The most basic rule of the conduct of a homicide investigation, is that you are going to "do everything". You have to check every lead (including the dumb obviously useless ones), interview every potential witness (especially the dumb obviously useless ones), and look for and process every piece of physical evidence. I recall one homicide in which some obvious evidence was collected but not preserved, because there was no way to analyze it usefully.  Two years later, it could have been analyzed, due to technological advances, and it became a very critical issue at trial, even more so because it seemed to link it to another murder.

You must account for every false suspect, document the end of every false trail, and in every other respect conduct a very thorough investigation. This is justified by the seriousness of the crime, and by the length of the jail term to which we are going to ask the court to sentence the killer. The jury is going to be asked to make a very difficult decision, frequently based on merely circumstantial evidence. They have every right to expect that the police have looked under every stone. They see this done on TV and may believe that it is the routine procedure for every serious crime investigated by the police.

The defense will make an issue out of any appearance that the police neglected to conduct a full investigation, regardless of whether or not some elements of the investigation were redundant or even frivolous.

The first few hours....

The best chance to solve any crime is right after it happens. The first officer on the scene and his/her immediate backups will determine if a solvable crime gets solved. Unfortunately, they seldom get credit (or blame) for the things that they do or don't do, to preserve the solvability of the crime. Typically the follow-up investigator gets any commendations for "solving" the crime. But the case where an investigator brilliantly deduces the identity of the mystery killer and ferrets him/her out is only common on TV.

The first officer at the scene has to screen the incident, and decide or confirm that a homicide may have occurred. Some intelligent decisions need to be made about priorities in the case, from the time the first responder realizes that it is a possible homicide. Once he/she has decided that it is a possible homicide, he/she should summon assistance; the scene should be evacuated and secured, witnesses must be identified and instructed not to discuss their observations; preliminary interviews might be called for; some preliminary crime scene searching might be in order. All of these things need to be done on a prioritized basis.

As backup arrives, the perimeter of the scene should be further secured, and the witnesses isolated from each other and spoken with by the officers. The first responder and his/her immediate backups should consider whether a gross survey of the scene is needed, prior to the arrival of the investigation team. In general, gross surveys do no harm if done with reasonable care by a knowledgeable officer, but even a minor mistake can do serious harm and they seldom do any good. If the first responder took a moment to look around prior to securing the scene, a survey is not likely to accomplish anything more.

By the time a significant number of officers have arrived, someone must take responsibility for coordination. If a supervisor is not on the scene, then someone else should cease doing the work itself and begin supervising. (Your department should have a clear policy that the senior officer on the scene of any major incident, in the absence of a ranking officer, WILL take command.) Once having assumed the supervisory role, the coordinator has to announce the fact that he/she has done so, and ensure that every other officer knows where to find him. He/she should see that he/she is briefed on the available information at that time, and on every development.

The supervisor's job is to prioritize the work and to see to it that the work gets done, not to do it himself. He/she or she should avoid getting involved in the actual work of interviewing, or any other attention-consuming tasks. If short of manpower, the supervisor might take a station guarding the scene, freeing up an officer for other work. He/she should assign tasks in order of priority, but also with a view to the abilities of the officers. You don't, for instance, assign the brand-new untested recruit to guard the suspect. Each officer should be instructed to report back to the supervisor (or coordinator) when he/she is done with an assigned task, and brief the supervisor on any new information. The supervisor should take notes on the progress of the work, and convert those notes to a report later.

Sooner or later, an "investigator" is probably going to show up at the scene. Let's be clear on the role of the "Investigator".  He/she is not a supervisor. After being briefed on the nature and status of the investigation, he/she might take over the assigning of priorities, but not the assigning of actual tasks. Assigning tasks is the supervisor's job. The investigator's main job should be attempting to get a confession from the suspect, or locating the suspect, or figuring out who the suspect is. The investigators job is to follow the main thread of the investigation; the most likely leads. If there are any leads at all, his/her involvement at the scene should be fairly limited. A supervisor who cedes his authority to coordinate the work is not helping the case, and is not doing the detective any favors.  An investigator who tries to run the show is letting his/her ego get in the way of the case, and is alienating a valuable resource.  And there are never enough resources.

