Working with suspects

Authorised Professional Practice

This page is from APP, the official source of professional practice for policing.

First published
Written by College of Policing
20 mins read

The identification of suspects, the trace/interview/eliminate (TIE) strategy and the arrest strategy are explored in this module. In terms of the arrest strategy, considerations around timings, background checks, searches, planning, pre-arrest briefings and post-arrest issues are included.

TIE strategy

The term trace/interview/eliminate (TIE) is taken from major incident investigation. Some forces use alternative terminology (trace/implicate/eliminate, trace/implicate/evaluate) but this does not affect the procedures for TIE enquiries.


  1. Constructing a TIE category
  2. Populating and prioritising a TIE category
  3. Setting TIE parameters for suspects and time
  4. Establishing elimination criteria
  5. Conducting TIE enquiries
  6. Managing not eliminated TIE subjects

Investigative effort can be focused on those implicated, in order to identify the suspect(s).

Investigators unfamiliar with this technique should seek assistance from experienced colleagues/supervisors. TIE strategies can be highly resource intensive and, unless managed effectively, have the potential to incorrectly eliminate the offender.

Investigators may also consider using intelligence-led mass screening enquiries as part of the TIE strategy in major crime investigations.

TIE strategy example

An investigation into the assault of a man in a bar has failed to identify the suspect and the investigator decides to use a TIE strategy.


The suspect and time parameters, based on witness testimony, are set as:

  • white man
  • 20–30 years
  • 5’10” to 6’2” tall
  • dark hair
  • gold earring in left ear
  • 22.15–22.30


All the men in the bar at the time of the offence are included in the category. Populating the TIE category is difficult because it is not known who was in the bar at the time.


The investigator adopts a snowballing technique, starting with the witnesses who are already known. Twenty people are identified and four fit the suspect parameters. Three were with others who provide credible alibis. The fourth states that he did not visit the bar that evening and was home alone at the time of the assault. This generates further enquiries to test the sighting of him in the bar.

Constructing a TIE category

A TIE category is a group of people sharing a common characteristic which is likely to include the offender.

TIE categories are typically based on those:

  • with access to the scene at the time of the offence
  • in the vicinity of the scene at the time of the offence
  • living in, or associated with, a certain geographical area or relevant premises
  • related to, or associated with, the victim
  • with previous convictions for similar offences (usually known as MO suspects)
  • with physical characteristics similar to the offender
  • with access to certain types of vehicle

This is not an exhaustive list, and the more that is known about the circumstances of the crime, the greater the chances of constructing an accurate TIE category. Some pre-work may be required to determine the most suitable characteristics to use.

As more material becomes available, the validity of TIE categories can be reassessed.

Populating and prioritising a TIE category

Having decided which groups are likely to include the offender, investigators must identify as many group members as possible. Sometimes this is known with a degree of certainty. For example, a check of company records could create a TIE category of ‘those employed in the named premises’. In other cases the exact membership of the group may be difficult to determine. Those in a TIE category are known as TIE subjects.

Ways to populate TIE categories:

  • official records, for example, membership lists, payrolls, electoral registers
  • police intelligence databases
  • media appeals
  • snowballing

Filtering TIE categories:

Sometimes the population of a TIE category is large and it is not possible to carry out enquiries on everyone. Investigators may choose to start with those members most likely to be the offender. In such cases it is possible to apply filters according to priorities chosen by the investigator, for example:

  • proximity to the scene
  • date of last conviction of MO suspects
  • age (where this is not known, investigators may wish to prioritise those who fall within the likely age range of offenders for that type of crime)
  • sex (where the sex of the offender is not known, investigators may wish to prioritise those of the sex most likely to have committed the crime but should avoid stereotyping)

TIE parameters

Suspect parameters

These are the known characteristics of the offender that can be used to implicate or eliminate those within a TIE category. They can include:

  • sex
  • age
  • physical characteristics
  • fingerprints
  • forensic characteristics such as DNA, fibres, footwear impressions
  • ownership of a particular make or colour of vehicle
  • ownership of particular clothing

The value of these characteristics varies. A fingerprint or DNA is likely to eliminate all but the offender, whereas knowing only the sex of the offender is of limited assistance, although it still eliminates half the population. Investigators should always consider setting parameters wider than those suggested by the material to allow for a degree of error in the descriptions given.

