1. Virtually all police departments already solicit direct feedback, both positive and negative, from citizens. This is just an effort to significantly increase the amount of feedback received through lower friction (phone based experience encouraged by officers rather than a paper form you ask the officer for or a website you go to later). For departments under supervision, DOJ generally requires that police departments or their oversight agencies accept positive and negative feedback, including anonymously, from the public. Part of the motivation of this kind of program is explicitly to increase the amount of positive feedback, because as a rule of thumb higher friction feedback processes result in more negative average feedback in this kind of situation (only people who are really pissed off go to the effort), but less friction to feedback also results in a better cross section of the population responding (traditional methods like paper forms tend to bias towards proficient English speakers, individuals with more free time, etc). Currently, even very busy police officers typically receive only a small handful of citizen comments and they are virtually always negative and resulting from a complex or extreme situation. This makes all sorts of biases and edge cases a major part of the data - in fact the majority of it. Significantly increasing the number of "reviews" will make them far more positive on average, and more related to actual officer conduct and not unusual interactions.
2. The fact that certain people give more consistently negative feedback and that certain officers, due to their assignments, interact with these people more frequently is well understood and already significantly influences police department metrics. For example, officers with advanced training in crisis intervention can be expected to receive more complaints, rather than less, because they are dispatched to interact with a community that is more likely to complain. The increase in overall volume of feedback from this type of program should make it easier to characterize this effect.
3. Police departments with well-functioning professional standards programs integrate many methods of overseeing officer performance, of which direct public feedback is only one element. For example, random audits of OBRD footage are typical (supervisor watches at least one randomly selected hour of body camera footage per supervised officer per month is an example policy). Negative public feedback, if handled as a complaint proper, will lead to a supervisor or full-time PS investigator reviewing body camera footage and potentially interviewing witnesses etc. in order to substantiate or dismiss the complaint.
4. Police officers in large departments receive a substantial amount of procedural criticism ranging from chain of command reprimands to being a subject of lawsuits. Police unions generally favor more direct feedback mechanisms because they allow for intervention in problems before they reach the stage of a lawsuit. From my experience working with direct feedback programs officer support for this concept is probably surprisingly high for that reason - better to have a contemporary, written negative review than a months after the fact IA investigation.
5. Officers in most departments are probably not going to be significantly more lenient due to this program, because their decisions to cite/not cite or arrest/not arrest is regularly reviewed by chain of command and in more serious cases outside professional standards as well. Officers are generally more likely to get in trouble for not citing than for citing, although there are a number of factors that can influence this (the situation can get complicated, e.g. blanket requirement to arrest on felony warrants is typical but may be waived in the case of e.g. DV victims). Like it or not, discretion is a well-established part of policing and departments are increasingly developing methods of monitoring and standardization in order to quantify and reduce e.g. racial bias.
6. Oversight, performance evaluation, and policy enforcement is a large function that typically involves multiple distinct commands in a police department (training, field training, performance, professional standards, etc). Most large departments are implementing some form of computer-based data system to surface trends and anomalies in officer performance measures, but particularly due to the significant power of police unions very few decisions are made based on actual data (zero in a lot of departments). Ultimately all of the collected feedback will be one of many inputs into the decision making process on advancement, discipline, etc. In practice, the subjective judgment of the chain of command virtually always wins, even in cases where it is difficult to square with objective data. Police union CBAs usually enshrine this concept in contract language (e.g. no discipline can be imposed without the subjective approval of a board of command staff based on a hearing, even in cases of clear and provable violation of policy that has a specified penalty, is a typical CBA term).
7. I like that this program seems intended to surface more subjective feedback on demeanor and professionalism, because these are huge problems in a lot of police departments that are extremely difficult to address because of the difficulty of establishing any evidence of a professionalism problem. Particularly, the lack of data on the issue makes it a bottom priority and something that the chain of command is encouraged to ignore rather than address, yet it is one of the biggest factors in public perception of police.
8. None of this program is actually new, just an effort to increase the response volume on existing complaint/commendation programs that have typically existed since WWII.