Yield is the most important selection trait for the ideal plant. Yield selection on single plants from segregating populations in the early selection cycle is unreliable because of unstable genotypes, low heritability and differential plant competitive abilities. Yield selection is best done in later cycles when sufficient quantities of the selected stable genotypes are available for replicated larger plot yield trials conducted over different locations and agronomic treatments. Breeding progress for yield in major crops has been generally slow at 1% -2% per year but nevertheless significant. Methods, e.g. breeding, physiological, biotechnological, to improve selection efficiency and shorten the selection cycles are not likely to substantially reduce the cultivar development time because of the mandatory extended cycles of yield testing. A smaller erect canopied palm with high harvest index that can tolerate higher density planting would be the oil palm ideotype for efficient yield enhancement. Such cultivars are unlikely to be available for the next 15 years. Nevertheless, plantations should accelerate replanting as improved cultivars particularly with better oil content are continuously being produced, and coupled with the simultaneous implementation of improved agro-management practices, larger quantum yields can be achieved. Existing planted materials already have high genetic yield potential, and the onus lies with the agronomist and manager to implement the prescribed agro-management practices to achieve the yield potential of the site and thus, narrow the gap between potential and realized yields.