Time parameters

The timeframe within which a crime was committed is a useful way of eliminating individuals from the TIE category. People may be able to prove that they were elsewhere when the crime was committed. Time parameters should be set as accurately as possible, based on the material available. When the exact time is not known, investigators should set the time parameters between the earliest and latest time at which the crime could have been committed.

Elimination criteria

Investigators need to establish the criteria by which they are prepared to eliminate members of the TIE category from the crime. The Home Office Large Major Enquiry System 2 (HOLMES2) database uses a tried and tested system of elimination codes. Investigators must decide which level of elimination to apply in each case. This depends on the material available, the nature of the offence and the characteristics of the TIE category.

HOLMES2 elimination codes

  1. Forensic elimination, for example, DNA, footwear impressions, fingerprints.
  2. Description (suspect parameters).
  3. Independent witness (alibi).
  4. Associate or relative (alibi).
  5. Spouse or common law relationship (alibi).
  6. Not eliminated.
Forensic elimination

Where forensic material or a fingerprint that uniquely links an offender with the offence is available, the relevant samples are obtained from TIE subjects for comparison against it. This method can provide definitive evidence which eliminates the individual from the enquiry.

A TIE subject can be considered as eliminated only if it can be shown with certainty that the forensic material or fingerprint must have been left by the offender and that no other person was involved. Where the offence has been planned, the offender may well have taken precautions to avoid leaving such evidence at the scene. In these circumstances the absence of such an evidential link does not confirm that the TIE subject was not involved.


In the absence of forensic evidence or a description, some TIE subjects can be eliminated through alibi enquiries. Alibi enquiries establish that a subject was not available to commit the offence because the person was in another location during the time that it was committed.

The degree of reliance that can be placed on alibi elimination depends on the credibility of the person providing the alibi. The HOLMES2 elimination criteria distinguish between alibi witnesses who are independent of the nominal, those who are associated with them in some way and those who are in an intimate relationship with them. Clearly, the value of alibi enquiries depends, to a large extent, on knowing with some certainty the location of the crime and the times during which it occurred. If it is not possible to set reasonably tight locations and alibi times, it is likely that many TIE subjects will remain not eliminated.

Not eliminated

Someone who cannot be eliminated using codes 1–5 must be recorded as not eliminated. This does not mean that they are a suspect in the case. Where the elimination criteria are broad, it is likely that fewer TIE subjects will be eliminated than where the criteria are more narrowly defined.

Implication and elimination testing

An important principle of TIE strategies is that implication and elimination are always provisional and should be rigorously tested against the material to hand and any new material that becomes available. It is good practice to regard TIE subjects as being either implicated or eliminated from the TIE category, not as being implicated or eliminated as the offender. For example, someone who has been eliminated against code 5, an alibi supported by a spouse, can always be re-examined if new material comes to light which allows forensic elimination criteria to be set.

Fingerprints, footwear impressions and samples given voluntarily for the purposes of elimination play an important part in many investigations. It is important to make sure that innocent volunteers are not deterred from participating. They must be fully informed of the reasons for requesting fingerprints, footwear impressions and DNA samples, and consent voluntarily to their being used for the purposes of a specific investigation. If the police or volunteer seeks to have the fingerprints, footwear impressions or samples retained for use after the specific investigation ends, the volunteer’s additional consent must also be fully informed and voluntary.

For further information see Police and Criminal Evidence Act 1984 (PACE) (Code D Annex F).

TIE enquiries

When carrying out TIE enquiries, investigators should be mindful that TIE subjects may also be potential witnesses. This applies particularly where they were present at the scene or are in a group which may know the victim or offender.

Interviewing TIE subjects

Each subject of the TIE category should be interviewed using the PEACE model of investigative interviewing to gather information on the characteristics that are of interest to the investigation. This could be a description, ownership of vehicles and their whereabouts at the time of the crime, together with the details of those who can verify this.

Where the elimination criterion is an alibi, investigators must thoroughly check the alibi with those who can verify it. This should enable some subjects of the TIE category to be eliminated from the investigation.

Where the elimination criteria include forensic material or fingerprints, these should be obtained and arrangements made to have the material tested. The results should be communicated to the investigator as soon as possible.

