🔗 Random Targeting Strategy

Random Targeting Strategy#

A random-selection baseline used to benchmark uplift models.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What it is#

A random targeting strategy selects customers (or units) at random for an intervention — a campaign, treatment or policy — instead of using a predictive or uplift model. In uplift modelling it is the baseline against which a model is judged: if a model cannot beat random targeting, it is not useful.

The diagonal baseline#

Treat a random proportion of the whole population — 20%, 50%, 100% — and, because the choice is random, the treatment effect spreads evenly. On a Qini or uplift curve (proportion targeted on the X-axis, cumulative incremental gain on the Y-axis) this traces a straight diagonal line: uplift accumulates linearly with the fraction treated.

A worked example#

With 10,000 customers and an average treatment effect of +5%, random targeting of 20% (2,000) yields about 100 incremental conversions; 40% yields about 200; and 100% yields about 500 — the campaign’s total incremental benefit. The line from (0, 0) to (100%, 500) is the random baseline.

Role, pros and cons#

That line is the benchmark: a good model’s curve starts steep — finding persuadables first — and stays above it throughout; a curve that hugs the diagonal adds no value. Random targeting is simple and fair (hence its use in A/B testing), but it wastes budget on sure things, lost causes and do-not-disturbs, delivers lower ROI, and cannot adapt to customer heterogeneity.

Theme: Causal Inference & Uplift · All terminology

Hint

Mind map — connected ideas

Qini Curve · Cumulative Uplift · Uplift Models · Uplift Score · Total Incremental Benefit (TIB) · Causal ML (Causal Machine Learning)

Hint

More in Causal Inference & Uplift

AUUC (Area Under the Uplift Curve) · Causal Effect · Causal Impact · Causal Inference · Causal ML (Causal Machine Learning) · Causal Trees · Cumulative Incremental Gain (CIG) · Cumulative Uplift · Incremental Conversions · Incremental Gain · Incremental Recovery Rate (IRR) · Incremental Revenue · Incremental Sales · Qini Coefficient