Once a suspect has been developed, the main thread of the investigation should follow the suspect, but there is still a lot of detail work to be done at the scene. Other officers at the scene should continue to report back to the supervisor, who will decide if the investigator needs to be briefed on new developments.

Preliminary Interviews.....

A "preliminary interview" is more of a conversation than a formal interview. Its purpose is only to identify the information most likely to lead to the suspect. During a formal interview, the officer should mostly just listen and take notes. But during a preliminary interview the officer should take a more active part in the conversation. The officer should ask the witness "what happened" and "who did it and how can we find him," and other questions designed to help answer those questions. The objective is to find out about any relevant information. Does the witness know anything directly related to figuring out who committed the crime, and/or how to find him? In a preliminary interview there is no need to be overly strict about the witness only relating his/her personal observations. We can accept opinions, in fact at this stage we should be seeking them out.

All of the information learned in preliminary interviews should be summarized to the coordinator, and any important descriptions or addresses should be given to the coordinator in writing. But don't give the coordinator or the investigator your own notes, you need those for your report.

A preliminary interview should just take a few minutes. If time and circumstances allow, the officer can and should go back for more information later. After the preliminary interview, again remind the witness that you do not want him/her to talk about his/her observations with any other witness.

Under no circumstances, should you attempt to conduct a "group interview" after the first few minutes. Interview each witness separately, out of hearing of the others. It is natural for  witnesses to rush up, in a group, to the first police officer they see, and begin telling him/her or her what they think they know. This may be an expedient technique for resolving a barroom brawl, but it has no place in a homicide investigation. Don't let witnesses do it to you. Once you have the basic idea of what happened, talk to each witness separately, even during the preliminary interviews.

Handling the suspect......

Some thought should be given to handling of the suspect at the scene. Violation of any of his/her rights should be scrupulously avoided. This is so basic that it may seem superfluous to mention it, but unintentional violations happen all the time and occasionally damage cases. Even unintentional violations by someone that the suspect just THOUGHT was a police officer, might be reason to throw out his/her statements (or even his/her arrest!) at trial. It is up to the police to justify our treatment of the suspect, and show that his/her rights were observed and protected. I once had a case where a suspect claimed that we tortured a 9-page confession out of him/her by allowing mosquitoes to bite him/her while he was in handcuffs. The judge didn't buy it, but only after I assured him/her that the I had asked the suspect if the mosquitoes were bothering him. (They were bothering me, but he was a tough guy and insisted he didn't care) The point is; think about your handling of the suspect.

Within the framework of the suspect's rights, he/she should be handled in the manner most likely to lead to the development of additional useful information (for instance, a confession). There is no law that says he/she has to be taken to jail right away. However, if you don't, be prepared to respond to the claim that you kept him/her at the scene in hopes of getting a confession out of him, which is probably true. Keep in mind that the courts have not (yet) prohibited us from trying to get a confession from a suspect.

There is also no law that says you have to Miranda a suspect as soon as he/she is arrested. He/she does not have to be Miranda'd, and probably should not be, until you begin to question him. Every officer who comes in contact with the suspect should be advised as to whether or not the Miranda rights have been waived.

If the suspect is to be kept at the scene for a while, he/she should be isolated from all other non-law enforcement people there. He/she should obviously be guarded, hopefully by an officer with some interviewing skills. The guarding officer should be briefed on the available information concerning the incident, and if possible updated occasionally by the coordinator. Whether or not to allow the suspect to overhear this is a matter of judgment, as is allowing the suspect to see what is going on. Depending on the suspect's personality and emotional state, it might be useful to let him/her see the body being carried out, and watch the activities of the police as we process the scene.

In making decisions about the handling of the suspect, consider that he/she may have trace evidence on his/her body or clothing. You may have to balance the likelihood of getting a confession out of him/her at the scene, against the likelihood of losing some trace evidence if he/she is not immediately processed.

