It is appreciated that improvements in feed form (FF) result in improved broiler performance. However, research has primarily focused on the finishing growth phase due to associated high feed consumption allowing the greatest opportunity to observe performance benefits. Due to lower feed volumes required in the starter growth phase, it may be more economical to improve FF in the starter phase if improvements in overall performance and processing characteristics are observed. Study 1 investigated the potential for interactive effects of high or low FF presented in each of three growth phases to influence broiler performance. These data demonstrated the potential for FF presented in the starter phase to interact with FF in the finisher phase influencing day (d) 46 ending body weight (BW). Due to starter FF impacting overall performance, this led to Study 2 which consisted of two experiments with the main objective of determining the optimal crumble particle size for improved starter (d 0-14) performance. Experiment 1 utilized 5 different crumble particle sizes ranging from 1202- 2172 µm; whereas Experiment 2 implemented 8 differing crumble particle sizes ranging from 1174- 3736 µm. These data demonstrated consistent improvements in feed conversion ratio (FCR) as crumble particle size increased, with improvements in BW gain being demonstrated in Experiment 2 for crumbles 2800 µm and larger. Due to associated performance benefits with large particle sizes, Study 3 examined the potential to feed pellets, in comparison to crumbles, at different qualities during the starter period. Additionally, two commonly used genetic strains were employed to determine if performance benefits due to FF and feed quality (FQ) would be similar among different strains. Lastly, common diets were fed following the starter phase to determine if benefits due to starter FF would translate to improved overall performance. Feed quality and FF interacted to influence d 18 BW and d 0-18 BW gain. Examining carryover effects, d 0-32 and 0-46 FCR were influenced by FF and FQ; whereas d 0-62 was not influenced. These data suggest that length of the growout should be considered for determining FQ and FF to present in the starter growth phase.