FASQ
Frequently Asked Smart Questions
This isn’t “What is your refund policy?”. These are the questions smart buyers ask when they want to de-risk a decision. Short answers up front. The real answer underneath.
Start here (the 8 questions that matter)
If someone only reads one section, it should be this. These are the questions that decide “worth a demo?” and “can procurement approve it?”.
What problem does VidiLex actually solve? Short answer: Video contains knowledge — but it’s unsearchable. VidiLex turns it into a searchable, timestamped knowledge layer.
Most organizations already have the videos: onboarding sessions, customer trainings, handovers, workshops, webinars. The problem is retrieval. People don’t rewatch 45 minutes to find one sentence.
VidiLex makes videos behave like knowledge: search, jump-to-timestamp, and question answering with citations back to the source moment. That means fewer repeated questions, faster onboarding, and less “tribal knowledge”.
If this matters to you: You’ll feel value when your team currently answers the same questions repeatedly, or onboarding depends on “the one person who knows”.
Is this just transcription with a UI? Short answer: No. Transcription is table stakes. The core value is retrieval: search + answers + citations + navigation.
A transcript alone doesn’t solve the real workflow: “Where in our training did we explain X?” or “What’s the approved process for Y?” VidiLex structures content into chapters, generates summaries, indexes meaning, and returns answers that point back to exact timestamps.
- Search results link to the exact moment.
- Answers include citations (so users can verify, not just trust).
- Works across libraries and folders, not just one video.
If you only need transcripts: you probably want cheaper tooling. VidiLex is for teams that need reusable knowledge, not raw text output.
What would make VidiLex a bad fit? Short answer: If your content is mostly public marketing video, or you need a consumer video platform, this isn’t it.
VidiLex is built for controlled environments where video is knowledge and access needs governance.
- If you want “public hosting + viral sharing”, you want YouTube/Vimeo-like products.
- If you want “per-seat LMS compliance tracking”, you want an LMS first (and VidiLex can complement it).
- If video is rare in your org, there’s not enough material to compound value.
How do you prevent AI hallucinations? Short answer: VidiLex anchors answers in your videos and returns citations + timestamps. If it can’t find support, it should fail clearly.
“Chat with your knowledge” is only useful if users can verify. VidiLex is built around retrieval from your actual content. Answers should reference the source moments so humans can confirm.
- Citations + timestamps are part of the workflow, not a nice-to-have.
- Users can jump to the source moment immediately.
- When the platform can’t confidently support an answer, it should respond with uncertainty instead of inventing.
If this matters to you: ask for a demo with your hardest “audit-style” question. If the system can’t cite the source, that’s a red flag.
Where is our data stored and processed? Short answer: EU-hosted by default (data residency). Designed for privacy-first operation.
“EU-hosted” only matters if it’s consistent across storage and processing. VidiLex is built to support EU-grade privacy by default. If you have stricter constraints (public sector / regulated), align requirements early.
Is customer data used to train AI models? Short answer: No. Customer data is not used for model training.
VidiLex processes data to deliver the service (transcription, indexing, retrieval). It does not use your content to train foundation models. If your policy requires specific contractual wording, handle that in procurement (Enterprise supports this).
How fast do teams see value? Short answer: When you upload “high-question” content first (onboarding, trainings, handovers), value is immediate.
If you start by uploading random recordings, it’ll feel underwhelming. The smart rollout is:
- Start with the top 10–20 videos that generate the most repeated questions.
- Make it discoverable (link from internal docs, onboarding flows, customer academy).
- Track the questions people ask — that tells you what to record next.
Product & Fit
Smart questions about scope: what it replaces, what it complements, and what “good usage” looks like.
Is VidiLex a replacement for an LMS or a wiki? Short answer: Usually no — it complements them by making video usable. It becomes the “video knowledge layer”.
Wikis and LMS platforms are text-first and module-first. VidiLex is built for the reality that teams record video because it’s faster than writing, then suffer because nobody can find anything later.
The best setups link between systems: your wiki/LMS points to VidiLex timestamps; VidiLex answers point back to the source video moment.