Questioning the suspect at the scene.....

First, isolate the suspect. If he/she is under arrest, secure him/her and place him/her in the custody of the most skillful available officer.

If you deal with the suspect do not be afraid to talk to him. In fact, not talking to him/her is probably a bad idea; it is likely to encourage a confrontational frame of mind in him. However, you have to observe some fairly complicated rules. Pre-Miranda, you should not question him, nor do anything even slightly coercive to encourage him/her to talk about the crime. This does not mean that you can't make him/her comfortable and at ease, in hopes of him/her saying something useful. You can and probably should tell him/her what is going on and why he/she has been arrested. You can tell him/her about the evidence against him/her (with the permission of the investigator, of course). You are also allowed to make extensive casual conversation with him, but only about things unrelated to the crime.

If the suspect starts to talk about the crime, you have to make a decision. The basic choices are 1) shut up and just listen; 2) Miranda him/her and question him, hoping for a full confession; 3) get someone else to talk to him.

  1. This is the universal fail-safe choice. Just let him/her talk. Nod your head, move around a little so he/she knows you're interested, and make inarticulate verbal noises ("hmmm", "oh", "ah", etc) at the appropriate times. And take notes as fast as you can. You are not required to Miranda him/her at this point, assuming that you really did not encourage him/her to start talking. (But, rest assured the defense will claim you did!) Resist the temptation to direct the conversation or to get him/her to give a full confession as to every element of the crime. That would be a Miranda violation. You can do that later. Just let him/her talk. Respond to his/her questions without asking any questions of your own. Once the suspect starts to talk about the crime, you are on a roll; don't blow it. You can Miranda and question him/her later. However, you are going to have to justify your contention that you did not illegally pressure him/her to tell you the truth.
  2. In some cases, it might be fruitful to Miranda him/her when he/she starts talking. That way, assuming you get past the waiver, you can ask him/her all the questions you want to. Hopefully, you could obtain a confession as to each of the elements of the crime and one that demonstrates intimate knowledge of the crime. This is obviously better than the suspect's rambling, disjointed, self-directed statements. The risk is that the suspect will change his/her mind about talking when confronted with the Miranda Warning. Once he/she says he/she wants a lawyer, getting an admissible confession is very difficult. Premature Miranda Warnings can make a disaster out of a case, if there is insufficient physical evidence or witnesses.
  3. Get someone else to talk to the suspect. Having gotten the suspect in the mood to talk, why take a chance on breaking the mood? If, through inexperience or oversight, you commit some error in good faith, the worst that will happen is that the statements he/she makes to you after the error, will be lost. On the other hand, it is a very good idea to try to get another officer to witness the suspect's statements to you. Your witness should take notes too, and leave a report. This makes any inculpatory statements by the suspect less deniable at trial. Trying to motion another officer into the car while your suspect is blubbering out his/her confession may break the mood; tape recorders are handy at times like that.
There is occasionally a problem with "fruit of the poisonous tree" evidence; evidence that was developed from earlier evidence that was illegal. If you get lucky and the suspect has made some confessions or inculpatory statements to you, and you are the least bit worried about its legality, review your constitutional procedures with the most experienced and knowledgeable officers at the scene. Do this BEFORE you tell any other officers the details about what the suspect said. If there is a problem with the statements being unconstitutional, just keep the statements to yourself. Document the fact that "some statements" were made by the suspect, and keep your notes, but do not document the statements themselves in your report. Make a separate report about the contents of the statement, and seal it until an authority on the legalities determines it should be unsealed. Treat the sealed report as evidence.  As soon as you make you potentially tainted information available to another officer, it is legally assumed to be available to every officer, and "furit of the poisonous tree" applies.

On the other hand, if the statements were obtained legally, no problem.

As you deal with the suspect, listen carefully for anything he/she says that might indicate a motive for the killing. Although motive is not an element of the crime of homicide, many juries are extremely reluctant to convict a killer without understanding his/her motive. Perry Mason always told the jury why the guy did it, and your jury will expect to be told.

