Thoughts on Analysis: The Humble Collator

Oft maligned and ignored, the humble collator, day after day, feeds our intelligence databases

In some organisations the collator posts are treated as a rite of passage for your aspiring analyst – learning the ropes from the shop floor upwards.

Other places, they are seen as dead-end and never-ending jobs for those who will never quite make an analyst.

The latter case, of course, represents the worst model as it often breeds an environment where the collator is isolated from the intelligence analysis leading to a ‘them and us’ situation and ambiguity over the purpose and ultimate aim of the collator function. Does this apply to your organisation? You would be surprised how often this is the adopted operating model.

But even in the first scenario, where collation is viewed as a necessary skill to be learned, it all too often is to be got out of the way as soon and as quickly as possible. The glory is all to be gained in the proper intelligence world and not in feeding a database?

What both of these two lines of thought are missing is that should the collation function fail, then the likelihood is that the mission will also fail. Intelligence assessments based on erroneous information will always be wrong.

Let us first examine the primary roles of the collator. The first function performed for your organisation is that of first line analysis, the collator being usually the first person to see the incoming information and so should be the person hotfooting it to your desk breathlessly saying ‘coup just happened in Guatemala!.’ You remember those indicators and warning that are gathering dust in your bottom drawer, well now is a good time to dust them off and stick them to the side of your collator’s computer screen. He needs to know that information and needs to be encouraged to act upon it.

So suddenly we now have a collator who understands the mission and their part in achieving it, and is starting to gain some glory along the way.

Now back to that blasted database and let us examine what went wrong in our transformation from card index systems (anyone remember that far back?) and our ever increasingly complex IT storage systems. On the face of it should be in a far better position now, but too often we are presented with poorly inputted information onto systems designed by computer experts with limited knowledge of the intelligence function. Both of these are a failure in intelligence management.

Anyone forced to work with the i2 Analysis packages will know what fantastically pretty pictures it can produce, which are really useful when briefing the board, but it lacks basic functionality to be helpful in analysis. In my experience, it is used a great deal by police ‘intelligence’ analysts who spend an inordinate amount of time making sure linkage lines do not cross over each other! Some people call that intelligence analysis, I do not.

Databases and their attached visualisation tools need to be centred on links between essential elements of information (EEIs). The EEIs represent basic intelligence, to be understood and to provide context for the current and applied intelligence, which in turn is almost always contained within the links between EEIs.

Databases must remain as simple as possible, whilst being mission focused. They are there to aid the analytical process, so unnecessary clutter must be avoided. How often have we seen databases full of information such as an individual who is linked to his mobile phone, his car, his girlfriend, all his relatives, his house, the houses of his relatives, his landline telephone, his place of work, his local pub – ad nauseam. This may well tell us something about his pattern of life, but tends to hide the crucial facts. I would argue that most of those mentioned links are intrinsically part of that individual and so should, in most cases, be part of his personal record and not EEIs in their own right. That situation changes when we find, for example, that he lends his car to another person of interest to our organisation, and so the car then becomes an EEI with defined links to the two individuals. Simple!

So the challenge facing the management of intelligence teams is to provide a database and visualisation tool that aids achieving the mission along with well-thought through guidance on its use and that they employ individuals as collators who understand the mission and their role in completing it. In my view collation is not a separate function from analysis, and whilst I would not advocate your most senior analysts spending their precious time feeding the database, perhaps some consideration should be given to this being an integral part of your junior and middle level analyst function.

Databases need to be able to produce clear PicIntSums (as per the fictional example below) that are of use to the decision-maker as well as the analyst. Can your system do this?

Is your intelligence team working efficiently?  Could they be more productive and enhance your business model?  To find out more about what QRO Global could do to assist your business or to learn more about the types of projects that we have worked on in the past, see our Services Page or email us today – info@qro-global.co.uk