What’s the “minimum viable library” to make this worthwhile? Short answer: If you have 10+ videos people regularly ask about, it’s already useful.
Quantity isn’t the point. “High-question density” is. Onboarding sessions, handovers, product walkthroughs, policy briefings — the videos people reference or re-explain constantly.
What makes VidiLex sticky inside an organization? Short answer: When it becomes the default place to ask “where was this explained?” — internally or for customers.
Adoption happens when three things are true:
- The content answers real recurring questions.
- People trust answers because they can click citations.
- Access and folders match how teams actually work (not a “dumping ground”).
Security & Data
The questions security teams ask when they’re serious: data residency, subprocessors, logging, access control, and audit trail expectations.
What should we expect in a security review? Short answer: Clear answers on data residency, subprocessors, access control, logging/auditability, and retention.
Security reviews stall when vendors hide behind “trust us”. Don’t do that. A productive review covers:
- Data residency: where data is stored and where it is processed.
- Subprocessors: who touches the data and for what purpose.
- Retention: what’s stored, for how long, and deletion behavior.
- Access control: roles, workspace isolation, and admin controls.
- Logging: what events are recorded and how audits are supported.
Do you support auditability and traceability? Short answer: Yes — the product is built around traceable answers (citations) and organizational access control.
“Auditability” is not just a PDF policy. In VidiLex, answers are anchored in source material through citations and timestamps. That is the difference between “AI said so” and “here is the evidence”.
Logging and role-based access control support organizational oversight (who has access, what actions occur, and what content is shared).
How do you isolate customers, teams, or departments? Short answer: With workspaces and role-based access — designed for separation from day one.
Workspaces are the boundary. This is crucial for customer academies and agencies: one platform, multiple isolated environments, no accidental mixing.
AI & Governance
Smart buyers don’t ask “which model?”. They ask: how do you control output, privacy, and operational risk?
Do we have to trust the AI, or can we verify? Short answer: You can verify. Citations and timestamps are the default behavior.
For regulated environments, “helpful answers” are not enough. People need proof. VidiLex is built around grounding answers in your recorded sources and letting users jump to the evidence.
What about multilingual teams — does this break in non-English environments? Short answer: Multilingual access is a first-class feature: ask in your language, get answers with links back to the original source.
The point is to avoid maintaining parallel knowledge bases per region. One source video can serve multiple languages through search and Q&A.
Deployment & Integrations
The practical stuff: how you roll it out, how it lives alongside existing systems, and what to expect operationally.
What’s the fastest path to a successful rollout? Short answer: Start with high-value videos, map folders to real teams, and make it the default place to ask questions.
- Pick the right content first: onboarding, handovers, product trainings, compliance briefings.
- Don’t dump everything: structure by team / customer / department from day one.
- Promote behavior change: “search VidiLex first” beats “ask Bob”.
Can we use this for customers (customer academy) without data mixing? Short answer: Yes — isolation via workspaces is the design pattern.
Many teams run one workspace per customer. That allows sharing training libraries and enablement content safely, without customers seeing each other.
Pricing & Limits
The questions finance and admins ask when they don’t want surprises: “What counts?”, “What happens when we hit limits?”, and “Is it per-seat?”.
Do we pay per user? Short answer: Plans are not seat-based. Pricing is governed by usage and concurrency.
This matters for customer academies and internal rollouts: seat pricing punishes adoption. Usage/concurrency-based pricing lets you invite the people who need access without playing licensing games.
What happens when we hit a limit? Short answer: Soft limits where possible: queueing and clear feedback. Sustained growth means upgrading for capacity.
The goal is predictable behavior, not surprise outages. If you hit concurrency or processing limits, the system should queue work and show users what’s happening. If your usage pattern becomes “peak all day”, that’s when dedicated capacity makes sense.
What should we budget for as we scale? Short answer: Concurrency + processing capacity drive scaling more than user count.
The cost curve comes from (1) how many people are active at the same time, (2) how much new video you process, and (3) how much storage you retain. That’s a more honest model than per-seat licensing when you want broad access.