Everything you need to know about using CP Analyser: getting started, uploading documents, configuring an analysis, understanding your results, and troubleshooting common questions.
On this page
- Category 1: Getting Started & Access (7 FAQs)
- Category 2: Creating an Assessment & Uploading Documents (4 FAQs)
- Category 3: Configuring the Analysis (4 FAQs)
- Category 4: Understanding the Results (6 FAQs)
- Category 5: Working Through Findings & Building the Counter (4 FAQs)
- Category 6: Validations (8 FAQs)
- Category 7: Feedback, Audit Trail & Accounts (6 FAQs)
- Category 8: Troubleshooting & Known Issues (8 FAQs)
Category 1: Getting Started & Access
Section 1.1: Introduction & Overview
1. What is CP Analyzer?
Answer: CP Analyzer is an AI-powered charter party (CP) risk-analysis tool for pre-fixture deal-making. It reads a charter party and the proposed recap, checks them against a set of validations, and flags risks and non-standard terms so chartering teams can negotiate with confidence - in minutes instead of hours.
2. What problem does it solve?
Answer: It replaces slow, inconsistent manual CP review. Instead of reading every clause by hand, the AI runs the same checks every time, catches things that are easy to miss (especially *missing* terms), and returns a prioritised risk list with suggested resolutions. Users report it roughly halves review time.
3. Is CP Analyzer pre-fixture or post-fixture?
Answer: Pre-fixture. It assesses risk *before* a deal is fixed. Once a voyage is fixed, the same documents (CP, recap, supporting files) feed the post-fixture products - but CP Analyzer itself stops at the negotiation stage.
Related: FAQ #1
Section 1.2: Accessing the Application
4. How do I access CP Analyzer?
Answer: It's a standalone web app. Customers log in with their email address and a password provided during onboarding. (The beta runs at a dedicated URL supplied in the credentials email - see caveats note on branding.)
5. Is CP Analyzer inside Sedna?
Answer: Not yet. It currently runs as a standalone app; Sedna integration is on the roadmap, not live. Treat it as a separate login for now.
Related: FAQ #4
6. Who gets access, and what permissions do they have?
Answer: Access is set up per customer (a dedicated tenant). Initial users are typically a 3–5 person task force (commercial leads + operations). All users are given admin access, meaning they can create and edit validation rules - not just run assessments.
7. I'm a brand-new customer - what should I do first?
Answer: Before running assessments, build your validation library. The system needs validations to produce meaningful results. We recommend having at least 8–10 validations in place before your first real assessment.
Category 2: Creating an Assessment & Uploading Documents
Section 2.1: Starting a New Assessment
8. How do I start a new risk assessment?
Answer: Create a new assessment, give it a name (naming is mandatory - it keeps your history traceable), then add the deal's basic details.
9. What documents can I upload?
Answer: Four kinds: the base charter party, the proposed recap (as received from the broker), rider clauses (if supplied separately), and any additional files (loose clauses, questionnaires).
10. How do I add the recap?
Answer: Paste the recap in exactly as you received it in the broker's email. The AI compares this recap against the base CP to detect conflicts and changes.
11. Do I need every document type for it to work?
Answer: No - at minimum you need the charter party and recap. Rider clauses and additional files improve coverage when the terms live outside the main form, but they're optional.
Related: FAQ #9
Category 3: Configuring the Analysis
Section 3.1: Perspective, Charter Type & Trade
12. What do I configure before running?
Answer: Three things: your perspective (Charterers or Owners), the charter type (Voyage Charter or Time Charter), and any trade-specific validations that apply (e.g. USA, West Africa, Pacific, Brazil, Period).
13. Why does perspective (Charterer vs Owner) matter?
Answer: The same clause carries different risk depending on which side you're on. Setting the perspective tells the AI whose interests to protect, so the flagged risks and resolution suggestions are written for your side of the deal.
Related: FAQ #12, #58
14. What's the difference between Voyage Charter and Time Charter here?
Answer: It selects which validation set and logic apply. Voyage Charter forms (e.g. GENCON, ASBATANKVOY) and Time Charter forms (e.g. NYPE) carry different risks, so choose the type that matches the deal.
Related: FAQ #12
15. What are trade-specific validations?
Answer: Extra checks tied to a trade lane or contract type (USA, West Africa, Pacific, Brazil, Period, etc.). Add the ones relevant to the voyage so the analysis includes region- or period-specific risks.
Category 4: Understanding the Results
Section 4.1: The Risk Matrix
16. How long does an analysis take?
Answer: Typically 2–5 minutes. You'll get a notification when it's finished.
Related: FAQ #17
17. How are results presented?
Answer: As a colour-coded risk matrix. Each finding is a card, grouped by priority category.
18. What do the risk categories mean?
Answer: Deal Breakers - must be addressed before fixing.
- Requires Approval - escalate to management before proceeding.
- Trader Discretion - negotiate if you can, but it's your call.
- Conflicts - inconsistencies the AI detected *between* documents (e.g. recap vs base CP).
19. What information is on each finding card?
Answer: A title and summary, the exact clause reference in the CP or recap, and a resolution suggestion (the recommended action).
20. What's a "Conflict" versus a "risk"?
Answer: A risk is a problematic term flagged by a validation. A Conflict is specifically a mismatch *between* your documents - for example, the recap says one thing and the base CP says another. Conflicts are surfaced separately so you can reconcile the paperwork.
Related: FAQ #18
21. What order are results shown in?
Answer: Findings are grouped by category. (Note: ordering by document position - recap → body → riders - is a known request that isn't fully in place yet; see Troubleshooting.)
