NIM and intelligence


NIM and intelligence

NIM and statistics

Job descriptions


Dysfunction of NIM

The role of Principal Analysts

Legal Status of NIM

What then must be done?

Joe's Telex


Copyright notice: text from this website may be reproduced freely, so long as is quoted as the source

About this website
Law enforcement intelligence practice in Britain is represented officially by National Intelligence Model, a 40-page book published by the National Criminal Intelligence Service in 2000 (download here). This publication - promoted by the Association of Chief Police Officers (ACPO) - describes nine ideas which it calls 'analytical products'.

The drawback with these products is that they bear no relation to intelligence analysis as practically undertaken. One such product is 'crime pattern analysis' defined, in part (we will come to the other part later), as "linkages between crimes and other forms of offending to reveal similarities and differences". One can compare cases, certainly, but there is no intelligence analysis product that begins and ends with comparing cases, since linking unsolved crimes does not address who is committing them.

Another alleged analytical product is 'network analysis', said to "[describe] linkages between people who form criminal networks [and] the significance of the links". Again, this does not constitute completed intelligence, since exploring criminal associations does not establish what the subjects are doing, or why they are suspected of criminality.

Neither 'crime pattern analysis' nor 'network analysis' equate to intelligence analysis, for their very definitions divide activity from association, thereby contradicting the need of the analyst to bring the two together in a seamless whole. The irrelevance of NIM products to intelligence can be demonstrated by the following example:

An intelligence analyst researches the modus operandi of burglary 1 and finds two other recent burglaries that may relate, due to similar property being stolen and means of entry. Further digging reveals a similar offence from several years ago which names a suspect (Smith) who was arrested at the time but not charged. The description of Smith is consistent with the current offender, sighted on one of the burglaries believed linked to burglary 1. A blue car seen at another of the burglaries leads the analyst to research vehicles linked to Smith, but no matching vehicles are found. However, the analyst researches Smith's associates and finds that one - Jones - owns a car that could be the suspect vehicle. The combination of previous similar MO, compatible description and similar vehicle is enough for the analyst to put up Smith as a suspect for the current series.

Now, at what point did the above intelligence work represent an array of different 'products' based around crime, network, commodity, etc? The intelligence analyst has merely investigated an offence and followed natural lines of inquiry. It does not represent three or four 'products' but a single, integrated intelligence picture of the available or possible truth of what has happened. The driving force is the analyst's investigative questions that stem from a single starting point. Had the analyst approached the matter using NIM methods, the intelligence could not have been reached.

Another NIM product which conflicts with intelligence analysis is 'target profile analysis'. A necessary implication of target profile analysis is that the target is chosen for analysis. This reverses the natural order of things whereby the identification of the target is a result - not the beginning - of intelligence analysis. 'Target profile analysis' may leave the intelligence analyst wondering how the subject was nominated as a target in the first place.


If we again look at the illustration of how intelligence analysis works, note the direction of the arrows. They go from left to right, that is, from the information to the intelligence. NIM reverses this direction by declaring a list of products that the analyst will produce, thereby causing the process to be started on the right. The effect is that the intelligence analyst is denied the ability to take their direction from the information that they are reading, and is bound instead by a range of pre-determined products which, by their nature, cannot represent the available or possible truth.

Why would the authors of National Intelligence Model do this? Why start the analyst in the wrong place?

The answer lies in the following statement on page 13, paragraph 1: "NIM is as much a management decision-making model as a description of intelligence process and products". Here is a scan:

Management decision-making is further emphasised on Page 7: "The Model provides important opportunities for law enforcement managers … the model can enable them to review intelligence systems and introduce more rigour into the management decision-making processes … It is also recognition of the changing requirements of law enforcement managers ... NIM gives such managers, individually and collectively, the framework for achieving that end [managing performance and risk, accounting for budgets]". For an image of the page, click here.

These statements raise the question: why is a document written for managers described as an intelligence model? According to Alan Buck, Chairman of the Superintendent's Association's Crime Advisory Committee, National Intelligence Model was originally intended for intelligence purposes rather than management purposes:

NIM, Page 6

Mr Buck confirms that, at the outset, NIM was to be what its title implies; a national intelligence model. At some point, however, it was appropriated for the benefit of managers. Thus, according to the Home Office website (their emphasis): "It is important to note that the NIM is NOT just about crime and NOT just about intelligence - it is a model that can be used for most areas of policing … The NIM is primarily a business model for use in allocating police resources."

