End-to-End eDiscovery for In-House Legal Teams: From Legal Hold to Production

In-house legal teams are taking on more discovery work just as the data behind it becomes harder to manage. The Thomson Reuters Institute's 2025 State of the Corporate Law Department report found that general counsel are three times as concerned about delivering value as they were the year before, with cost control a higher priority than ever.

Collection is where that pressure shows up first. Onna's 2026 Collaboration Data Risk & Readiness Report, a survey of 241 senior legal and IT leaders, found that organizations run an average of 10 collaboration platforms and hold 48TB of collaboration data per matter. Teams spend 26 hours per matter coordinating collection alone, and only 2% of collections are sized correctly.

Handoffs Are Where Corporate eDiscovery Breaks Down

Most corporate eDiscovery programs were assembled one tool at a time. A hold is issued in one system, a collection request goes to IT as a ticket, data is exported to a vendor for processing, and review happens somewhere else. Each step can work fine by itself. Problems start in the gaps between them.

The same report shows how those gaps play out in practice:

• Scope drifts. In 39% of matters, teams discover new data sources partway through collection and have to restart.

• Collections miss the mark. Incomplete scope forces recollection in 36% of cases, while 39% over-collect because of uncertainty or limited tooling.

• IT becomes the bottleneck. Nearly 70% of organizations rate their dependence on IT at 7 out of 10 or higher.

• Costs climb. Four in five organizations say collaboration data causes cost overruns.

The common cause is structural: no single system knows who is on hold, what has been preserved, and what has already been collected.

Five Stages Take a Matter From Hold to Production

Every matter moves through the same five stages. Programs differ in how many of those stages still depend on manual handoffs.

1. Issue the Legal Hold

The duty to preserve begins when litigation is reasonably anticipated. The hold names custodians, sets scope, and tracks acknowledgments. This comprehensive guide to legal holds covers notice content, custodian follow-up, and release.

2. Preserve Data in Place

Notices alone do not stop automated deletion. Preserving data where it lives, inside each collaboration and cloud source, suspends retention for in-scope content without copying it to new storage. When the hold system is connected to the collection platform, issuing a hold can trigger that preservation directly. Teams increasingly need to preserve AI-generated content in collaboration platforms as well, since summaries, drafts, and meeting notes often land outside the assistant that created them.

3. Collect With Targeted Scope

When preserved data is indexed and searchable, legal teams can filter by custodian, date range, channel, and keyword before collecting. Teams collect what they need the first time, which cuts down on both recollection and over-collection.

4. Review the Culled Set

The culled set should move into review without manual exports or metadata mapping. Processing steps such as deduplication, email threading, family grouping, and PII detection can run as the data arrives, so reviewers start with organized documents and their context intact, including the thread around a message and the prompt behind an AI response.

5. Produce With a Full Chain of Custody

Production draws on the same record that began with the hold. Every action, from notice to export, is documented in one chain of custody that can support a defensibility challenge. When review and production run on a tool the in-house team can operate, routine matters no longer need to go to outside vendors at outside rates.

An eDiscovery Platform for Corporations Needs Five Capabilities

When legal operations, IT, and information governance leaders evaluate eDiscovery for corporations, these capabilities determine whether the workflow holds together:

• Direct connections to data sources. Prebuilt connectors to chat, email, file storage, video, and AI assistants remove the need for custom exports and IT tickets for each matter.

• Hold linked to preservation. The custodian list in the hold should map directly to what is preserved, so no one is on hold without their data being protected.

• Coverage for AI-generated content. Prompts, responses, and AI outputs saved in other tools now fall within scope. This overview of legal holds for AI-generated content explains why these sources need the same treatment as email.

• Review the in-house team can run. Routine matters stay in-house when review and production do not require a specialist or an outside vendor.

• One chain of custody. A single record of every action from hold through production is easier to defend than logs spread across several systems.

This is the model behind the Corporate Suite. Reveal Hold issues holds and preserves data in place, Onna collects and culls it across collaboration sources, and Logikcull takes the culled set through review and production, all under one chain of custody. Teams that want to see the workflow on their own data sources can request a demo.

A Connected Workflow Turns Discovery Into a Repeatable Process

In-house eDiscovery becomes predictable when the hold, preserved data, collection, review, and production share one record. Fewer handoffs mean fewer restarts, less over-collection, and a clearer answer when opposing counsel or a regulator asks how data was handled. For legal departments that have to show value, this matters as much as speed: discovery stops being rebuilt for each matter and runs the same way every time.

To map your current workflow from hold to production and identify where handoffs add time and cost, contact the Onna team.