Related: FAQ #18, #52
Category 5: Working Through Findings & Building the Counter
Section 5.1: Actioning Cards
22. What can I do with each finding?
Answer: Three actions: Flag for negotiation (add it to your counter), Accept risk (acknowledge it but don't negotiate - e.g. an obsolete COVID-19 clause), or Dismiss (not relevant).
23. What is "Build Counter" / the negotiation tab?
Answer: It's a shortlist of every item you flagged for negotiation. It gives you a single view of your negotiating points so you can turn them into a counter-offer.
24. How do I turn flagged items into an email to the broker?
Answer: Open the negotiation tab, copy the flagged points into your email, add anything you spotted in manual review, and send your counter. (A built-in counter-draft feature is on the roadmap; today it's copy-into-email.)
Related: FAQ #23, #52
25. Can I add points the AI didn't flag?
Answer: Yes. The counter is yours to edit - add anything you caught manually before sending.
Related: FAQ #24
Category 6: Validations
Section 6.1: What Validations Are
26. What are validations?
Answer: Validations are the checks the AI runs your documents against - your library of "things to flag and why." They're what turns a generic read into a company-specific risk assessment.
27. Who creates the validations - us or the customer?
Answer: The customer. All users have admin access and can create and edit their own validation rules. They're specific to each customer's business.
28. How is a validation structured?
Answer: Each validation captures what to check, why it matters, its severity (which category it lands in), and when it applies (perspective, charter type, trade area). Clear, AI-friendly wording matters - vague validations produce vague flags.
29. Can we help a customer write their validations?
Answer: Yes. If a customer sends a high-level list of topics they want checked (or an Excel file of them), we can help structure them into AI-optimised wording and upload them. This is a normal part of onboarding.
30. Why should customers wait until they have 8–10 validations before running?
Answer: Without a meaningful validation set, results are thin and not representative of the tool's value. Building the library first is the single biggest driver of a good first experience.
Section 6.2: Validation Governance
31. Can everyone change the shared validations?
Answer: Technically all users are admins, but the recommended model (proven with our anchor customer) is a single admin who owns universal validation changes, with the rest of the team requesting changes through them. This keeps the library consistent.
32. Is there a record of validation changes?
Answer: Yes - validation changes are tracked via a change log, and changes require a double-confirmation step to prevent accidental edits. (An expanded audit trail for validation edits is on the roadmap.)
33. Should one customer's validations ever be shared with another?
Answer: Never. Validations are customer-specific and often reflect confidential commercial policy. They must not cross tenants.
Related: FAQ #27
Category 7: Feedback, Audit Trail & Accounts
34. How do users report a bug or request a feature?
Answer: There's an in-app feedback module for bugs, feature requests, and UX issues. Feedback is captured with traceability so it can be actioned.
Related: FAQ #35, #53
35. What happens to feedback once submitted?
Answer: It flows to the product team. For the anchor customer this runs as a weekly feedback loop. Set expectations accordingly - it's a route into the roadmap, not instant resolution.
Related: FAQ #34
36. Can I export a record of an assessment?
Answer: Yes - audit-trail export is live. It gives a trail of what was checked and what was decided, which is useful for compliance and internal review.
Related: FAQ #37
37. Why does auditability matter for this product?
Answer: Chartering decisions need a defensible record of which risks were raised and how each was handled. The audit trail + validation change log together provide that.
38. How do we add more users for an existing customer?
Answer: The customer sends the list of email addresses; we create accounts and issue credentials. New users also get admin access by default.
Related: FAQ #6
39. How do we set up a brand-new customer (tenant)?
Answer: Internally: a dedicated tenant environment is created, then user accounts are provisioned against the customer's email addresses, and credentials are sent with the getting-started guidance (build validations first).
Category 8: Troubleshooting & Known Issues
Section 8.1: Common Questions
40. Why am I not seeing useful results after logging in?
Answer: This almost always means you haven't added any validations yet. CP Analyzer needs a validation library to produce meaningful results - we recommend building at least 8–10 validations before your first assessment. See the Validations section of this guide to get started.
41. Why are the suggested resolutions written for the wrong side of the deal?
Answer: Check the perspective setting (Charterer or Owner) on your assessment. It tells the AI whose interests to protect, so if it's set incorrectly the resolutions will be written for the other side. If the perspective is already correct and results still look off, let us know through the in-app feedback option so we can take a look.
42. Why did I get a different number of results running the same documents twice?
Answer: Result consistency between repeat runs is an area we're actively improving. If you notice a meaningful difference between two runs of the same documents, please flag it via the in-app feedback option with the assessment name so we can look into it.
Related: FAQ #43, #53
43. What happens if I refresh the page or switch tabs while reviewing an assessment?
Answer: Right now, refreshing the page or switching tabs mid-review can cause you to lose your place. We'd recommend avoiding this while reviewing an assessment - we're working on improving this. If it happens to you, let us know via the feedback option.
44. Does CP Analyzer support LOI (Letter of Indemnity) analysis?
Answer: Not yet - LOI analysis isn't available today, but it's a high-priority item on our roadmap. Let us know if this would be valuable for you; it helps us prioritise.
Related: FAQ #45
45. Is there a built-in clause library or preferred-clause lookup?
Answer: Not yet. Today you'll need to look up preferred clauses outside the tool, but an integrated clause library is on our roadmap.
Related: FAQ #44
46. Can I control the order results are shown in?
Answer: Not yet - findings are currently grouped by risk category rather than by document position (recap, body, riders). Ordering by document position is a common request we're looking at for a future release.
Related: FAQ #21
47. How do I get help if I run into an issue I can’t resolve?
Answer: Use the in-app feedback option to report bugs, request features, or flag anything that isn't working as expected - it's tracked so our team can follow up. For anything related to your account or access, reach out to your usual support contact.
Comments
0 comments
Please sign in to leave a comment.