In addition to paid media there may be other marketing actions taking place that
affect the KPI of interest.

## Organic media variables

Organic media variables are media activities that have no direct cost. These can
include, but are not limited to, impressions from newsletters, a blog post,
social media activity, or email campaigns. Organic media variables have the
option of being modeled with reach and frequency, and have Adstock and Hill
effects, just like paid media variables. The only difference between organic and
paid media is that organic media does not have an associated cost. So, ROI
priors cannot be used with organic media and results pertaining to ROI, such as
response curves and budget optimization, are not available for organic media
variables.

Causal effects and % (percent) contributions are given for organic media
variables and are calculated in the same manner as paid media, as described in
[Incremental outcome
definition](https://developers.google.com/meridian/docs/post-modeling/roi-mroi-response-curves#incremental_outcome).
As with paid media, the incremental outcome of an organic media channel is
defined as the expected difference in the outcome under the observed channel's
media execution compared with the counterfactual scenario of not running that
channel.

## Non-media treatment variables

Non-media treatment variables are marketing activities that are not directly
related to media, such as running a promotion, the price of a product, and a
change in a product's packaging or design. They have no direct marketing cost
associated with them but, unlike organic media variables, they are not media
related and there are no Adstock and Hill effects. They differ from control
variables because they are considered to be intervenable and therefore are
treatment variables under the causal model. As such, incremental outcome and %
(percent) contributions are provided for non-media variables.

Similar to paid and organic media, the incremental outcome of a non-media
variable is defined as the difference in expected outcome between two
counterfactual scenarios. The first scenario is setting the non-media variable
to the observed historical value for each geo and time period. The second
scenario is setting the non-media variable to either the minimum of the
non-media variable (default), the maximum or a user supplied value (see
`non_media_baseline_values` argument in `ModelSpec`). The incremental outcome
is the expected outcome under the first scenario subtracted by the expected
outcome under the counterfactual scenario.

The reason the second counterfactual is not setting the non-media variable to
zero, as is the case with paid and organic media, is that zero is often not an
appropriate counterfactual for non-media variables. For example, if the
non-media variable is price it may make sense to think of the causal effect of
setting the price to its observed value compared with the minimum price the
product sold for but it does not make sense to set the price to zero.

## Deciding if a variable is a non-media treatment variable or a control

The main difference between a non-media variable and a control is that non-media
variables are considered to be intervenable, and therefore are treatments under
the assumed causal model. Controls, otherwise known as confounding variables,
are not intervenable and are assumed to affect both the treatment variables and
the outcome. For more information, see [Causal
Graph](https://developers.google.com/meridian/docs/causal-inference/causal-graph).

As an advertiser, if you can intervene and change a variable's value (such as
changing the price or running a promotion), the variable is more likely a
non-media variable than a control. If the variable is outside the control of the
advertiser, such as, broad economic indicators or geo or national level
demographics then it is most likely a control.

## Deciding if a variable is an organic media variable or a non-media treatment variable

Organic media variables behave like paid media variables with no associated
cost. They usually are impression or reach and frequency based, and are
typically advertising activities with no direct cost, such as social media posts
and email campaigns. Non-media variables also have no direct cost but are not
media related. Typically, non-media variables are related to changes in the
underlying product, such as the price, a promotion or a change in the product's
packaging. Another way to determine what is appropriate for your variable is
that organic media have Adstock and Hill effects applied, but non-media
variables don't.

## Differences between the types of input variables

The following table can help determine what input variable is appropriate:

| Input Variable | Cost | Adstock/Hill | Intervenable | Effect (% Contribution) |
|---|---|---|---|---|
| `media` | x | x | x | x |
| `non_media` | - | - | x | x |
| `organic_media` | - | x | x | x |
| `controls` | - | - | - | - |

For more information about assumed causal relationships between input variables
in Meridian, see [Causal Graph](https://developers.google.com/meridian/docs/causal-inference/causal-graph).