Also listen for expressions of intent, and of knowledge of the details of the crime. Take notes on everything the suspect says. Anticipate that you will have to leave a report on the conversation, to include your efforts to observe his/her rights. (Even if the suspect says nothing, you still have to leave a report.)

The magic words......

As your contact with the suspect progresses, listen to the flow of the conversation. Don't encourage the suspect to say the magic words, "I want a lawyer, " unless of course, he/she actually wants to. If you find your own conversation leading in that direction, change the subject. This is not to say that you should try to prevent the suspect from demanding a lawyer if he/she wants one; he/she absolutely has that right. But we are not required to bring the lawyer business to his/her attention until we start to question him. Once the suspect utters the magic words, there is probably not going to be a confession.

For new officers, let me empathize that.  Once the "L" word has been uttered, we are done talking to the suspect, for the duration of the case.  The next public words he/she utters about the case, are going to be on the last day of the trial, after he/she has heard what every other witness has to say.  Since the suspect is the only person who knows the truth about his/her guilt, it is a big handicap not being able to talk to them.

On the other hand, if a suspect has made a casual expression of interest in a lawyer, but continues to talk about the crime, we can continue to listen to him. You obviously can't question him/her anymore, even if you have a prior Miranda consent. At the next opportunity, you have to ask ask him/her if he/she really meant that he/she wanted a lawyer, or if he/she was just talking. It is good, at that point, to take a break, even if he/she says he/she didn't mean he/she wanted a lawyer, to counter any argument that you pressured him/her into changing his/her mind.

Furthermore, the suspect is entitled to change his/her mind about having a lawyer. So even after he/she has demanded one, if he/she indicates an interest is discussing the crime, you can ask him/her about the Miranda waiver again. You are not, however, allowed to TRY to talk him/her out of wanting a lawyer, once he/she has demanded one.

Gross Survey of the Scene....

A gross survey is not a crime scene search. It is just looking around in the crime scene (the area that is going to be secured for later searching and processing) to see if anything needs to be done right away. Examples might be a bootprint in the snow that is being covered up by falling snow or wind; it should be protected prior to securing the scene. Another example might be a piece of evidence (say, a wallet lying on the floor, out of place) that might immediately identify a likely suspect; it should be examined in the least intrusive manner, right away.

The danger of a gross survey is that it involves carelessly planned and inexpert movement through the scene. It creates a risk of destruction or alteration of trace evidence, addition of extraneous material to the scene, or the movement of physical evidence.

A gross survey serves three purposes: to determine whether any perishable evidence is deteriorating; to determine if anything can lead immediately to a suspect; and to ensure that there is no one left in the scene behind the last officer out. If done carefully, surveys do no harm, but there is a powerful temptation to do more than is necessary.

If done, surveys should be done with notebook in hand. The officer takes notes on everything he/she does, everyplace he/she goes, anything he/she is forced to move or alter. Later, he/she can fully document his/her activities in the crime scene prior to the crime scene search. Note-taking like this preserves the admissibility of the results of the crime scene search. It also naturally discourages you from indulging idle curiosity.

Surveying might include moving through the scene to look quickly at other rooms. This might or might not be a good idea, depending on the circumstances. The odds of making a serious blunder and messing up some important evidence are fairly small, but so are the odds of accomplishing anything important by walking around in the scene. The more physical evidence that seems to be present at the scene, the stronger the temptation to "analyze" it immediately, and the greater the chance of messing some of it up.

If you are not reasonably sure that you can accomplish something positive in the scene, secure it and leave. You or someone else can re-enter later if leaving was the wrong choice.

I am not trying to encourage gross surveys. The decision as to whether they need to be done can only be made on a case-by-case basis. Usually, they are not fruitful, nor are they usually necessary unless you have nowhere else to look for leads. However, I personally investigated one case, that would never have been prosecuted, much less convicted, if we had not decided to look in the crime scene before the search team showed up. Because we quickly found the critical lead in the scene, we were able to locate the suspect and gather overwhelming evidence from his vehicle, before he got home. It would have been gone if we had waited.

