# Creating a population

How to fill in the population form in the New Research wizard, the field limits, how creation works, what to do if it fails and how to reuse an existing population.

A population is the group of synthetic consumers (personas) who answer your questions in a study. Personas are sampled from Türkiye's population, survey and income data ([How synthetic consumers work](/en/docs/getting-started/how-synthetic-consumers-work)). You can create a population once and reuse it in later studies.

You create populations in the population step of the **New Research** wizard while setting up a new study. The **Populations** library has no create button. You can also ask [Argus](/en/docs/argus/working-with-argus) to create a population for you.

## Before you start

- Decide on the vertical. Each population belongs to a single vertical and cannot be used in another vertical's flow ([Verticals](/en/docs/research/verticals)).
- Prepare your research topic: a concrete phrase that describes the product or category, for example "Premium coffee subscription".
- Check your balance. To submit the form, you need a balance of at least 500 credits in General Research, or at least 50 credits in Fashion and Ad Testing.

> [!NOTE]
> Creating a population does not spend credits; these amounts are only the minimum balance you need. Credits are spent when you start a study with the population, and the cost depends largely on the number of personas ([Credits](/en/docs/account/credits)).

## Opening the form

### General Research

1. In the sidebar, click **New Research**.
2. Select the **General Research** card.
3. On the **Choose Your Research Type** screen, select a method card.
4. On the **Choose Population Mode** screen, click the **Create New Population** card.

### Fashion and Ad Testing

1. In the sidebar, click **New Research**.
2. Select the **Fashion** or **Ad Testing** card.
3. In Fashion, select a method card. Ad Testing has only one method, so this step is skipped.
4. On the **Choose a population** screen, click the **Create a new population** card.

Both paths open the same form (the browser tab reads **Step 3 · Create Population**). In Fashion and Ad Testing, the form has one extra, vertical-specific tab.

## Filling in the form

The name, size and topic fields sit at the top of the form. Below them are the **Population features** tabs on the left and the **Population summary** panel on the right.

1. In **Population Name**, enter a descriptive name (e.g. "Young Urban Professionals"). The field is optional, but unnamed populations are hard to tell apart in lists.
2. In **Population size**, enter the number of personas. The default is 40, or 20 for Real-Time Focus Group.
3. In **Research topic**, enter the product or category.
4. Click **Find segmentation**.
5. Accept the suggested segmentation with **Use segmentation**.
6. If needed, narrow your target audience in the **Demographics**, **Economic**, **Lifestyle**, **Employment** and **Social Media** tabs on the left.
7. Optionally, add a custom feature in the **Custom Feature** tab.
8. Optionally, set the distributions in the **Fashion Attributes** tab in Fashion, or the **Reklam Testi Attributes** tab in Ad Testing.
9. Check your selections in the **Population summary** panel.
10. Click **Create population** once and wait for the next screen to open.

### Finding and accepting a segmentation

Segmentation splits the population into market segments based on your research topic. Each persona is assigned a market segment, and in reports you can see the results for each segment separately ([Segmentation](/en/docs/results/segmentation)).

- The **Find segmentation** button stays disabled until you fill in **Research topic**. While the search runs, the button reads **Searching...**.
- If a match is found, a card with a **Match found** badge appears. If there is more than one suggestion, you can pick another one with **Show other suggestions**; the best match is marked with ★.
- After accepting, you can go back to the suggestions with **Change** or clear the selection with **Remove segmentation**.
- For product groups that have ready-made segment rules, the card reads **Segment rules found (v2)**, lists the segment names and accepts the selection automatically.
- If there is no saved match, the system tries to prepare a new segmentation for your topic, so the search can take a while. If nothing comes back, the message "No matching segmentation found for this product/brand" appears; describe the topic more clearly and search again.

> [!IMPORTANT]
> Until you accept a segmentation, the filter area looks blurred and cannot be clicked, and the **Create population** button stays disabled.

### Filters and ratios

Each attribute card shows its mode in its header: **All** (no restriction), **Custom selection** (only the options you tick) or **Custom ratios** (a percentage split across the options). For attributes such as **Age**, **Gender** and **SES** (socio-economic status), you can give each option a percentage with **Set ratios**. The total must be exactly 100%, and you must confirm the ratios with **Use these ratios**. You can find every attribute and its options on the [Filters and attributes](/en/docs/populations/filters-and-attributes) page.

### Custom features

In the **Custom Feature** tab, you can give a set percentage of personas a free-text feature, for example "Prefers eco-friendly products". Each row has **Feature Description**, **Ratio** and **Smart Distribution**. A persona carries at most one custom feature, so the ratios cannot add up to more than 100% ([Custom features](/en/docs/populations/custom-features)).

### Vertical-specific tab

In Fashion, the **Fashion Attributes** tab asks for a percentage split for **Primary Shopping Channel** and **Style Confidence**; in Ad Testing, the **Reklam Testi Attributes** tab does the same for **Kategori İlgilenimi** (Category Involvement) and **Reklama Açıklık** (Ad receptivity). If you leave a group empty, the default split is used; if you fill it in, it must total exactly 100%.

