This case study demonstrates the material cost, time, and risk reduction achieved by avoiding a “direct-to-RSMF” workflow for chat data and instead applying search, filtering, and contextual expansion prior to RSMF formulation using StreemView.
By operating on native chat messages first—rather than prematurely committing data into static 24-hour RSMFs—the legal team produced a dramatically smaller, more relevance-dense review population. The outcome was a 97% reduction in review volume, a 96% reduction in review cost, and substantial downstream efficiencies that would not have been achievable in a traditional workflow.
Background and Challenge
Modern chat data (Slack, Teams, text messages, and similar platforms) is fundamentally conversational—not document-centric. Traditional workflows that promote all collected chat data directly into 24-hour RSMFs impose several structural problems:
-
- Entire 24-hour conversations become indivisible “documents,” with the potential to contain thousands of non-responsive messages in a single document for large channels.
-
- Search terms applied after RSMF creation over-capture content because hits anywhere in the day pull the full days wroth of conversation into review.
-
- Review teams are forced to perform extensive redaction, splicing, and privilege work on bloated records that were never relevant in the first place.
-
- Hosting and review costs scale with volume, not relevance.
In this matter, the client faced a chat population large enough that a direct-to-RSMF approach would have created a review set measured in many millions of messages, most of which were non-responsive.
Data Profile
-
- Messages processed into StreemView: 39,275,846
-
- Hosted size in StreemView: 1,923 GB
-
- Negotiated search terms: Highly targeted, complex Boolean criteria
Rather than committing this population to RSMF up front, the team used StreemView to interrogate the data in its native conversational form.
StreemView-First Workflow
1. Search Before Structure
All negotiated search terms were applied before any RSMF files were created. This ensured that relevance decisions were made at the message level, not at the 24-hour document level.
-
- Direct search hits identified: 97,695 messages
2. Contextual Expansion (±5 Messages)
Recognizing that individual messages lack meaning in isolation, StreemView was used to automatically expand each hit to include five messages before and after the hit.
-
- This preserved conversational context
-
- Eliminated the need for manual splicing during review
-
- Maintained defensibility while avoiding over-capture
3. Selective RSMF Export
Only messages deemed relevant plus their contextual window were promoted into RSMF for downstream review.
-
- Final export set:
-
- 1,155,283 messages
-
- 57 GB total
-
- Final export set:
What a Direct-to-RSMF Workflow Would Have Produced
Analytical modeling showed that if the same data had been promoted directly into 24-hour RSMFs before searching:
-
- Messages requiring review: 13,798,199 (Search hits + same-day messages)
-
- Hosted review size: 676 GB
Because every search hit would have pulled in all messages from that calendar day, the review population would have ballooned with irrelevant content.
Comparative Results
Message Volume Reduction
-
- Direct-to-RSMF: 13,798,199 messages
- Reduction: 65%
-
- StreemView-first: 1,155,283 messages
- Reduction: 97%
Review Time and Cost Impact Modeling
Assumptions
-
- Review rate: 600 messages/hour (Estimated)
-
- Review cost: $40/hour (Estimated)
| Approach | Messages Reviewed | Review Hours | Review Cost |
| Direct-to-RSMF | 13,798,199 | 22,997 hrs | $919,879 |
| StreemView-First | 1,155,283 | 1,925 hrs | $77,018 |
Estimated Review cost reduction: $842,861
Hosting Cost Impact (18-Month Lifecycle)
Assumption: $12/GB/month (Estimated)
| Approach | GB Hosted | Total Hosting Cost |
| Direct-to-RSMF | 676 GB | $145,925 |
| StreemView-First | 57 GB | $12,218 |
Estimated Hosting Savings: $133,707
Total Estimated Project Cost Comparison
| Approach | Total Cost |
| Direct-to-RSMF | $1,135,032 |
| StreemView-First | $158,464 |
-
- Cost per message (Direct-to-RSMF): $0.03
-
- Cost per message (StreemView-First): $0.004
Why This Matters
This case study illustrates a core truth of modern eDiscovery: Once you commit chat data to RSMF, you inherit all of its inefficiencies. By delaying RSMF creation until after relevance decisions are made, legal teams:
-
- Eliminate massive volumes of non-responsive content
-
- Reduce redaction and splicing complexity
-
- Accelerate review timelines
-
- Lower review workspace hosting and review costs by orders of magnitude
-
- Improve reviewer accuracy by delivering relevance-dense records
These gains compound further when accounting for additional StreemView capabilities not included in the cost model, such as sophisticated de-duplication, over-the-wire attachment retrieval, and flexible RSMF export options.
Conclusion
This matter demonstrates that RSMF should be an output—not a starting point. Applying search, filtering, and contextual logic upstream transforms chat review from a blunt, document-centric exercise into a precise, conversation-aware workflow.
For modern data, purpose-built solutions are not a luxury. They are the difference between operational drag and strategic advantage.
