End User License Agreement

1. Acceptance of Terms

By creating an account, subscribing to a plan, or using Omniscia ("the Platform") and its analysis engine ("the Service"), you agree to be bound by this End User License Agreement ("EULA"). If you do not agree with these terms, do not use the Service.

2. Service Description

Omniscia provides AI-powered creative analysis for video advertisements through its suite of products:

• Lens analyses uploaded video and image ad sets to score creative diversity across 6 weighted categories, detect similarity patterns via clustering analysis, map hook fingerprints and creative branch coverage, score landing page alignment across 5 dimensions, and generate prioritised recommendations
• Pulse tracks score history, category trends, velocity, and percentile benchmarks over time (Pro+)
• Nexus (formerly Radar) provides cross-platform intelligence — unifying Meta, Google Ads, and TikTok performance data into a single model with platform grading, budget reallocation recommendations, cross-platform ROAS/CTR/CPA pattern analysis, fatigue-risk detection, and creative fingerprint matching across platforms (Starter+)
• Intel provides competitor intelligence — monitoring competitor ads via the Meta Ad Library, scoring their creatives through the Lens pipeline, and inferring performance tiers from serving duration (Starter+)
• Scia is an AI-powered creative strategist with access to your scores, campaigns, competitors, and trends (Trial+)
• Forge generates data-backed creative direction briefs from your Lens scores, Pulse trends, and campaign outcomes (all plans)
• Launch publishes ads to Meta, Google Ads, and TikTok with Cortex-informed budget, duration, and targeting recommendations
• Signal is a market intelligence pipeline that surfaces real-time trends, platform changes, and competitive shifts from 32+ sources. Includes the Helix analysis engine, which identifies trend signals from cross-platform data. Cortex-validated trends are cross-referenced against real ad performance (Starter+)
• Cortex is the self-improving ML engine at the core of the Platform, learning from the correlation between creative analysis scores and real advertising outcomes to adapt scoring weights, estimates, and recommendations over time
• Vault provides analysis history organisation with folders, search, sort, and batch operations (all plans, consolidated under Lens)
• Atlas is an interactive onboarding guide for new users

The Platform also uses creative fingerprinting to match and track ad creatives across Meta, Google Ads, and TikTok. The Service is provided on an "as-is" basis and all results, including scores, estimates, recommendations, trend analyses, and competitor analyses, are advisory and do not guarantee advertising performance.

3. Subscription & Billing

Paid plans are billed on a recurring basis via our payment processor, Paddle. Monthly plans renew on the same calendar date each month. Annual plans renew on the same date each year. By subscribing, you authorise recurring charges at your selected plan's rate. Annual billing offers a 20% discount compared to monthly pricing.

You may upgrade, downgrade, or cancel your plan at any time from the Profile page. Upgrades take effect immediately: you pay only the prorated price difference for the remaining days in your current cycle and receive a proportional share of the new tier's monthly analysis allowance. Your billing date does not change. Downgrades and cancellations take effect at the end of your current billing cycle.

3.1 Plan limits

Each plan includes monthly credits. Credits are consumed by video analysis (10 credits) and Intel deep analysis (10 credits). All other features are unlimited.

• Trial (free, 14-day Starter trial, no card required): 300 credits, up to 15 ads per set, 5 Intel brands. Full Starter feature access during trial

3.2 Overage charges

If you exceed your monthly credits, additional credits are available at your plan's overage rate: per credit. Top-up credit packs are also available. Overage charges are billed via Paddle.

4. No Refund Policy

All payments are final and non-refundable. Due to the nature of the Service, each analysis consumes third-party AI processing resources (including computer vision, native audio analysis, audio transcription, and language model inference) that incur immediate, non-recoverable costs on our end. These costs are incurred at the time of analysis and cannot be reversed.

For this reason, we strongly encourage all users to:

• Use the 14-day free trial to fully evaluate the Service before committing to a paid plan
• Start on a lower tier and upgrade only when you are confident the Service meets your needs
• Review your plan's analysis limits and features before subscribing

If you experience a technical issue that prevents the Service from functioning as described, please contact support and we will work to resolve it. However, this does not entitle you to a monetary refund.