Treating TIE subjects as suspects

Where there is forensic evidence or other compelling material linking a TIE subject to the offence, or where their alibi is shown to be false, it is likely that they will be treated as suspects. People should not be considered to be suspects just because their alibi cannot be verified. Further enquiries should be carried out to gather material that may implicate or eliminate them.

Mass screening enquiries

This involves taking DNA samples or fingerprints from a selected part of the population and submitting those samples to the force forensic provider. This attempts to match a sample found at the scene of a crime to a member of the population. It is generally used in major investigations.


Such enquiries can be long-running and expensive, both in terms of human and financial resources, and may have a significant impact on the community. It is, therefore, essential that careful consideration is given to the likely benefits for the investigation. The use of this technique should be restricted to when more routine enquiries are yielding little or no information, or are proving more costly and time-consuming than conducting a mass screening.

For further information see:

Not eliminated TIE subjects

It is likely that some TIE subjects will be eliminated from the investigation. This enables decisions to be made about how to progress enquiries in relation to those who are not eliminated. The unique circumstances of the case and any new material available should direct enquiries. It is not appropriate to simply reissue the investigative action with the instruction to ‘continue enquiries’ as this is unlikely, on its own, to advance the investigation.

Large numbers

Where there are a large number of not eliminated TIE subjects in a particular category, it may be necessary to accept that little can be done until more material can be gathered. In these circumstances it may be more productive to put resources into locating and gathering material, rather than to continue to focus on individuals.

Small numbers

Where there are a smaller number of not eliminated TIE subjects, it may be possible to carry out further research into them, or to raise actions to implicate or clear them by more intrusive investigative methods such as searches of property, surveillance, interview or identification enquiries.

Arrest strategy

An investigator must decide whether the suspect can or should be arrested. The decision to deprive an individual of their freedom should not be taken lightly, and advice should be sought from the Crown Prosecution Service (CPS) if needed. Once a suspect has been identified, a strategy development approach can be adopted to gather material that will either implicate them in the offence or eliminate them. Every individual who falls within the suspect category must be treated in the same way. If there are specific reasons for not following the same procedure, these should be recorded in the crime report or policy file.

Power of arrest

PACE s 24 (as substituted by the Serious Organised Crime and Police Act 2005 (SOCPA) s 110) provides the statutory power of arrest for a person involved in a criminal offence. To make an arrest, an officer must first have grounds for believing that it is necessary, as detailed in PACE Code G. This may involve considering the intelligence case and/or evidential material. Before making an arrest, investigators must consider when and how the arrest should take place.

Timing an arrest

The way in which the identity of the suspect is discovered has a bearing on how and when an arrest should be made. If there is a choice, it is usually between making an early arrest and conducting a planned arrest. The decision about timing depends on a number of factors, which should be kept under continuous review. If the circumstances alter, the decision to make an immediate arrest or to delay it may have to be amended and the reasons for this recorded.

Factors to consider

  • Does the suspect pose a serious risk to the safety of the victim, witnesses or the general public?
  • Is there a likelihood that the suspect will commit further or more serious offences?
  • Is the suspect likely to destroy, conceal or falsify evidence that will obstruct the investigation?
  • Is further surveillance or other covert means of surveillance required? (This will need authorisation under the Regulation of Investigatory Powers Act 2000.)

For further information see Necessity to arrest.

Background checks

When a suspect has been identified, checks should be conducted into their background and lifestyle. These checks can strengthen and assist the case.


  • Is the suspect violent?
  • Does the suspect possess any known firearms and/or is there a history of use of firearms?
  • Does the suspect have or use any other premises?
  • Does the suspect have access to any vehicles?
  • Does the suspect have any previous criminal convictions, arrests?
  • What is their MO?
  • Are there any known associates?
  • Is there any habitual behaviour associated with the suspect, for example, are they a known gambler or a known drug addict?
  • Is there any intelligence regarding their previous behaviour in interviews?

In volume crime investigations, research into a suspect’s background and previous offending behaviour may reveal similarities to the offence currently under investigation. It may also lead to the identification of a particular MO which can assist investigators to identify other linked offences and to plan for subsequent searches and/or suspect interviews.


