> For the complete documentation index, see [llms.txt](https://docs.mediarithmics.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.mediarithmics.io/audience-and-activation/build-your-audiences/expand-your-audience-with-lookalike.md).

# Expand your audience with Lookalike

Grow an audience that is too small to be worth activating, using your own first-party data.

## Your goal

Reach more users who resemble an audience you already have, without buying third-party data.

## Expected outcome

A lookalike segment, calibrated on a minimum similarity index you chose, refreshed daily, and sized in the region you need for activation.

## When this is useful

You have identified the right audience and there is not enough of it: a high-value customer segment too small to buy media against, a converter segment you want to prospect against, a subscriber base you want to grow. A lookalike raises reach using only your first-party data.

## Prerequisites

* **A seed segment that is persisted and is not itself a lookalike.** Any other persisted segment can be a seed.
* **The feature enabled.** Cohort-based Lookalike is not on by default, and the cohort model has to be configured. Contact your Account manager.
* **A stable seed.** Wait for the seed segment's calculation to finish, and check its volume has been stable over the last few days before expanding it.

## Concepts you need first

| Concept          | Where it is defined                                                                                                                                    |
| ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------ |
| Cohort           | [What a cohort is](/audience-and-activation/build-your-audiences/segment-reference/segment-types/cohort-lookalike-segment.md#what-a-cohort-is)         |
| Similarity index | [The similarity index](/audience-and-activation/build-your-audiences/segment-reference/segment-types/cohort-lookalike-segment.md#the-similarity-index) |
| Overlap          | [Overlap](/audience-and-activation/build-your-audiences/segment-reference/segment-capabilities/overlap.md)                                             |

## Choose the variant

| Variant          | Use it when                                                                                                                            |
| ---------------- | -------------------------------------------------------------------------------------------------------------------------------------- |
| **Cohort-based** | You want the platform to do the modelling, from the cohorts assigned to every user in your datamart.                                   |
| **Score-based**  | You have your own algorithm - a score, a gender or socio-demographic model - that you want to integrate. Contact your Account manager. |

## Steps

{% stepper %}
{% step %}

### Sanity-check the seed

Confirm the seed is persisted and stable, and look at its [overlap](/audience-and-activation/build-your-audiences/segment-reference/segment-capabilities/overlap.md) with your other segments. If most of the seed is already inside a much larger segment, expansion may be redundant.
{% endstep %}

{% step %}

### Launch the lookalike

Open the seed segment, click **Create Lookalike**, select **Cohort based Lookalike**, and name it.
{% endstep %}

{% step %}

### Calibrate the trade-off

In the **Lookalike calibration** panel, move the slider to set the **minimum similarity index**, and choose whether to **include** or **exclude** the seed segment's own users. The expected user count updates in real time. Click **Save those settings**.

Higher index, closer resemblance, smaller audience. Lower index, larger audience, looser resemblance. The index is a floor on quality: the platform never goes below it to hit a size.
{% endstep %}

{% step %}

### Wait for the computation

Like any other segment, the lookalike is populated at the next segment computation, and refreshed daily afterwards.
{% endstep %}
{% endstepper %}

## Verification

Compare the final count against the expectation shown during calibration. A shortfall is normal and is documented behaviour, not a failure - see [What happens at each update](/audience-and-activation/build-your-audiences/segment-reference/segment-types/cohort-lookalike-segment.md#what-happens-at-each-update).

Two things make the final number move:

* The seed segment itself varies between updates.
* Cohorts are calculated live, so a cohort's population changes between the segment update and the segment calculation.

## Troubleshooting

<table data-search="false"><thead><tr><th width="300">Symptom</th><th>What it means</th></tr></thead><tbody><tr><td>The lookalike is much smaller than the expected size</td><td>The target could not be reached without dropping below your minimum similarity index. Precision was preserved instead. Lower the index if you need the volume more than the resemblance.</td></tr><tr><td>The lookalike contains the seed users and you did not want that</td><td>The seed was set to <strong>include</strong>. Recalibrate with <strong>exclude</strong>.</td></tr><tr><td>Consent cannot be enforced on the lookalike</td><td>Expected. Lookalike is one of the three types that cannot carry a processing activity. See <a href="/audience-and-activation/build-your-audiences/enforce-compliance-on-your-audience.md">Enforce compliance on your audience</a>.</td></tr><tr><td>The lookalike volume swings day to day</td><td>The seed is unstable, or cohorts are moving. Stabilise the seed first.</td></tr></tbody></table>

## Related pages

<table data-view="cards" data-search="false"><thead><tr><th></th><th></th><th data-hidden data-card-target data-type="content-ref"></th></tr></thead><tbody><tr><td><strong>Cohort lookalike segment</strong></td><td>The type, the cohort model, and the index arithmetic.</td><td><a href="/audience-and-activation/build-your-audiences/segment-reference/segment-types/cohort-lookalike-segment.md">Cohort lookalike segment</a></td></tr><tr><td><strong>Overlap segments</strong></td><td>Checking whether the expanded audience is actually new.</td><td><a href="/audience-and-activation/analyze-your-data/overlap-segments.md">Overlap segments</a></td></tr></tbody></table>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.mediarithmics.io/audience-and-activation/build-your-audiences/expand-your-audience-with-lookalike.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.