Full details, including how this applies to subscriptions, top-ups, seats, and the UK Consumer Contracts Regulations 2013 waiver, are set out in our Refunds Policy.

5. Data Usage & Privacy

For a plain-English summary of how we handle your data, see our Trust & Security page. For full GDPR details including your data subject rights, see our Privacy Policy.

5.1 What we collect

• Account information: Email address, display name, and authentication credentials (managed via Firebase Authentication)
• Uploaded content: Video and image files you upload are encrypted and stored on secure servers for playback and reference within your account. They are retained only for as long as the associated analysis exists. When you delete an analysis from Lens or delete your account, the files are permanently removed
• Analysis results: Scores, categories, recommendations, and metadata generated from your uploads
• Usage data: Analysis counts, subscription status, and feature usage for billing and service operation

5.2 How we use your data

• Your uploaded content is processed solely for the purpose of generating your analysis and is retained on secure servers for your playback and reference. Only you (and your team members on a team plan) can view your uploads
• Files are permanently deleted when you delete the analysis from Lens or delete your account. We do not retain your content beyond your control
• Analysis results and scores are stored in your account for as long as you retain them
• Video frames are sent to AI services (Anthropic Claude) and audio tracks are sent to AI services (Google Gemini, OpenAI Whisper) exclusively for analysis via zero-retention API calls. They are not stored or used for model training by these providers
• When you link Meta, Google Ads, or TikTok ad performance data to analyses, only the AI's derived analytical classifications and aggregated performance metrics are used to train Omniscia Cortex. No video content, audio, images, transcripts, or personal information is included in training data (see Section 5.4)
• Auto-analyse (paid plans): When you connect a Meta ad account, Omniscia automatically downloads and analyses all your historical video creatives through Lens at no credit cost, and links the results to your ad performance data. This process runs automatically after connection and on subsequent syncs. Your analyses generate anonymised analytical outputs that sharpen the Cortex model for the entire platform

5.3 What we do NOT do

• We do not sell, share, licence, or distribute your data to any third party
• We do not use your videos, images, ad creatives, or personal data for training AI models — Cortex trains exclusively on Omniscia's own anonymised analytical outputs (creative classifications and aggregated metrics), which contain no identifiable content
• We do not access your Meta, Google Ads, or TikTok ad accounts without your explicit authorisation via OAuth. You control which accounts are connected and can revoke access at any time
• We do not store payment card details. All payments are handled by Paddle, our payment processor
• We do not serve advertisements or monetise your data in any way beyond the subscription fee
• We do not display your content to other users or provide Omniscia staff access to your raw creative files

5.4 Data retention & deletion

You may delete individual analyses from Lens at any time. Deleting an analysis permanently removes the associated files, thumbnails, and results. You may delete your entire account from the Profile page, which permanently removes all your data including analyses, uploads, subscription records, OAuth connections, and account information. Deletion is immediate — no 30-day hold or hidden backups.

Anonymised analytical outputs: When you link ad performance data (from Meta, Google Ads, or TikTok) to an analysis, Omniscia generates its own derived analytical outputs for the purpose of training the Cortex engine — specifically, creative classifications (e.g. "visual style: UGC", "hook type: question") and aggregated performance metrics (e.g. average ROAS, CTR). These outputs are Omniscia's intellectual property and derived intelligence, not your personal data. No video frames, audio content, images, transcripts, filenames, account identifiers, or any personal information is stored. This data is purely analytical — it describes what the AI observed, not the content itself, and cannot be traced back to you or your account. As fully anonymised data under UK GDPR Recital 26, these outputs fall outside the scope of personal data regulation and persist after account deletion to preserve the accuracy of the global ML model for all users.

Mandatory training: By using the Platform, you acknowledge that Omniscia generates anonymised analytical outputs from linked analyses as part of its core operation. This is integral to the platform's network-effect intelligence — every analysis contributes to a sharper model for all users. These outputs are Omniscia's derived intelligence and contain no personal information, video content, audio, or identifiable data.