Here are screen shots of the Home Office website with the text highlighted:

Consequently, the name of the publication, National Intelligence Model, is misleading, since the published document is intended as a "National Management Model". This is further indicated by the word 'manager' appearing nineteen times within its pages, whereas the word 'detective' does not appear once, even though intelligence is more relevant to the latter than to the former.

Replacing an intelligence model with a management model, while still wearing the mantle of 'intelligence', has consequences for the intelligence analyst profession. To understand why these consequences are negative, we must recall the problem to which intelligence analysts are intended as the solution, which is that law enforcement bureaucracy hampers intelligence. The intelligence analyst solves this problem by taking information out of law enforcement's administrative boxes, reassembling it according to offender, and handing it back as intelligence on which others can act.

As its text reveals, however, NIM is not written for intelligence analysts or other investigative and intelligence staff, but managers, and so does not recognise the organisational boxes over which managers preside to be a problem; instead, analysts are assumed to be the problem, which NIM seeks to 'solve' by providing managers with a list of products that they can demand. NIM does not acknowledge that managers, dwelling in their administrative units, are incapable of directing analysts, or that any attempt by them to do so will perpetuate an incomplete or distorted picture. The effect of NIM may be illustrated as follows:

Whereas the intelligence analyst should be located in position IA, a tasking from the manager of, say, the Robbery Squad on East Division (represented by RS) risks starting the intelligence analyst at RS. This means the intelligence analyst may be constricted by bureaucratic considerations into following route IA², not route IA which draws upon all sources of information in an unhindered way. Consequently, the "intelligence product" on the right is an illusion.

Further reading of NIM shows that this arrangement is not accidental. Page 22 states that analytical work must be "directed", and that, "The [intelligence] unit will only deliver the intelligence products that lead to the greatest impact if its work is kept on course, and strong management is required to prevent its wandering".

Note that NIM does not say that strong analytical work is required to "prevent intelligence wandering". Its stated position is that an intelligence analyst who is following lines of enquiry is "wandering"; but a manager is "on course" by virtue of being a manager. NIM is a doctrine of managerial infallibility.

NIM further undermines intelligence analysts by stating, "It is vital that an intelligence manager of appropriate status be appointed to ensure that meaning and significance are added to the analytical techniques and products before they are presented, as completed intelligence products, to the tasking and co-ordination group" (p21, para1).

Thus, it is a manager, not the intelligence analyst, who adds "meaning and significance" to analytical work. Meaning and significance is not deemed a result of intelligence analysis itself, nor is an intelligence analyst regarded as capable of producing "completed intelligence products". The implication is that managers - who have undertaken none of the investigation - can present themselves with 'intelligence' of their own choosing, quite regardless of the intelligence analyst who appears to perform no meaningful function other than make 'products' at their 'direction'.

NIM-thinking is illustrated on a grand scale by the Weapons of Mass Destruction debacle, whereby the twin notions that Iraq possessed WMD, and that Saddam Hussein was linked to Osama Bin Laden, were determined by politicians in advance as a pretext for war in Iraq. For instance, according to the "Downing Street Memos" of July 2002, "…intelligence and facts were being fixed around the policy". And former CIA official Tyler Drumheller said in an interview with Sixty Minutes that politicians were entirely selective in which pieces of intelligence they used, ignoring those facts that did not fit the case: "The policy was set. The war in Iraq was coming, and they were looking for intelligence to fit into the policy, to justify the policy".

Such dynamics take place in police forces every day; for instance, a manager, or even a detective, wanting to make the right noises can seize upon whichever offender is the flavour of the month and inflate their significance out of all proportion. Individuals who are promotion or image-obsessed can see the analyst as a threat, for the analyst will not blindly endorse the direction in which the said individuals want to go. The analyst's aim is to produce intelligence, that is, the available truth, not what is politically desirable.

