SAGD (steam-assisted gravity drainage) has been proven as an effective technology to enhance heavy oil/bitumen recovery. The main shortcoming of this method is its inefficiency, a result of high water and energy consumption. As a solution to SAGD efficiency improvement, we propose the addition of chemicals resulting in higher recovery and reduced steam consumption. The objective of this paper is to screen new generation chemicals as additives and study the mechanisms and optimum injection strategies. This screening was achieved through Hele-Shaw-type macroscopic visual experiments. We previously screened a wide variety of chemical additives for steam flooding. As a continuation of this work, these chemicals were tested for SAGD conditions using a new visual experimental design where the optimal injection strategies were identified, eventually providing a reference for the selection of chemical additives for field applications. Eleven conventional and new generation chemical additives (heptane, biodiesel, dimethyl ether, LTS-18, Tween 80, Span 80, Novelfroth 190, ionic liquid [BMMMIM BF4 ], silicon dioxide nanoparticle, DES 9, and DES 11) were selected based on both their strong thermal stability and enhanced oil recovery capability. The recovery improvement mechanisms for the different chemical additives and different injection strategies were identified through flow characteristics, emulsifying ability, viscosity reduction capability, and wettability alteration. Simultaneously, the mechanisms were studied from a macro perspective via analyzing areal sweep efficiency and microscopic oil displacement efficiency together with observing the images acquired during the process. Three different injection strategies were applied for each chemical: (1) chemicals were injected at the beginning, (2) in the middle, and (3) at the end of the steam injection. The chemical additives played different roles in recovery improvement, and different chemical addition strategies yielded different mechanisms. Heptane exhibited extraordinary characteristics with maximum "steam saving" (34.52%) when the middle injection strategy was applied, and maximum ultimate oil recovery (64.75%) was obtained for the end injection strategy due to the ability to reduce the viscosity of heavy oil by dissolving around the chamber edge. Steamflooding with Novelfroth 190 showed an excellent performance for the middle and end injection strategies because of its ability to develop rapid oil drainage "channels". The addition of surfactant LST-18 presented the ability to improve the EOR by forming emulsions. Additionally, the distributions of the steam chamber in the Hele-Shaw cell were different because of the changed flow characteristics when the same chemical additive was injected at different times, thus showing the ability to reduce viscosity and form emulsions with different strengths.