| Standard brand lift, reasonable overlap | ipw and aipw | They keep every respondent and target the ATT directly. AIPW is usually the more precise of the two. |
| The sample was quota-matched on demographics | nn_matching with matching_features | Exact matching within the quota cells respects the design, with the score handling everything else. |
| Thin overlap, a few very large weights | overlap_weights, or trim | Stable estimates for the comparable population. State that the population changed. |
| A small control group | ipw or overlap_weights | Matching reuses the same few controls and discards the rest. |
| You need the effect for the whole audience | ipw and aipw with estimand="ATE" | Requires overlap in both directions, so check the low-score exposed respondents too. |
| You doubt the propensity model's functional form | aipw, and try gradient_boosting | The outcome model gives a second chance at getting the adjustment right. |