6. Omniscia Intel — Competitor Intelligence

Omniscia Intel allows you to discover and analyse publicly available competitor advertisements via the Meta Ad Library API. By using this feature, you acknowledge and agree that:

• Competitor ad data is sourced exclusively from Meta's publicly accessible Ad Library, which exists to comply with advertising transparency regulations. Omniscia accesses this data through Meta's official API
• Competitor ad scores are generated by the same Omniscia analysis engine and are advisory. They reflect creative quality indicators, not guaranteed performance outcomes
• Performance tiers (Testing, Promising, Performing, Winning) are estimates inferred from ad serving duration and active/inactive status. They do not reflect actual spend, revenue, ROAS, or any confirmed performance metric. An ad classified as "Winning" may not be profitable. It simply indicates extended serving duration
• Deep analysis temporarily downloads competitor ad media from Meta's public Ad Library to run through the Omniscia analysis pipeline. No competitor video, audio, or image content is stored after analysis. Only derived analytical classifications (scores, creative attributes, similarity data) are retained. Source media is deleted immediately after processing
• You are responsible for ensuring your use of competitor intelligence complies with applicable advertising, competition, and intellectual property laws in your jurisdiction
• Cortex-backed performance estimates on competitor ads are advisory and do not represent actual competitor performance outcomes
• Omniscia does not guarantee the completeness, accuracy, or availability of competitor ad data, as it depends on Meta's Ad Library API
• Competitor tracking is subject to plan-based limits (number of watched brands and refresh frequency)
• If you are an advertiser whose creatives appear in Intel analysis results and you wish to request removal, you may contact us at legal@omniscia.ai or follow the process described in Section 12.3

7. Omniscia Nexus — Cross-Platform Intelligence

Omniscia Nexus provides cross-platform intelligence by unifying Meta, Google Ads, and TikTok performance data into a single model. By using this feature, you acknowledge and agree that:

• Estimates are advisory only and do not constitute a guarantee of advertising performance, return on investment, or any specific outcome
• Estimate accuracy depends on the volume and diversity of aggregate platform data. Estimates with higher confidence levels are based on larger sample sizes
• Your individual ad performance may vary significantly from estimates due to factors outside Omniscia's analysis, including audience targeting, bid strategy, market conditions, seasonality, landing page quality, and product-market fit
• Omniscia uses anonymised, aggregate data from all platform users to train scoring models. No individual user's data is identifiable in the model
• Fatigue detection and survival analysis are statistical estimates based on historical patterns and do not represent future ad longevity or fatigue timing
• Budget reallocation recommendations are advisory and based on cross-platform creative fingerprinting and performance correlation. They do not guarantee improved outcomes
• Platform grades (A–F) are composite scores factoring in ROAS efficiency, CPA trend, creative quality, and audience saturation. They are advisory assessments, not guarantees
• You should not rely solely on Omniscia estimates for advertising budget or campaign decisions

8. Omniscia Cortex & Adaptive Scoring

Omniscia includes Omniscia Cortex, a machine learning engine that learns from the correlation between creative analysis scores and real advertising performance outcomes across Meta, Google Ads, and TikTok. By using the Platform, you acknowledge and agree that:

• Adaptive weights: Scoring category weights may change over time as Cortex's ML models learn from aggregate performance data. This means the same ad set may receive a different score as the model improves. This is by design and reflects increased accuracy
• Global model: Cortex trains on anonymised, aggregate performance data from all platform users. No individual user's ad creatives, performance metrics, or account data is identifiable in the global model
• Personal intelligence: Users who link 30 or more ad-analysis pairs receive personalised scoring weights that blend their individual performance patterns with the global model. Personalised weights are derived solely from your own linked data and the global model, not from other users' individual data
• Launch recommendations: Budget, duration, and campaign setting recommendations provided through Omniscia Launch are advisory. They are based on statistical analysis and Cortex's ML estimates, and do not represent specific advertising outcomes
• Signal intelligence: Helix trend analyses and market-shift estimates within Signal are advisory. They are statistical projections based on historical cross-platform data and do not represent future market conditions or performance outcomes
• Percentile benchmarks: Percentile rankings shown across the Platform are relative positions within the current dataset. They reflect how your scores compare to other analyses, not absolute performance guarantees
• Fatigue-aware scoring: The engine may adjust scores based on similarity to your existing fatigued ads. This is designed to prevent waste but does not guarantee improved performance
• Persistent analytical outputs: Omniscia's derived analytical classifications (e.g. "visual style: UGC") and aggregated performance metrics are stored independently of your account as Omniscia's intelligence. No video frames, audio, images, transcripts, or personal information is retained — only the AI's analytical outputs. As fully anonymised, non-personal data, these outputs persist after account deletion to preserve model accuracy. See Section 5.4 for details
• Data quality safeguards: Omniscia applies minimum spend ($50), minimum duration (7 days), outlier trimming, SHA-256 deduplication, and statistical significance filters to training data. Despite these safeguards, ML model outputs should be treated as informed estimates, not certainties

