Platform
One governed lifecycle, from source to sustained training.
Snaptiq covers the full production lifecycle of training content: analysis of the source, instructional design, production, verification, publishing and sustainment. Every step leaves a record, and every decision is made by the person entitled to make it.
Products
Six products. One governed platform.
Each product does one job for one group of people. Together they take training from source to learner, and keep it current.
Training Studio
Design and produce training with AI, or from your own footage
ExploreProduction Line
Run your whole library through one governed line
ExploreSource Library
Authoritative sources, revisions and change detection
ExploreReview & Decisions
SME review and customer gates, on the record
ExploreTraining Library
Publish to learners, with the source on the player
ExploreOperations & Insights
Activity, learner viewing, health and cost
ExploreTraining Studio
Design and produce training with AI, or from your own footage.
One workspace for the whole instructional design: describe the training you need, and the Studio drafts it, tracks its coverage, produces it and takes it through review.
- A conversation that asks what it needs to know, then drafts the brief, objectives and requirements from your references
- Storyboards with narration, visuals, warnings, common errors and knowledge checks
- A traceability view of every requirement: planned, produced and verified
- Production from generated shots, presenters, narrated cards or your own footage, then review and approval in place
Production Line
Run your whole library through one governed line.
Every asset in the library moves through the same steps, from intake and change identification to customer acceptance, and back into monitoring when it’s done.
- A board of every asset by stage, with what each is waiting on and whose move it is
- Customer decision gates for priorities and for the baseline before production
- Delivery checklists, SME review and QA acceptance before anything reaches learners
- Accepted content returns to monitoring, so the library maintains itself against its sources
Source Library
Authoritative sources, revisions and change detection.
The manuals, procedures and policies your training must match, held by revision and broken down to the paragraph, so every change can be traced to the training it affects.
- Sources registered by document and revision, split into citable paragraphs
- Revisions compared paragraph by paragraph: citation gone, value changed, wording changed, renumbered
- For each source, the training that still cites it
- Safety paragraphs that no scene covers, reported as discrepancies
Review & Decisions
SME review and customer gates, on the record.
A desk for the people who check the content and a desk for the people who decide, each showing exactly what’s waiting on them and why.
- SME review of baselines and delivered packages, with the source citation on every scene
- Customer decisions recorded against stated criteria; approving against “not met” is refused
- Approvals tied to the exact version approved, so a later edit shows as stale
- Blocked actions explain themselves, and name who can take the next step
Training Library
Publish to learners, with the source on the player.
Learners watch accepted training only, with captions, chapters and a transcript, and the citation for whatever is playing on screen.
- Only accepted content reaches learners; reviewers can watch work in progress
- The source citation shown on the player for the scene that is playing
- Captions, chapters, transcript and an audio-described version
- How far learners get in each training, fed back to the people who run the library
Operations & Insights
Activity, learner viewing, health and cost.
The view for the people who run the program: what’s moving, who’s using it, what learners watched, and what the AI is costing.
- Activity, people and sign-ins, with nothing estimated that can’t be measured
- What learners watched, and the state of the library
- AI usage and estimated cost, with monthly and per-project budgets
- Content flagged for review when a source changes, and on a calendar
The platform
The services every product runs on.
Human-in-the-Loop AI
AI drafts briefs, storyboards, narration and questions. People review every draft and decide every gate.
Auto-Produce
Turns an approved storyboard into a first cut in one run: generated shots and a presenter per scene, footage scenes left for your authors, every clip labeled. Where the deployment allows AI video.
Traceability
Every scene, objective and question linked to its source paragraph and revision. When a source is revised, only the scenes it affects are flagged.
Assessment Engine
Question banks written from the approved design, rule-checked and SME-approved, exported in every major format.
Integrations
SCORM 1.2, SCORM 2004 and cmi5 packages for your LMS, single sign-on through OpenID Connect, SMTP or cloud email, and the AI services each deployment enables.
The lifecycle
Analyze → Design → Develop → Verify → Publish → Sustain
Instructional designers, producers, subject-matter experts and customer reviewers work in one place, with one record.
- 01
Analyze
A versioned training brief and a discrepancy check of the source.
- Mission need, audience, desired performance, scope
- Source analysis: wrong or outdated revision, untraced content, uncovered warnings
- Triage by risk, safety and learner demand
- 02
Design
Measurable objectives, traceable requirements, storyboards, questions.
- Objectives (LO-001…) with how each is measured
- Requirements (REQ-001…) by category, each with a source locator
- Storyboards and a question bank tagged to objectives and Bloom’s level
- 03
Develop
Generated by AI, curated from your footage, or both.
- Coverage tracked per requirement as you produce
- Captions, transcript and chapters timed to the narration
- AI-generated clips labeled as illustration on frame
- 04
Verify
Each requirement checked against evidence. The customer decides.
- Reviewer verification per requirement — never on narration alone
- SME approval of every question
- Customer gates enforced by the platform, with criteria recorded
- 05
Publish
LMS-ready packages, plus open formats your team can use.
- SCORM 1.2, SCORM 2004 and cmi5 packages for your LMS
- Captions, transcript and an audio-described version
- Traceability matrix, question sets and storyboard in open formats
- 06
Sustain
When the source changes, only what it affects is redone.
- New revisions compared paragraph by paragraph
- Affected scenes flagged; the rest keep their approval
- Review triggered by source change, and by the calendar
The loop Accepted content returns to monitoring. When its source is revised, Sustain flags exactly what to rework, and the cycle runs again for those scenes only.
Human in the loop
Generate it. Curate it. Approve it.