Whether or not to do a gross survey of the crime scene, beyond just securing it, is not usually a decision for the field officer. Supervisory and investigative backup is usually close enough that the decision can be deferred until more information is available, and someone experienced enough to make the decision (and with enough rank to take the heat for it) has arrived and briefed.

Securing the Crime Scene.....

Even in the ideal case when the murder was witnessed by several reliable and cooperative witnesses, and the killer has confessed, a full crime scene search should be done. This means that the scene must be protected until a proper search can be organized; frequently a matter of several hours. A formal crime scene search will be done in most homicide cases, even if it is not likely to produce any useful evidence. This is a part of the "do everything" philosophy that has to guide a homicide investigation.

Securing the scene may require some creativity; or some manpower, but it must be done. It takes some time to assemble a proper search team, and there are frequently other priorities. The formal search is frequently not done for hours, sometimes even the next day. There is no problem with this delay, if no perishable evidence is lost and the scene is secured in the meantime.

Securing the crime scene serves two important functions. It prevents evidence from being removed or destroyed, and it prevents extraneous evidence from being added. By far, emergency personnel (including police officers) are the worst offenders in terms of adding material to the scene. Ambulance workers should be instructed as to where they may move and place objects if they have to work in the crime scene. It might be better for them to move an injured person out of the scene, if that is practical. But tell them to leave their debris behind.  They can take the living victim, but only the living victim, out of your crime scene.

Professional firefighters don't seem to have the mentality for careful movement in a crime scene; keep them out if you can.

After the first rush of excitement, there always seems to be one or two officers with nothing to do. In addition, in some departments, officers of such high rank that they can't be assigned tasks, may start to arrive. By this time, hopefully, a door guard will have been posted at the crime scene, and his/her most difficult job will be to keep his/her fellow officers out.

After the first responder has left, no one else has any business in the crime scene until the crime scene search team is finished. This includes the Sheriff or Chief, the investigator(s) and the supervisor; they should all just stay out.

Absolutely everyone who was present in the crime scene after the crime, should document their movements and observations in writing. This includes every witness, ambulance attendant, fireman, and the coroner or medical examiner; get written statements from them. It also includes every cop regardless of rank; if you enter the crime scene, you leave a report documenting your movements and observations, in detail.

Crime Scene Processing.....

Before beginning a crime scene search, consider whether a search warrant will be needed. The simple fact that you are already inside the house, does not mean you can search it. While the police are generally allowed to be at the scene of a death despite the objections of any property owner, we do not necessarily have authority to search the scene over such objections.

In every case where a person with a property interest in the crime scene might be or later become a suspect, a warrant should be obtained. Consent searches are legal, but more risky than a warrant search. IF there is consent get, the consent in writing. Ideally, you would obtain both written consent from every property owner, AND a warrant. (This is an example of the "do everything" philosophy; you now have redundant legal authority to conduct your search, and the defense has no possible challenge.) Retain custody of the crime scene until you have obtained the warrant, but do not search it until the warrant is signed.

Even a freely given consent can come back to bite you.  For instance if a parent or roommate gives consent to search, the suspect can still claim exclusive privacy rights to their own room.  Get a warrant if you can.

The search itself and related processing must be planned out and organized. A team should be assembled, and jobs assigned based on appropriate skills. (The old rule that the junior man "gets" to be the recording/collecting officer is baloney; assign jobs based on skills.) Someone has to take pictures, someone has to take measurements, and someone has to collect, record, and take custody of any evidence that is found by the other searchers.  Someone, hopefully several people, have to do the actual searching. Only one person does the actually collecting. Depending on the available manpower, of course, these jobs may have to be divided up or combined. Each job has to be done, and it has to be clear who is going to do it.

