EWJ August 62 2025 web - Journal - Page 26
In urban areas it is common to encounter users’
phones with bulk advertising texts from several
different drugs lines.
Potential vs ‘Confirmed’ transactions
IDMU has teamed up with a data analytics company
to automate the analysis of phone traffic data. The
above ‘how often’ figure provides an estimate for the
number of potential transactions involving the line on
any particular day.
The algorithm does underestimate the number of
transactions by restricting the potential customers to
numbers receiving bulk outgoing texts, not accounting
for multiple transactions from the same number on the
same day, and not counting potential orders placed via
text or other messaging services, however this effect is
usually more than outweighed by the gross overestimation of the quantity involved in each transaction.
The phone traffic analysis has been refined to identify
all incoming numbers which are subsequently contacted back by the drugs line, or the number of a line
associate (e.g. runner), within a limited time period
that day. This figure is treated as the number of ‘confirmed’ transactions for any particular day.
The Alternative IDMU Algorithm
Starting from first principles, the quantity of drugs
supplied by a line can be estimated using the formula:
HOW OFTEN x HOW MANY x HOW MUCH
Caveats
The standard Crown statement lists factors not taken
into account (e.g. not on customer list, text orders, ordering via a friends phone not on list, collective purchases, introductions, drugs supplied as payment ‘in
kind’, communication direct with runner, pre-arranged deals or drugs seized by police or stolen via
rival dealers) all of which are used to paint their figure
as an underestimate.
How Often? – Number of Unique Incoming Contacts
per Day (potential) or subsequently contacted by the
line (confirmed)
How Many? – How many wraps (of each type) per
transaction
How Much? – The average quantity of drugs in each
wrap
Factors leading to an overestimate would include
transactions spanning midnight, callers who run out
of credit and use a different phone, who ring when
temporarily out of stock, or who are waiting for a new
batch not yet delivered, callers asking for credit (tick)
or who have insufficient funds to complete the purchase, callers unwilling to wait if quoted a long delivery time, callers who contact the line after all the day’s
stock has been sold, or where an exchange is aborted
due to visible police presence of members of the
public getting too close.
How Often – Phone data is analysed to generate the
total number of unique incoming contacts each day,
representing successful incoming phone calls or voicemails (longer than specified number of seconds) or
successful text messages to the drugs line. Second or
subsequent contacts are presumed where there is a
significant time gap between successive incoming
events from the same number. Network messages and
numbers of associates (e.g. co-conspirators or family
members) are excluded. Transactions which span
midnight are attributed to the date contact was
initiated rather than double-counted on both days.
Summary
It is not reasonable to presume that drugs line users
will each purchase on average 0.9g crack cocaine and
heroin (combined) each and every day. Although such
a figure represents mean consumption, median usage
is a small fraction of this (0.166g combined), and heavier users are not only more likely to purchase drugs in
larger units (not from retail drugs lines), but are also
more likely to receive methadone or similar substitute
drug prescriptions reducing their demand for ‘street’
heroin.
How Many – This data is derived from phone
messages where quantities are mentioned, and ledgers
(paper or electronic) where retail sales have been
recorded. All such sales are recorded in a database
which is kept updated as soon as further reports come
in. Where there are significant numbers of orders in
a particular case these figures can be used, however
overall figures were as follows (at the time of writing):
£5-£10 lines (50-120mg per wrap) – 303x reports, 23x
cases – 1.63x wraps crack, 1.71x wraps heroin – total
3.35x wraps per transaction (value £22.04, or £6.59
per wrap)
Furthermore the drugs market is competitive and
users, even in relatively remote areas, will normally
have a range of suppliers or ‘lines’ to choose from. The
algorithm unreasonably presumes that every single
potential customer will purchase exclusively from the
line they are investigating.
£15-£20 lines (140-220mg per wrap) – 593x reports,
6x cases – 1.96x wraps crack, 1.09x wraps heroin –
total 3.05x wraps per transaction (value £43.94 or
£14.36 per wrap)
The empirical evidence from message contents and
phone records suggests average purchase quantities
are 3-4x wraps in any one transaction, involving
353mg to 469mg combined crack or heroin (depending on wrap size).
How Much – This is derived from forensic analysis of
wraps seized in the case and attributed to the line in
question.
The algorithm underestimates the potential number
of transactions, but overestimates the quantity
purchased by users per day from any particular line.
Note that where there are fewer than 10x outgoing
events (calls or messages) the line is treated as ‘fallow’
or out of stock, such dates are typically characterised
by numerous failed calls or messages. On such dates
any customers would need to source their drugs from
another supplier.
EXPERT WITNESS JOURNAL
© IDMU Ltd 27 June 2025
www.idmu.co.uk
24
AUGUST/SEPT 2025
