Why AL Designation Not Set Causes Documents to Be Skipped in Active Learning

Outline (skeleton)

• Grab attention: in e-discovery, Active Learning helps reviewers focus on what matters, but some docs still slip through.

• Quick refresher: what Active Learning does and how search conditions guide what gets flagged or skipped.

• The core question in plain terms: which condition, when paired with AND, signals documents that reviewers skipped?

• The correct answer and why: AL Designation is Not Set. This shows items that haven’t received a designation yet.

• Why the other options don’t indicate skipped reviews.

• Practical takeaways: how to set up filters in Relativity to surface gaps and improve coverage.

• Quick tips and common misreads.

• Wrap-up: the big idea in one clean takeaway.

Article: A clear lens on skipped documents during Active Learning

If you’ve ever poked around Relativity and watched an Active Learning workflow in action, you know one thing for sure: the system loves to separate the signal from the noise. It flags the documents it thinks reviewers should focus on, and it quietly marks others as—well—potentially less important. But there’s a twist that can trip people up: some documents aren’t flagged or categorized at all. They sit in a kind of neutral limbo, waiting for a decision.

Let me explain what Active Learning does in this space. Think of it as a smart assistant that helps reviewers home in on the most relevant material. It uses feedback from humans to fine-tune what the system expects to see. That means the right search conditions become the gatekeepers. If you want to know which documents bypass the usual review path, you’re essentially looking for a signal that the document has not yet received a designation—or any human labeling at all.

Here’s the thing about the specific question you’re weighing: which search condition, when used with the AND operator, will pull up documents skipped by reviewers during Active Learning? The correct answer is AL Designation is Not Set. Why does this reveal skipped material? Because “Not Set” flags items that haven’t been assigned a designation yet. No reviewer said “this is responsive” or “this is irrelevant.” It’s a clean, neutral state. When you combine that with another condition via AND, you’re effectively asking the system to return only the docs that meet both criteria: they fall into a particular category (as defined by the second condition) and they haven’t been designated at all. In other words, you’re shining a light on the gaps—documents that slipped through the cracks, so to speak.

Now, what about the other options? Let’s walk through them with a practical mindset.

• Option B: Active Learning Project Reviewers > These Conditions > Active Learning Project Reviewers is Not Set. On the surface, this sounds like it would surface docs where reviewer assignments aren’t in place, but the logic isn’t as direct to “skipped” items. It turns more into a condition about who has access or who is assigned to review, rather than a direct indicator of whether a doc has been reviewed or designated. In short, it’s a different flavor of missing metadata, not the core signal that the document itself hasn’t been labeled yet.

• Option C: Active Learning Project Reviewers > These Conditions > Active Learning Project Reviewers is Set. This one indicates that a document has already been touched by a reviewer—designated and reviewed at least once. It’s the opposite of what you want if you’re hunting for items that were skipped. It’s a sign of completion, not omission.

• Option D: Active Learning Completed is Set. When this is true, the Active Learning cycle for that doc has finished. That means the document got processed through the system, labeled, or otherwise acted on. It’s again the opposite of “skipped.” If you want to catch items that didn’t get a designation, you’re better off looking for the absence of a designation, not the presence of a completed status.

So, in the real world, the right pairing to reveal unreviewed documents is AL Designation is Not Set, especially when you’re trying to map out gaps in coverage. It’s a straightforward signal: no label, no decision, no final tag yet. This is exactly the kind of flag that project managers and review teams use to ensure there aren’t blind spots in their dataset.

Turning this into practical practice on your end

If you’re using Relativity in projects that employ Active Learning, here are a few grounded tips to keep the workflow smooth and transparent:

• Build a neutral-state filter. Create a saved search or a quick filter that includes AL Designation is Not Set. Then, to tighten the net, add a second condition that ties to the dataset’s focus (for example, a keyword or a tag that you know should have been reviewed). The AND connection is your friend here because it filters for both criteria at once.

• Don’t confuse missing with completed. It’s easy to slip up and think “not set” means not relevant, but it often signals a real review gap. Treat it as a to-do item for coverage checks.

• Run periodic gap analyses. Set up a lightweight cadence to re-run the Not Set filter after initial passes. If the same docs keep showing up, there’s a structural issue in the workflow—perhaps a department’s review queue isn’t pulling those items in.

• Pair with a completeness audit. Alongside your Not Set signal, include a count check: how many docs exist in total, how many have a designation, and how many remain unassigned? This gives you a quick health check on the review process.

• Use it as a conversation starter. If you’re coordinating a team, bring the Not Set results to a meeting. It’s a concrete way to discuss where coverage is thin and what manual review might need to happen to close the loop.

Let’s connect this to a broader picture. Active Learning is designed to optimize human attention, but it works best when you treat your filters like a map, not a rigid rulebook. A well-chosen filter helps you quickly spot gaps, not just confirm what you already suspect. And in a project of any size, those gaps can hide critical documents that affect outcomes—from risk assessments to regulatory disclosures.

A few gentle reminders to help keep your workflow sane

• Keep the language consistent. In your searches, use the exact field names and labeling conventions your team uses. If “AL Designation” shows up with a slightly different name in a field you’ve customized, you’ll get inconsistent results. A little housekeeping goes a long way.

• Save common views. If you find yourself frequently needing the Not Set filter plus a second criterion, save that as a view. This saves time and reduces the chance of human error during fast-paced review cycles.

• Balance precision with speed. Not every gap needs a full drill-down. Start with a broad Not Set screen, then drill into the subcategories to identify where exact gaps exist.

• Document your reasoning. A short note about why a Not Set result matters helps teammates understand the logic, especially when new folks join the project.

A quick mental model you can carry forward

Consider Active Learning as a relay race. The baton moves from the system to the reviewer and back again, iterating to sharpen the answers. The Not Set signal is like a clean baton drop—indicating a handoff didn’t occur yet. When you catch those drops, you can reallocate attention and ensure nothing important stalls in limbo.

Toward a more intuitive understanding

If you’re listening for a simple rule of thumb, here it is: when you want to identify documents that have slipped through review, look for the absence of a designation. Combine that with a relevant attribute or category using AND, and you’ve got a precise signal that highlights gaps. That’s the essence of the approach behind AL Designation is Not Set.

A few practical prompts to keep in mind

• Always verify field names before building queries. A small mismatch stops the filter from returning any results.

• Use concise second conditions. Pair Not Set with a realistic attribute (like a document type, a date window, or a keyword tag) to narrow the pool without losing sight of the bigger picture.

• Treat Not Set as a flag, not a verdict. It tells you there’s more work to do, not that a document is irrelevant by default.

• Periodic reviews beat one-off looks. Re-check Not Set results after a queue refresh or after a workflow adjustment. You’ll spot trends you’d miss otherwise.

A closing thought

In the end, the value of this approach isn’t just about catching skipped documents. It’s about building a steadier, more transparent review process. It’s about making sure the system’s intelligence and human judgment work in harmony. The Not Set signal is a small lever, but pull it thoughtfully, and you illuminate the edges where things might have slipped through the cracks. And when you can see those edges clearly, you’re in a better position to guide the team toward a more complete, well-documented outcome.

If you’re implementing this in your own environment, keep the focus on clarity over complexity. A simple, well-tuned filter with clear naming and routine checks often yields the most reliable improvement. After all, in a fast-moving review project, a dependable signal about what still needs labeling can save days of effort—and that’s a win worth aiming for.