Part of the planning of the search should involve the order in which the various parts of the crime scene will be searched. Keep in mind that the photographer and collection officer will have to move back and forth between various searchers, and you don't want them moving through un-searched or un-processed areas. It is sometimes useful to search those areas least likely to produce important evidence, first. This gives you more room to work, as various areas become cleared.

Pictures should be taken of the scene itself before anyone re-enters, and taken progressively as the team moves through the scene. Pictures should be taken of each piece of evidence before it is collected or otherwise moved. Measurements showing the location of the evidence, to an appropriate degree of accuracy, should be made.

If you have never done a crime scene search, here are a few very basic rules: Set a pattern for your searching of the assigned area (it does not matter what pattern you use, just use one). Search the floor first, then move up. When you find something useful, stop. Don't touch the evidence. Call the photographer and collection officer, assist them as necessary, and then move on.

Always walk or crawl FORWARDS.

It is highly undesirable to add trace material (fingerprints, fibers, footprints, etc) to the crime scene or to alter the anything before it has been photographed and documented. It is even more important however, when you make the inevitable mistake, to be aware of the fact that you did and to make a record of it.


There are a few bad crime scene habits that police officers have accumulated over the years, which deserve mention here:

Follow-Up Investigation.....

In a homicide investigation, we have to "do everything". We have all lost good solid cases, and it can happen in homicide cases too. There is just no such thing as an open and shut case. Even if you have a completely legal confession, you could lose the confession to an off-the-wall ruling.

Furthermore, anything that has been neglected in your investigation, will be brought up by the defense in the courtroom. The argument goes like this: If the cops had only checked that one last oddball lead, they would have found out that the defendant could not possibly committed this crime. The cops didn't do their job, so the jury is told to get mad at them, not the defendant.

The defense attorney will definitely interview any potential witness that you don't get to. Of course, if the witness knows nothing useful, or something damaging to the killer, then the attorney will say nothing to anyone about it. He/she 's a defense attorney; he/she doesn't have to. But if the witness's testimony is favorable to the killer, you had better be prepared to refute it in court. If you never even talked to the witness, you are in trouble.

For the same basic reasons, field officers frequently find themselves assigned to do some pretty stupid-seeming things during a homicide investigation. It may seem like you are just confirming the obvious....because that's exactly what you are doing.

Media Relations.....

Minnesota statute 13.82 and 13.83 defines information that the police must release to the media when an arrest has been made. Beyond that, the "public's right to know" is just a slogan, not a law. There are a number of good reasons why a decision may be, and usually is, made to not inform the media about the details of a case.

America is blessed with the world's best news apparatus. Good use can be made of it in broadcasting descriptions and in asking the public for help in identifying possible suspects. Sometimes the decision to put out this information must be made quickly, even at the scene. But in general this decision should be left to a fully-informed, well-experienced ranking officer, after consultation with both the investigator and the prosecutor.

Unless he/she has been assigned to do so, no field officer at the scene of any crime should brief the media. For a major crime, a press conference should be scheduled. Every officer at a the scene should be informed when and where the press conference will be held. He/she should give the time and location of the press conference to every news hound who approaches him. The only other thing that any officer should say about the crime to any non-police person, is that "no other information is being released at this time." Even "I don't know," is a bad answer; stick to "no other information is being released at this time."

There is no such thing as "Just one question, officer?" If you answer the first question, they ask you a follow-up question. Et Cetera. Until you realize you've been had. This technique is a part of every journalist's training. Don't let it get you.

Reporters work under a scoop system. They don't get in trouble with the editor just because they fail to get information. But if some other reporter gets information that they didn't get, then they are in trouble. Once you convince them that there is only one source of information (the press conference), they will get off your back. On the other hand, if you develop a reputation as the cop with the loose lips, they will give you no peace. No officer should discuss the facts or his/her opinions with the media, and witnesses should also be asked not to.

Whodunits...."Who did it and how can we find him?"

The initial investigation of a whodunit should be fast-paced. While mistakes should be avoided if possible, minor ones can be tolerated if the case continues to develop satisfactorily. The important thing is to develop information as quickly as possible.

