Overlay Behavior sets how Relativity applies incoming values to multi-select and multi-object fields during a load file import. It can merge new values with existing data or replace them, helping preserve the intended record history. That small setting can make a big difference when imports meet existing matter data.

Multiple Choice

Overlay Behavior governs what aspect of loading data?

Overlay Behavior is about how data imports handle fields that can hold multiple values. When you load a file, fields like multi-select choices or multi-object links can already contain data, so the loader needs a rule for how to apply the incoming values. Overlay Behavior determines whether the new values are merged with what’s already there (adding to the existing selections), or whether they replace the existing values. This ensures imported data for those multi-valued fields is integrated in a controlled way rather than unintentionally altering or discarding current content. The other options relate to rendering HTML, storing long text in a grid, or viewer interactions with extracted text, which are not about how loading handles multi-valued fields.

In Relativity e-discovery, an import can do more than add a document or update a text field. Some fields hold several values at once: a document might have multiple issue tags, or link to several people, companies, or other records. When new data arrives for one of these fields, Relativity needs to know what to do with the values already there. Overlay Behavior controls that decision during load file import.

The rule matters because “update this record” can mean two different things. The incoming values might be added to the values already stored, or they might replace them. Those actions can produce very different results, especially in a large matter where reviewers or earlier imports have already populated the field.

Why multi-value fields need special handling

Many database fields hold just one value. A document may have a single date, a single status, or one short description. If an import brings in a new value, the system can update that field in a fairly direct way.

Multi-value fields are different. A multiple choice field can contain several selections, such as “Privileged,” “Needs Review,” and “Key Document.” A multiple object field can connect one record to more than one related object. For example, an item could be associated with several custodians or organizations.

That creates a small but important question: if a load file supplies a new value, should Relativity keep the old values too? Overlay Behavior supplies the rule. Depending on the selected behavior, incoming values can be merged with existing values or can replace them.

Think of a multi-value field like a set of labels on a folder. If a new label arrives, you might stick it beside the old ones. Or you might remove the old labels and use only the new set. Neither approach is always right; the right choice depends on what the import is meant to do.

Merge or replace: what changes?

When the incoming values are merged, the new values are added to the existing set. Existing selections remain, unless the import process or field configuration calls for a different result. This can be useful when a load file contains additional tags or links and the goal is to preserve information already in Relativity.

When the incoming values replace existing values, the field reflects the values provided by the new import rather than retaining the previous set. This can be appropriate when the imported data is intended to be the authoritative, complete version of that field.

A simple example makes the difference clear. Imagine a document already has the choices “Contract” and “High Priority.” A new load file supplies “Key Custodian.” With merge behavior, the resulting field may retain the first two choices and add the new one. With replace behavior, the field may contain only “Key Custodian.” The same incoming value can therefore lead to a very different record, depending on the overlay rule.

That’s why Overlay Behavior isn’t just a technical detail tucked inside an import setup. It affects the meaning of the data after the load.

Where this matters in an e-discovery workflow

Overlay Behavior is most relevant when a load file updates records that already exist and includes values for multiple choice or multiple object fields. It helps decide how the import treats those multi-valued fields as it applies incoming data.

This can come up in several situations:

  • A later data delivery adds new issue tags to documents already in the workspace.

  • A revised file links records to additional custodians or related objects.

  • A vendor or processing team sends updated metadata for documents that have already been loaded.

  • A correction file changes the set of values associated with existing records.

The details vary from matter to matter, but the core concern stays the same: will this load preserve existing values, add to them, or replace them? Knowing the intended result before importing can prevent surprises later.

A field setting with real consequences

It’s easy to focus on the more visible parts of a load file—document identifiers, extracted text, dates, or file paths. Yet a setting that governs multi-value fields can have a broad effect across many records. If an import unexpectedly replaces existing selections, reviewers may lose useful categorization. If it merges values when a clean replacement was intended, old labels or links may remain and create confusion.

Neither outcome necessarily means the import failed. The data may have loaded exactly as configured. The issue is that the chosen behavior may not match the purpose of the load.

This is where clear communication between the people preparing data and the people managing the workspace helps. The team should know whether an incoming file is meant to add information, correct a previous set, or provide a complete replacement. That intention should guide the import configuration.

What Overlay Behavior doesn’t control

The name can sound broad, but Overlay Behavior has a specific role in this context: it governs how Relativity handles multiple choice or multiple object fields during load file import.

It does not determine how HTML appears inside a field. Nor does it define how long text is stored or displayed in a data grid. It also isn’t the setting that controls how extracted text interacts with the Viewer. Those are separate aspects of the platform, even if they can all appear in the same general e-discovery workflow.

Keeping those concepts separate makes troubleshooting easier. If multi-select values have changed after an import, Overlay Behavior is worth checking. If a field’s HTML looks different, or a long-text value is hard to view, the cause is likely elsewhere.

A few checks before loading data

Before importing a file that updates existing records, it helps to review the fields included in the load and identify any that allow multiple values. Then confirm what the incoming data is supposed to represent.

Is the file adding new tags to an existing set? Does it contain the full, final list of selections? Is it correcting links between records? These questions point toward whether merging or replacing makes sense.

It can also help to examine a small sample of records before relying on the same configuration across a larger load. Compare the values already present in Relativity with those in the incoming file, and verify that the resulting values match the intended outcome. A quick review can reveal an assumption that would otherwise affect many records at once.

Finally, keep a record of the import settings used. In a matter with multiple data deliveries, notes about how multi-value fields were handled can save time and reduce uncertainty when someone later asks why a set of tags or links changed.

The takeaway

Overlay Behavior governs how Relativity handles multiple choice and multiple object fields when data is imported through a load file. It determines whether incoming values are merged with existing values or replace them.

That may sound like a narrow setting, but it shapes how information is preserved across updates. Treat it as part of the data’s meaning, not merely a checkbox in an import workflow. When the intended result is clear and the setting matches that intent, multi-valued metadata can be updated in a controlled, predictable way.