The timing of an arrest provides the investigator with an opportunity to plan a search of the suspect’s home address (or other premises linked to the suspect) and their vehicles (see Search Strategy). It may also provide opportunities to recover incriminating or corroborating material before it is altered, disposed of or destroyed. A search of a suspect’s premises may also identify property from other offences, or intelligence which can be used to identify other offenders or associates. Searches of the suspect’s premises should be carried out in accordance with PACE Code B.

Locard’s principle of exchange

When conducting searches or arrests, the investigator should be aware that evidential material may have been transferred from the victim or crime scene onto the clothing of the suspect. The suspect is, therefore, a potential crime scene which can be harvested, particularly if arrests and searches take place shortly after the commission of an offence. The possibility of cross-contamination must be considered and, wherever practicable, investigators who have visited the crime scene, or who have been in close contact with a victim, should not be deployed to search or arrest the suspect unless they have changed their clothing. Similar considerations should be applied to the use of vehicles used to convey victims and suspects.

Planning an arrest

The arrest should be planned, where circumstances allow. The formulation of a plan should take into account resources and logistics. If several suspects are involved, investigators must decide if the arrests need to be coordinated. They must also consider the most appropriate method of searching to use and whether search warrants are required.

Custody arrangements

Specific arrangements may be required in some cases. For example, if several suspects are going to be arrested simultaneously, it must first be decided if they should be held at the same or separate police stations, and whether sufficient facilities are available for this.

Other considerations

The vulnerability of the suspect, for example, age, disability, and the possibility that the suspect may have caring responsibilities for other vulnerable individuals should be taken into account. If the suspect does not speak English, an interpreter should be arranged. See also APP on Police response to concern for a child – Arrest strategy.

The timing of an arrest should be planned to make best use of the time the suspect spends in custody. Likely defences should be considered and catered for in interview plans (see investigative interviewing). Preparing an initial interview plan at this stage helps to maximise the detention time available to the investigator post arrest. Prior to arrest in serious or complex investigations, specialist interview advisers should be used or consulted to develop an initial interview strategy, especially where there are multiple suspects or unusual features associated with the commission of the offence.

Pre-arrest briefings

Arrest and search teams should receive a briefing prior to both the arrest and search phases, to ensure that all personnel are aware of each officer’s role and the reasons for the arrest.

Information to consider during a briefing

The following list is not exhaustive but covers some of the areas that may have to be considered:

  • circumstances of the offence
  • authority for search
  • nominated officer in charge of the arrest team
  • nominated officer in charge of the search team
  • communications
  • method of entry
  • items to be searched for (for example, clothing, footwear, weapons, trophies)
  • recording significant statements or silences
  • methods of recovery and cross-contamination
  • exhibits officer
  • loggist
  • health and safety
  • transport of suspects and location of the custody suite
  • interview teams
  • debrief – time and location

Post arrest

Once in custody, the suspect’s detention is controlled by PACE. At the time of arrest and before the suspect is delivered to the police station, anything that the suspect says must be fully recorded. In particular, any significant statements, for example, ‘I don’t know what you’re talking about’, made by the suspect post arrest and before formal interview should be recorded.

Where a suspect is released under Investigation (RUI’d), or no further action is taken, relevant information must be passed to the suspect. See custody for further information.


Arrest teams should be specifically briefed not to question or interview suspects further – unless there are imminent threats to life or property. All questioning of a suspect must be conducted in accordance with PACE.

For further information see APP on briefing.


At the conclusion of the arrest and search phases, a nominated officer should debrief the teams to ensure that all relevant material has been identified and documented.

Significant statements

A significant statement is defined in PACE Code C as:

One which appears capable of being used in evidence against the suspect, in particular a direct admission of guilt.

Identification of suspects

In some investigations the suspect is identified beyond reasonable doubt from the outset. They may have been caught red-handed, arrested during the initial response or identified from material gathered during the investigation. Whenever a suspect’s identity becomes known, consideration must be given to formal identification procedures.

Points to consider

Where investigators are conducting interviews with witnesses and identification is likely to be an issue, guidelines contained in R v Turnbull (1977) QB 224 and captured in the mnemonic ADVOKATE will help to secure best evidence.

Descriptions should be obtained while the recollection is fresh in the mind of the witness. In some cases the suspect may still be in the locality and may not have had the opportunity to alter their appearance or conceal/dispose of evidence.

