Identify all images containing vehicles," then follow up with "Of those, which show a dark-colored pickup truck?"

Check That the AI Understood You

If a result looks off, ask the AI to explain how it got there or restate what it searched for. Rephrasing your request with different wording often clears up misunderstandings fast.

Try: "Explain what criteria you used to flag those messages" — then adjust your request based on the answer.

You Stay in Control — Always Verify

ShieldView AI accelerates review — it doesn't replace investigator judgment. Treat AI findings as leads to confirm, not conclusions. Every flagged item links back to the original evidence so you can verify it yourself before it goes in a report.

Need help with something specific?

Our support team answers platform questions, walks through workflows, and helps troubleshoot — usually within one business day.

Email support@shieldviewai.com Request a Demo →