---
title: "Update network settings"
method: POST
path: "/forecaster/network-settings"
tags: ["Yield Forecast"]
---

# Update network settings

`POST /forecaster/network-settings`

Update the Yield Forecast settings of a network.

## Request body

- object
  - `sampling` integer — The level of sampling to use (between 0 and 3). 3 provides the quickest response and 0 the highest accuracy.
  - `dateRange` integer — The default number of days for which forecasts will be calculated and results shown, starting from the current day.
  - `dateAggregation` 'date' | 'week' | 'month' — The default date aggregation in the `groupBy` of the inventory results forecasts.
  - `groupBy` string[] — Break the forecast results down by the given `GroupBy` keys. Some caution should be used when choosing the `GroupBy` keys, as some keys may have thousands of possible values and can produce very large responses. For more details and the list of [available keys](https://dev.kevel.com/reference/kevel-forecast-overview#group-by-fields), see this [section](https://dev.kevel.com/reference/kevel-forecast-overview#grouping-by-and-time-zone-considerations). When grouping by fields that are arrays, the default behavior is to group by the whole set. Alternatively, it is possible to expand the elements of the array and group by each of them by suffixing the field name with `[]`, or by using the singular of the field (if it is one of the known array fields). It's also possible to expand the key-value pairs of `$user.custom` by suffixing the field name with `{}`.
  - `filterGroupBy` string, nullable — Filter the forecast grouped results breakdown by the given rule (Zerkel query) at the ad level. This rule operates on the following set of numeric fields for all forecast types: `impressions`, `clicks` and `uniqueUsers`. Additionally, the Available forecast supports the `bookedImpressions`, `bookedClicks` and `bookedUniqueUsers` fields, and the Deliverable and Existing forecasts the `adId`, `flightId`, `campaignId`, `advertiserId`, `reservedAdId`, `reservedFlightId` and `reservedCampaignId` fields. An extra key, `groupedFiltered`, is added to the forecast result containing all the results that were filtered out by the rule. If used in combination with the `ResultsBy` parameter, the `advertiser`, `campaign` and `flight` levels won't include the `uniqueUsers` count.
  - `inventoryPriorities` integer[] — The priority IDs to be considered as the first paid priority of each channel. It should be set one per channel.
  - `salesProbabilityWeight` integer, nullable — The minimum sales probability weight threshold used to filter reserved campaigns. Reserved campaigns with a `salesProbabilityWeight` below this value will be excluded from forecasts. Set to `0` to include all reserved campaigns regardless of their sales probability.

## Response `200`

The JSON object of the network settings, if successful.

- Schema — unresolved $ref

## Other responses

- `400` — The entity in the body is invalid.

---

[API](https://skmtc.net/kevel/apis/engine-apis.md) · [All operations](https://skmtc.net/kevel/apis/engine-apis/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/kevel/engine-apis/revisions/0ca55f45aa7d/schema)