9. Team Collaboration & White-Label

Agency plans include team collaboration features. By using these features, you acknowledge and agree that:

• Team roles: Company accounts support 4 roles: Owner (full control), Admin (manage members, approve reports, configure branding), Editor (run analyses, submit for review, comment on reports), and Viewer (read-only access). You are responsible for managing role assignments within your team
• Approval workflow: Editors can submit analyses for review. Admins and Owners review and approve or reject submissions from the Approvals queue
• Report comments: Team members can add threaded comments to individual reports for internal discussion
• Team seats: Agency includes 15 seats ($79 per additional seat). Each seat is a user account with the team role you assign
• White-label branding: Agency plans can rebrand shared reports and PDF exports with a custom brand name, logo, primary colour, and footer text, and optionally hide all Omniscia branding
• Shared access: All team members on a company account can view analyses created by other members. Analyses shared via public link are accessible to anyone with the link

10. API Access

Agency plans include programmatic API access to the Platform. By using the API, you acknowledge and agree that:

• Authentication: API access requires an API key generated from your dashboard. You are responsible for keeping your API keys confidential. Treat them like passwords. If you believe a key has been compromised, revoke it immediately from your dashboard. You can generate up to 10 active keys per account
• Rate limits: API requests are subject to per-key hourly rate limits: 1,000 requests/hour for Agency (full access). Exceeding your rate limit will result in temporary throttling. Do not attempt to circumvent rate limits by generating multiple API keys
• Scopes: API keys can be created with read-only or read-write scopes. Write-scoped keys can submit analyses and manage campaigns programmatically. You are responsible for all actions taken via your API keys
• Usage costs: API calls that trigger analysis (e.g. programmatic analysis submission) count against your monthly analysis limit and are subject to the same overage rates as dashboard-initiated analyses
• No SLA: The API is provided on a best-effort basis. We do not guarantee specific uptime, latency, or throughput beyond reasonable commercial standards. We may modify, deprecate, or version the API with reasonable notice
• Prohibited use: You may not resell, sublicence, or redistribute API access to third parties. You may not use the API to build a competing service. Automated scraping or bulk data extraction beyond normal analytical use is prohibited

11. Third-Party Services

The Service relies on the following third-party providers for core functionality:

• Anthropic (Claude): AI vision and language analysis of video frames
• Google (Gemini): Native audio analysis including music classification, mood detection, and sound effect identification
• OpenAI (Whisper): Audio transcription of video speech content
• Firebase (Google): User authentication and account management
• Paddle: Payment processing and subscription billing
• Railway: Backend infrastructure hosting
• Vercel: Frontend application hosting
• Meta Graph API: Ad performance data sync and campaign publishing (authorised via OAuth)
• Google Ads API: Ad performance data sync (authorised via OAuth)
• TikTok Marketing API: Ad performance data sync (authorised via OAuth)
• Brave Search API: Web search for Signal intelligence enrichment

Each provider operates under their own terms of service and privacy policies. Data shared with these providers is limited to the minimum required for their function and is governed by their respective data processing agreements.

12. Intellectual Property & Copyright

12.1 Your content

You retain full ownership of all videos, images, creatives, and content you upload. The Service does not claim any rights over your content. Analysis results generated from your content are provided for your use and are associated with your account.