Let AI draft and produce the training, or have your authors bring in the video they already have. Mix both in a single training. Either way, Snaptiq does the heavy lifting and your people make every decision.
Generate it
AI drafts and produces it from your source.
Describe the training you need and point Snaptiq at the source. It drafts the brief, objectives, requirements, storyboard, narration and questions, then produces every scene in one run with generated shots and a presenter in the voice you choose.
- From source to a full first cut, fast
- Every draft traced to the paragraph it came from
- Generated shots labeled as illustration on frame
Curate it
Your authors bring the video in.
Upload your own footage, screen recordings or existing training. Snaptiq transcribes it and reads its on-screen text, suggests where each scene starts and ends, and your authors map the footage to the design. Real footage is what proves a step was demonstrated.
- Keep the footage you already trust
- Suggested scene ranges, confirmed by authors
- Mix curated and generated scenes in one training
Either way, people approve it.
Every draft is reviewed, every link to source is confirmed, and every gate is decided by your people. Then Snaptiq breaks the finished training down into everything you deliver: captions, transcript, chapters, an audio-described version, script and storyboard, traceability matrix, question bank, and SCORM and cmi5 packages for your LMS.
Traceable
Traceability per scene, not per video.
A trace link binds one scene to one paragraph of one revision of one source document. Objectives, requirements and test questions are linked the same way.
- The citation for whatever is playing sits on the player
- Approval to produce is refused until every scene cites a source
- Numbered objectives (LO-001…) and requirements (REQ-001…), each with a source locator and acceptance criteria
- Mandatory procedural steps are tracked one by one — a requirement covering six steps can’t be “covered” while two are missing
- Traceability matrix exports to Excel and CSV, at scene and at script-line level
TRACEABILITY MATRIX · excerpt · synthetic
| Req. | Source | Scenes | Verified |
|---|---|---|---|
| REQ-006 | Rev C §3.2.3 | S03 | Verified |
| REQ-007 | Rev C §3.2.4 | S04 | Verified |
| REQ-008 | Rev C §3.2.5 | S05 S06 | Pending review |
| REQ-009 | Rev C §3.3.1 | — | Gap |
Governed
The customer decides. The producer produces.
Decision gates are enforced by the platform itself, not the interface. A producer account can’t pass a gate by any route — including an administrator’s.
Gates with criteria
Each gate records an answer against every decision criterion. Approving against a criterion marked “not met” is refused, with the alternative named.
Approval bound to content
A scene stores the content fingerprint it was approved at. Edit it afterwards and the approval is visibly stale, never silently carried.
Safety content blocks, it doesn’t warn
A scene citing a WARNING or CAUTION that isn’t marked safety-critical stops approval. A manual warning no scene covers shows up as a discrepancy.
Verification needs evidence
Reviewers verify each requirement against confirmed evidence. A procedural step is never verified on narration alone.
Every seat at the table
Instructional designers, producers, SMEs, customer reviewers, QA and auditors each see what their role needs — and refusals name who can act.
A complete audit trail
Every transition, decision and refusal is recorded with who, when and against which version.
Current
A revision only reworks what changed.
Register a new revision of a source already in the library. Snaptiq compares the two paragraph by paragraph, flags exactly the scenes that cite what moved, and carries everything else forward with its approval intact.
- Content that’s out of date is ranked above content that’s merely old
- Calendar-based review for everything else, routed to the right reviewer
- Accepted content returns to monitoring — the library maintains itself against its sources
Rev D registered · 2 of 14 scenes flagged · illustrative
Human-in-the-Loop AI
AI proposes. People decide.
AI takes the slow first drafts off your team. It never takes a decision off them.
| Where AI helps | What a person still does |
|---|---|
| Drafts the training brief, objectives and requirements from your source | Edits, confirms or questions each one; human edits are tracked and kept |
| Drafts storyboards and speakable narration | Reviews the design; the customer approves it before production |
| Writes assessment questions tied to objectives and Bloom’s level | Rules check each item; a subject-matter expert approves every one |
| Proposes links between scenes and source paragraphs | Confirms the link — a proposal counts for nothing until then |
| Generates illustrative video clips | Clips are labeled on frame and never accepted as proof a step was demonstrated |
Snaptiq’s AI capabilities (document reading, speech, language models and video analysis) are each optional and switched on per deployment; with none configured, the workflow runs on its built-in tools.
Assessment
Questions that only test what the training teaches.
The question bank is written from the approved design — objectives, requirements and narration — so nothing is tested that isn’t taught.
- Multiple choice, select-all, true/false, ordering and short answer
- Each item tagged to objectives, Bloom’s level and difficulty
- Rule checks: one keyed answer, no “all of the above”, readable stems, the right answer not the longest
- SME approval of every item before a test is ready
- Question sets exported as QTI 1.2 (Canvas, Blackboard, D2L), QTI 2.1, Moodle XML, GIFT, Excel and Word, with answer keys and a syllabus
Publish
A delivery package that outlives the tool.
A package your LMS can load, plus open formats and the instructions to rebuild it. Nothing proprietary is needed to play, edit or re-render your content.
LMS
SCORM 1.2, SCORM 2004 and cmi5 packages that play the video, give the test and report completion and score
Media
MP4 (H.264), captions, transcript, chapters, poster frame and an audio-described version
Design
Brief, storyboard and script as Word documents
Evidence
Traceability matrix (Excel, CSV), decision and audit record
Assessment
Question bank and sets in QTI, Moodle XML, GIFT, Excel and Word, answer keys, syllabus
See your own source material become traceable training.
Bring a manual, an SOP or a policy. We’ll walk it through the lifecycle with you.