The capstone asks you to produce a full end-to-end EDRM protocol for a matter of your choosing. Every module in this course connects. Every deliverable in the kit is available to you. The point is not to demonstrate that you memorized the modules — it's to demonstrate that you can think through a matter start to finish and produce something you'd stake your name on. Praise is nice. Criticism is useful. Weird ideas are often where the gold is.
What you'll produce
An end-to-end eDiscovery protocol package for one matter — real or synthetic — containing at minimum:
- Matter scoping memo (2 pages) — parties, dispute, likely custodians, likely sources, timeline, budget envelope.
- Collection protocol (using DLC-M01-002).
- Custodian interview script (using DLC-M01-003), tailored to your matter.
- Processing and loading spec (using DLC-M03-001, DLC-M03-002).
- Review protocol — including TAR / GenAI approach if applicable (using DLC-M04-003).
- Analytics plan (using DLC-M05-001).
- Production spec (using DLC-M06-001).
- Privilege log plan — jurisdictional considerations, threading approach (using DLC-M07-001).
- AI-use policy — either disclosure clause or explicit non-use (using DLC-M08-002).
- Capstone essay (~1,500 words) — walking through your decisions and the trade-offs. This is where you show your reasoning.
Matter selection
You may choose:
- Option A · Real matter: A matter you've worked on or observed, anonymized. Do not use identifying details of any real client.
- Option B · Faculty scenario: A synthetic matter provided in class (choose from the three scenarios in the cohort portal).
- Option C · Custom scenario: Design your own matter. Must include at least: 3+ custodians, 2+ source types beyond email, at least one preservation complication, at least one AI-adjacent question.
Section 1 · Matter Scoping
Two pages max. If you can't scope a matter in two pages, you don't understand it yet.
1.1 · Matter Selection
Chosen option:
Matter title (anonymized):
1.2 · Dispute Summary
In 2–3 sentences, what is the dispute about?
1.3 · Parties & Roles
Who is producing? Who is receiving? Any joint-defense arrangements or common-interest privileges?
1.4 · Likely Custodians
Name (or role) — expected data volume — special considerations:
1.5 · Likely Sources
Source types in scope (by name and volume estimate):
1.6 · Timeline & Budget Envelope
Meet-and-confer date; production deadlines; total budget assumption:
1.7 · Preservation Complications
Auto-deleting messaging? Departing employees? Ephemeral platforms? Non-US data?
Sections 2–9 · Attach Your Deliverables
Complete each kit template below for your capstone matter, then attach them to your submission. Check off each as you attach.
Attachment Manifest
Please describe each attached file (filename, section, page count):
Section 10 · Capstone Essay
~1,500 words. This is where the assessment lives. Explain your reasoning; show your work; call out trade-offs you accepted; identify what you would do differently next time.
Structure
- What kind of matter is this, and why did I make the workflow choices I did? (~400 words)
- Which decisions were hardest and how did I resolve them? (~400 words) — TAR vs. GenAI, threading approach, redaction strategy, vendor selection, etc.
- What are the defensibility risks in my plan and how have I mitigated them? (~300 words) — With reference to case law from DLC-M08-001.
- If I had 20% more budget or 20% less, what would change? (~200 words) — Shows you understand what's marginal.
- What did I learn from this course that changed how I would approach this matter versus how I would have approached it before? (~200 words)
Essay text
Section 11 · Rubric & Self-Assessment
Score yourself before submitting. Faculty grade against the same rubric.
| Dimension | Proficient (3) looks like | Self | Faculty |
|---|---|---|---|
| Matter scoping | Section 1 is complete, specific, and defensible. Custodian and source estimates are grounded. Preservation complications identified. | ||
| Workflow integration | Every kit template is filled in consistently — the UID scheme in §2 is honored through §7 production. The threading choice in §8 is compatible with the review approach in §5. | ||
| Analytical rigor | Cost estimates use published rates. TAR / analytics decisions are supported by DLC-M05-001 break-even math. Sample sizes computed correctly. | ||
| Defensibility | Case-law citations are appropriate. Preservation, TAR, privilege, and AI positions could withstand a motion. Redaction workflow includes post-redaction OCR. | ||
| Essay quality | Essay demonstrates reasoning, not just choices. Trade-offs surfaced honestly. Delphoria voice throughout. Under word count is fine; over is not. | ||
| Delphoria posture | Advisor stance throughout. No language implying Delphoria performs collections, processing, or hosting. "Delphoria Platform" (never Splunk). Vendor evaluations even-handed. |
Self-total (of 18): Faculty-total (of 18):
Contribution to course grade: Capstone counts for 20% of your course total. A score of 15/18 corresponds to Proficient across the board and comfortably meets the certificate threshold.