12.2 Omniscia intellectual property

The Omniscia platform, including its analysis engine, scoring algorithms, intelligence models, product names (Lens, Pulse, Nexus, Launch, Intel, Atlas, Signal, Cortex, Helix, Scia, Forge, Vault), user interface designs, documentation, and all associated software, are the intellectual property of Omniscia and are protected by applicable copyright, trademark, and intellectual property laws. The Omniscia name and logo are common-law trademarks of Omniscia.

You may not copy, modify, distribute, sell, or lease any part of the Platform or its source code, nor may you reverse-engineer or attempt to extract the source code of the Service, unless applicable law expressly permits such activity.

12.3 Copyright infringement (DMCA)

Omniscia respects the intellectual property rights of others. If you believe that content available through the Service infringes your copyright, you may submit a notice under the Digital Millennium Copyright Act (DMCA) or equivalent local law to legal@omniscia.ai. Your notice must include: (a) identification of the copyrighted work; (b) identification of the infringing material and its location on the Service; (c) your contact information; (d) a statement of good-faith belief that the use is not authorised; and (e) a statement under penalty of perjury that the information is accurate and you are the copyright owner or authorised to act on their behalf.

12.4 Feedback & feature suggestions

Any feedback, suggestions, feature requests, or ideas you provide about the Service — including via the roadmap voting and feature suggestion tools — may be used by Omniscia without restriction or compensation. You grant Omniscia a non-exclusive, worldwide, royalty-free licence to use and incorporate such feedback. Roadmap votes and feature suggestions are non-binding — Omniscia retains sole discretion over product direction and development priorities.

13. Changelog & Roadmap

Omniscia maintains a public changelog and product roadmap. The changelog documents platform updates and is provided for informational purposes. The roadmap allows users to view planned features and vote on priorities. Roadmap items and timelines are indicative and non-binding. Omniscia reserves the right to modify, reprioritise, or remove roadmap items at any time without notice.

14. Limitation of Liability

The Service is provided "as is" without warranties of any kind, either express or implied. We do not warrant that the Service will be uninterrupted, error-free, or that analysis results will achieve specific advertising outcomes.

To the maximum extent permitted by law, Omniscia shall not be liable for any indirect, incidental, special, consequential, or punitive damages arising from your use of the Service, including but not limited to loss of revenue, advertising spend, or data.

14.1 Liability Cap

Our total aggregate liability to you for any and all claims arising out of or relating to the Service shall not exceed the total fees paid by you to Omniscia in the twelve (12) months immediately preceding the event giving rise to the claim. This cap applies regardless of the form of action, whether in contract, tort, strict liability, or otherwise.

15. Indemnification

You agree to indemnify and hold harmless Omniscia, its officers, and employees from any claims, damages, losses, or expenses (including reasonable legal fees) arising from: (a) your use of the Service; (b) content you upload, including ad creatives that infringe third-party intellectual property rights; (c) your violation of these Terms; or (d) your violation of any applicable law or regulation.

16. Force Majeure

Omniscia shall not be liable for any delay or failure to perform obligations under these Terms resulting from circumstances beyond our reasonable control, including but not limited to: natural disasters, acts of government, internet or telecommunications failures, cloud infrastructure outages (including our hosting providers), third-party API provider failures or rate limiting, cyberattacks (DDoS, ransomware), power failures, or pandemic-related disruptions. During any such event, our obligations are suspended for the duration of the force majeure event.

17. Account Termination

We reserve the right to suspend or terminate accounts that violate these terms, engage in abusive use of the Service, or attempt to circumvent usage limits or billing. You may terminate your account at any time from the Profile page.

18. Changes to Terms

We may update this EULA from time to time. Material changes will be communicated via email or an in-app notice. Continued use of the Service after changes take effect constitutes acceptance of the updated terms.

19. Governing Law & Jurisdiction

This Agreement shall be governed by and construed in accordance with the laws of England and Wales. Any disputes arising from or in connection with this Agreement shall be subject to the exclusive jurisdiction of the courts of England and Wales.

20. Contact

For questions about this agreement or the Service, contact us at support@omniscia.ai.