The first priority in the whodunit is to figure out who committed the crime, AND to find him. Statistically, the odds of solving the case go up astronomically, if the cops know the names of the victim and suspect, and the location of the crime scene(s), within 24 hours.  In fact, knowing those three things within 24 hours is the single biggest predictor of success in a homicide.  So it is better to take a chance on making some mistakes, than to fail to develop information rapidly.

It is much easier to obtain a confession and critical physical trace evidence from the killer, shortly after the crime. Even if the crime happened a long time before the investigation begins, you are more likely to get a confession from the killer if he/she is confronted shortly after a major, well-publicized development in the case, such as when the body is found. This should be reflected in the conduct of the investigation from the time the first responder shows up until the killer has been identified and is in custody. It is especially crucial during the first few hours.

The first responder and his/her arriving backups should concentrate their efforts on finding and talking to potential witnesses. The two important questions to ask the witnesses are "who did it" and "how can we find him". Officers should also conduct a gross survey of the scene, for evidence that will readily identify the killer. Examples of such evidence might be bootprints patterns, the likelihood of blood on the killer's clothing, the type of weapon used (if it might still be in the killer's possession), or evidence of stolen property; anything that will help someone coming across the killer to recognize him/her as such.  Blood, fingerprints, and other technical trace evidence may be critically important to the final resolution of the case, but are not so useful in immediately identifying the suspect, and getting to the guilty party quickly is paramount.

Witnesses must be asked about descriptions, names, addresses and associates of any person who might have done it or who might know who might have done it. Those potential witnesses and suspects must be located and evaluated quickly. Every officer at the scene should feel a sense of urgency in developing good suspects as quickly as possible.

The media might be asked to broadcast descriptions or other information to help in locating the killer. The decision to inform the media is not to be taken lightly, there are complex ramifications. However, if you need the help of the public in identifying or locating a suspect, it is best to get the information out as quickly, while the information itself is "hot", while the memory of the public is fresh, and before the killer has gotten to a safe place.

In the initial investigation of a whodunit, there is no substitute for a competent, well-trained, widely experienced group of field officers who can work as a team, and work quickly with a minimum number of mistakes.

This is not to say that the scene should not be protected, nor that witnesses should not be handled in a careful and planned manner, and their information fully analyzed. Even though it is imperative to quickly figure out who did it and how to find him, you are still going to have to establish proof of guilt. Reasonable cause, or even probable cause, to make an arrest is not the same a proof. Sooner or later, you are going to have to do a full and methodical crime scene search, and do formal interviews of all potential witnesses, and check out every lead, including extraneous leads. Some of this will have to be delayed, of course, but it cannot be ignored if you want to get a conviction.

The chances of successfully prosecuting a whodunit are dramatically increased if you can get to the killer quickly.

Killer-in-Custody....."let's prove it"

Where the whodunit is a fast-paced, intuitive style of investigating, where doing the wrong thing is better than doing nothing, the killer-in custody case is slow, methodical, well-planned, detailed ad nauseum and deductive. With the killer safely in police custody, or at least satisfactorily identified and any physical or trace evidence on his/her person preserved, priority should be placed on careful planning of the development of the case.

If the obvious suspect is present at the scene, then the case is well on its way to a successful conclusion. Keep in mind, however, that you may have the wrong guy. Trying to prove that the wrong guy did it, can be very harmful to a case. The likelihood of catching the right guy decreases with every hour spent working on the wrong guy. Even if you later catch the right guy, the defense attorney will have access to all the evidence you developed against the original suspect(s). He/she will use it to divert suspicion from his/her client in the eyes of the jury.

With the suspect in custody, high priority should be given to the best possible crime scene search. Physical and trace evidence is not typically useful in directing us to a suspect, but once we have him, the evidence can prove that he/she did it.

The scene should be secured until it can be fully processed by the most qualified set of searchers that can be assembled. If you have access to evidence specialists, they should be called out, even if it means a substantial delay. There is usually no need to rush into a crime scene search, especially if the (right!) suspect is in custody.