## Limits and validation

| Field | Rule |
| --- | --- |
| **Population Name** | Optional, up to 255 characters |
| **Population size** | 10–2,001 personas |
| **Population size** (Real-Time Focus Group) | 1–20 personas in General Research, 10–20 in Fashion |
| **Research topic** | Up to 255 characters; required to search for a segmentation |
| Segmentation | The form cannot be submitted until one is accepted |
| **Custom ratios** | Each option 0–100; the total must be exactly 100% and confirmed |
| **Custom Feature** | **Ratio** 0–100; rows total at most 100%; in General Research, the description is limited to 1,000 characters |
| Vertical-specific tab | Each group either stays empty or totals exactly 100% |
| Credit balance | At least 500 credits in General Research, at least 50 in Fashion and Ad Testing |

If you enter a number outside the limits in **Population size**, the value snaps to the nearest limit when you leave the field. While any ratio or distribution breaks these rules, the **Create population** button stays disabled.

If your balance is too low, General Research opens the **Insufficient Credits** dialog, and **Buy Credits** takes you to the credits page. In Fashion and Ad Testing, the message "Creating a population needs at least 50 credits." and your current balance appear below the top fields.

## After you submit

When you submit the form, the study is saved as a draft and the design step (**Design Survey** in the step bar) opens. Personas are created in the background, and you can prepare your design in the meantime. Real-Time Focus Group follows a different flow: in General Research, the **Participant Selection** screen opens instead of the design step, while in Fashion the **Odak Grubu Hazırlanıyor** (Preparing Focus Group) screen is shown until the population is ready ([Real-Time Focus Group](/en/docs/research/focus-group)).

The badge next to the design step's title shows the population's status:

| Status | General Research | Fashion and Ad Testing |
| --- | --- | --- |
| In progress | **Generating Personas...** | **Popülasyon oluşturuluyor** (Creating population) |
| Ready | **Personas Created** | **Popülasyon hazır** (Population ready) |
| Failed | **Generation Failed** | **Popülasyon oluşturulamadı** (Population could not be created) |

Until the population is ready, the **Next Step** button reads **Waiting...** and does nothing. The page updates the status on its own; you do not need to refresh it.

The design page of the Comparison method in General Research has neither this badge nor the **Waiting...** button; even so, you cannot start a comparison until the personas have been created.

The app does not show an estimated time; the more personas there are, the longer creation takes. For Ad Testing populations, an extra step runs after the personas are created to assign viewer archetypes.

General Research populations appear in the **Populations** library only after their personas have been created. Fashion and Ad Testing populations are listed straight away; their detail pages read "Population generating…" until creation finishes.

## If creation fails

If creation fails, the badge on the design step reads **Generation Failed** (General Research) or **Popülasyon oluşturulamadı** (Fashion, Ad Testing). The **Next Step** button stays on **Waiting...**, and you cannot go any further with this study.

In Fashion, the design pages for Price Analysis (WTP), Competitive Analysis (Conjoint), Priority Ranking (MaxDiff) and Feature Price Impact have no failure badge; their badge keeps reading **Popülasyon oluşturuluyor**. In that case, check the population's detail page in the **Populations** library: for Fashion and Ad Testing populations, the error message appears there in a red box. A failed General Research population, on the other hand, never appears in the library.

The app has no retry button, and the system does not retry creation on its own. To create the population again from scratch:

1. In **New Research**, select the same vertical and method.
2. Click the **Create New Population** card (General Research) or the **Create a new population** card (Fashion, Ad Testing).
3. Fill in the form again.
4. Click **Create population**.

Because creating a population does not spend credits, a failed attempt does not affect your balance. If the problem keeps happening, write to the support team from the **Support** page in the sidebar ([Support and troubleshooting](/en/docs/account/support-and-troubleshooting)).

## Using an existing population

If you already have a suitable population, you do not need to create a new one: the new study runs with the same personas and the same number of personas. Because the personas are not generated again, you do not have to wait for creation either.

For the steps, see the "Using a population in a new study" section of the [Population library](/en/docs/populations/population-library) page.

## Tips and common mistakes

- Make the topic concrete. Topics of 3 characters or fewer (e.g. abbreviations) only return a result when there is a near-exact match, and they do not trigger the preparation of a new segmentation.
- An attribute left in **All** mode keeps the Türkiye-wide distribution; every filter you add narrows the population.
- Avoid conflicting filters. For example, if you select age **18-24** together with **Retired**, the system does not stop creation; it relaxes the filter for those personas.
- If you want a mixed distribution, use ratios. Instead of selecting only **Female**, set a ratio of 60% Female and 40% Male, and remember to click **Use these ratios**.
- Very narrow SES selections combined with other filters can result in fewer personas than you asked for.
- Larger populations give more stable percentages, but they raise the credit cost of every study you run with that population.

## Related pages

- [Filters and attributes](/en/docs/populations/filters-and-attributes)
- [Custom features](/en/docs/populations/custom-features)
- [Population library](/en/docs/populations/population-library)
- [Starting a new study](/en/docs/research/new-study)
- [Credits](/en/docs/account/credits)