NIM-thinking is implemented via training agencies such as the National Policing Improvement Agency, "… dedicated to supporting the implementation of the National Intelligence Model". NPIA provides a 5-day training course for analysts on Major Incident Rooms, which it summarises as follows:

"The aim of this course is to provide trained analysts with the knowledge and skills necessary for effective analytical support of a major incident investigation … requires the analyst to provide a briefing to a senior investigating officer based upon analysis performed within the scope of agreed terms of reference" (stored image / cache).

Note the last seven words. The notion that analysis should be conducted "within the scope of agreed terms of reference" means that a senior investigating officer (SIO) can veto any line of enquiry that the analyst may wish to explore, simply by refusing to agree it. Such training conflicts with the intelligence analyst's need to take direction from the information itself.

The danger of allowing police managers to decide the outcome of cases ahead of intelligence analysis is illustrated by Sussex Police who in 1998 decided that 39-year old Sion Jenkins had murdered his foster daughter Billie Jo. To support this view, Sussex Police presented what it considered compelling evidence; for example, Mr Jenkins made false claims about qualifications on a job application form some years previously.

The alternative view is that the decision by Sussex Police to charge Mr Jenkins had less to do with evidence than it did with internal politics. During 1997, Sussex Police had a series of failed murder investigations, meaning that SIO Supt Jeremy Paine was under political pressure to get a result; this accounts for Sussex Police's single-minded desire to convict Sion Jenkins to the total and otherwise inexplicable exclusion of alternative suspect, "Mr B".

An intelligence analyst on such a case would explore both possible suspects, as well as any other others, and present all to the SIO for a decision. The SIO may decide to investigate one, some or all of these possibilities, or even reject them entirely in favour of some other suspect. The salient point is that the SIO makes his decision in the light of the intelligence placed before him, and justifies it accordingly. The effect of NPIA's training, that analysis be performed "within the scope of agreed terms of reference", is that parameters can be created before analytical work is undertaken, meaning unexpected or undesired outcomes can be prevented. The result is that the SIO can avoid being seen to make a decision in the light of intelligence, for it is never acknowledged.

The National Anti-Intelligence Model

National Intelligence Model can therefore been seen to place two barriers against intelligence analysis; first, by enabling managers to direct the analyst via the selection of products, for analysis is subject to products rather than the other way around; and second, by enabling managers rather than intelligence analysts to supply meaning to intelligence. The effect of these distortions is that intelligence conclusions can be made at the outset of the process, rather than at the end.

There is a third obstacle; NIM's statement that intelligence analysis must be 'sanctioned'. An example of implementation is the intelligence process of Gwent Police:

The highlighted text states that analysis can only be undertaken following 'authorisation'. In a non-NIM environment, a manager who wishes to prevent an intelligence analyst from carrying out their function must intervene in a status quo in which the intelligence analyst is assumed to be able to carry out that work. Such prevention must be expressed and can therefore be challenged or disputed by the intelligence analyst. The effect of NIM is to reverse this assumption, so that the manager conveys authorisation. In other words, prevention of intelligence analysis is automatic; it need no longer be stated or justified, but is assumed to be the natural state.

Now, when did this happen; how did it happen? The intelligence analyst did not sign a contract of employment with a manager; the contract is with the employing organisation. If one employee, a manager, interferes with another, an analyst, the latter may appeal to the employer; but NIM, by putting in print that an intelligence analyst may no longer be assumed to carry out intelligence analysis, seeks to place the police manager above the employing organisation. In other words, NIM introduces a feudal arrangement in which the analyst exists vis-à-vis a manager, not the employer.


National Intelligence Model is not a model about how intelligence works; it is a sales document written to appeal to a particular mentality, that of police managers who are unable to trust staff they cannot control. Such individuals are not managers in the correct sense of the term, but bureaucrats at best, and harassers at worst.

At no point does NIM recognise that such attitudes are themselves a problem. Nowhere does NIM acknowledge that intelligence analysts are responsible for the stage between information and intelligence, or for producing actionable intelligence upon which others choose whether to act. There is nothing in NIM to oblige managers to let go of intelligence, or to acknowledge this inability to let go as a problem.

Next > NIM and statistics