The first principle of identification evidence is that failure to accurately record the description of a suspect as soon as it is available, and failure to use the description during the identification process, can seriously undermine the chances of a conviction. Investigators should bear in mind the difference between identification and recognition and that, where a witness claims to recognise a suspect, identification procedures must still be followed.


A Amount or length of time the witness had the suspect under observation.
D Distance between the witness and the suspect during the observation.
V Visibility conditions during the observation.
O Obstructions to the observations - whether they temporarily or partially inhibited the observations.
K Whether the suspect is known to the witness in any way.
A Any particular reason the witness has for remembering the suspect or event.
T Time the witness had the suspect under observation and the amount of time elapsed since the event.
E Errors in the description provided by the witness compared with the actual appearance.

Identification and recognition


This is a formal legal procedure that draws on a combination of established procedures, including the use of competent eyewitnesses who are likely to become key witnesses for the prosecution. It tests their ability to identify a person suspected of committing the offence under investigation as the same person they saw on a previous occasion


This is where someone who is not an eyewitness to the offence under investigation claims to recognise a person, depicted in an image, for example, a CCTV image. A successful recognition process can trigger other investigative options which may result in a chain of evidence leading to, and in support of, a formal identification procedure. In the absence of other supporting evidence, evidence of recognition can be adduced providing it is gathered in a robust manner that is open to scrutiny. R v Smith and others [2008] EWCA Crim 1342 illustrates recognition.

Identification procedure

An identification procedure must be held whenever:

  • a witness has identified a suspect or purported to have identified a suspect prior to any formal procedures being held or
  • there is a witness available who expresses an ability to identify a suspect or there is a reasonable chance they will be able to identify a suspect or
  • the suspect disputes being the person the witness says he or she has seen

An identification procedure is not required:

  • when it is not practicable
  • when it would serve no useful purpose in proving or disproving that the suspect committed the offence, for example, when it is not disputed that the suspect is already well known to the witness who claims to have seen them commit the crime
  • if the witness could describe clothing only or would not be able to recognise the suspect again.

The identification officer should be consulted when deciding on which identification procedure to adopt during an investigation. Due consideration must be given to the procedure most suited to the witness. PACE Code D provides guidance to investigators.

Suspect not known

The following identification procedures should be used where the suspect is not known:

  • street identification – when a witness can be taken to the particular location to see whether an identification of a suspect can be made
  • showing photographs to a witness
  • showing video footage or photographs of an incident to a witness
  • using facial imaging techniques, which may include artists’ impressions, composites and electronic facial imaging technique (E-FIT™)

For further information see ACPO (2009) Facial Identification Guidance.

Suspect known

In this sense ‘known’ means that the police have sufficient information to justify the arrest of a particular person as a suspect for an offence.

Where the suspect is known and the witness is available to take part in an identification parade, the identification procedures should be used in the following sequence:

  • formal identification parade or one of the approved video ID techniques
    group identification
  • a videofilm
  • a confrontation

Failure to ensure that the procedures set out in PACE Code D are followed can seriously undermine the strength of the prosecution case or give the defendant grounds for appeal against a conviction. However, there may be circumstances where, following the judgment in the case of R v Long [1991] Crim LR 453, the court will accept identification evidence that has not been gathered in accordance with PACE Code D, for example, where the witness recognises the suspect when they meet accidentally in the street.

Video identification techniques

Using video identification procedures means that the suspect does not need to attend. This reduces the time required to arrange formal identification procedures and gives witnesses an early opportunity to view the parade. Witnesses may also feel less intimidated when picking out an image from a computer screen than by attending a formal identification parade at a police station. However, video identification procedures preclude the witness from seeing the suspect walk, move or use a particular phrase or words.

Voice identification

In some cases this can be used to support other identification evidence, for example, where the suspect’s facial features are obscured, but they speak clearly to the victim or witness. It should not, however, be regarded as proof of identity in its own right.

Categories of voice identification:

  • where the witness identifies the voice of someone they know
  • recognition where the witness does not know the voice of the suspect, but would be able to pick out the voice as being distinct from other voices
Was this page useful?

Do not provide personal information such as your name or email address in the feedback form. Read our privacy policy for more information on how we use this data

What is the reason for your answer?
I couldn't find what I was looking for
The information wasn't relevant to me
The information is too complicated