Reading the original query sheet

September 2020 queries · review note added 1 October 2026. These are retained query texts supporting the pictured walkthrough, not rerun results or a checked contemporary Neo4j tutorial.

The patterns show how a question selects connected records. They also show why the unit of a result matters: count(i) and count(t.Description) count matched rows/non-null values, not necessarily distinct investments or programmes. Multiple paths can repeat the same entity. The supplier comparison requires identifier/name reconciliation; the start-delay query groups activity observations by business function, not independently established “programme types”. The metric query uses Project_delay > 100, so its threshold is strictly more than 100 days.

The spend query uses the historical property name Enhancement_spend_$m without identifier quoting, converts values to integers and reports no currency conversion. Its original £82m claim is therefore not a verified monetary result. Inspect types, property names, distinct identifiers and missing records before any future reproduction. No queries have been executed during this review.

2026 terminology references: Neo4j aggregation and duplicate counting and identifier naming rules. These clarify the reading notes; the historical query text has not been modernised.

Original query sheet, retained unchanged

Cypher queries

These are the queries used to return the results in this note.

How many programmes are dependent on our largest single supplier ?

MATCH (s:Supplier)<-[r:uses_Supplier]-(i:Investment) WHERE i.name IS NOT null RETURN s.name,count(i) ORDER BY count (i) DESCENDING

MATCH (s:Supplier)<-[r:uses_Supplier]-(i:Investment) WHERE s.name=’INTERNATIONAL BUSINESS MACHINES CORPORATION’ AND i.name IS NOT NULL OPTIONAL MATCH (i)<–(j:Bureau) OPTIONAL MATCH (i)<–(k:Agency) RETURN s,i,j,k

What metrics are being used to control these particular programmes?

MATCH (s:Supplier)<-[r:uses_Supplier]-(i:Investment)–>(m:Metric) WHERE s.name=’INTERNATIONAL BUSINESS MACHINES CORPORATION’ AND i.name IS NOT NULL RETURN m.Measurement_category,count(m.Description) ORDER BY count(m.Description) DESCENDING

MATCH (m:Metric) WHERE m.Metric_results=”Met” RETURN m.Measurement_category,COUNT(m.Description) ORDER BY m.Measurement_category

** Which programmes are given low scores by clients?**

MATCH (m:Investment)–(n:Service) WHERE (m.Evaluation_by_CIO=’1’OR m.Evaluation_by_CIO=’2’) AND m.name IS NOT NULL RETURN m,n

** How much money is being spent on similar HR programmes ?**

MATCH (s:Service)–(i:Investment) WHERE s.name=’756-Human Capital / Workforce Management’ AND i.name IS NOT null RETURN s,i

MATCH (s:Service)–(i:Investment) WHERE s.name=’756-Human Capital / Workforce Management’ AND i.name IS NOT null RETURN sum(toInteger(i.Enhancement_spend_$m))

MATCH (s:Service)–(i:Investment)–(p:Project) WHERE s.name=’756-Human Capital / Workforce Management’ AND i.name IS NOT null AND p.Project_delay>0 RETURN i,s,p

How much delay to projects on these programmes?

MATCH (s:Service)–(i:Investment)–(p:Project) WHERE s.name=’756-Human Capital / Workforce Management’ AND i.name IS NOT null RETURN i,s,p

MATCH (s:Service)–(i:Investment)–(p:Project) WHERE s.name=’756-Human Capital / Workforce Management’ AND i.name IS NOT null AND p.Project_delay>0 RETURN i,s,p

MATCH (s:Service)–(i:Investment)–(p:Project) WHERE s.name=’756-Human Capital / Workforce Management’ AND i.name IS NOT null AND p.Project_delay>0 RETURN sum(p.Project_delay)

22 activities listed for these delayed projects

MATCH (s:Service)–(i:Investment)–(p:Project)–(a:Activity) WHERE s.name=’756-Human Capital / Workforce Management’ AND i.name IS NOT null AND p.Project_delay>0 RETURN a.output_type

MATCH (s:Service)–(i:Investment)–(p:Project)–(a:Activity) WHERE s.name=’756-Human Capital / Workforce Management’ AND i.name IS NOT null AND p.Project_delay>0 RETURN avg(Tointeger(replace((a.Actual_start_delay),” days”,””)))

Which programmes start late?

MATCH (a:Activity) RETURN avg(Tointeger(replace((a.Actual_start_delay),” days”,””)))

MATCH (b:Business_Function)<–(i:Investment)–>(p:Project)–>(a:Activity) WITH b,i,p,a,Tointeger(replace((a.Actual_start_delay),” days”,””)) AS x RETURN b.name AS Type of programme,Tointeger(avg(x)) AS Average start delay (days) ORDER BY Tointeger(avg(x)) DESC

MATCH (b:Business_Function)<–(i:Investment)–>(p:Project)–>(a:Activity) WITH b,i,p,a,Tointeger(replace((a.Actual_start_delay),” days”,””)) AS x WHERE b.name=’301-Controls and Oversight’AND i.name IS NOT null RETURN b,i,p,a

Business functions for one particular service

MATCH (m:Service)–(i:Investment)–(o:Business_Function) WHERE m.name=’743-Knowledge Discovery’ AND i.name IS NOT null RETURN m,i,o

What metrics might help with very late projects?

MATCH (m:Project)–(:Investment)–(t:Metric) WHERE m.Project_delay>100 AND t.Metric_results=’Not Met’ RETURN t.Measurement_category,count(t.Description) ORDER BY count(t.Description) DESC

the work of 1 department

MATCH (p:Investment)<-[:responsible_for]-(c:Bureau) WHERE c.Name=’Food and Drug Administration’ OPTIONAL MATCH (p)-[:pays_for]->(n:Project) OPTIONAL MATCH (n)-[:has_task]->(m:Activity) OPTIONAL MATCH (p)-[:judged_by]->(v:Metric) OPTIONAL MATCH (p)-[:lets_contract]->(w:Contract) OPTIONAL MATCH (p)-[:uses_Supplier]->(a:Supplier) OPTIONAL MATCH (p)-[:is_a_type_of]->(b:Service) OPTIONAL MATCH (p)-[:has_business_function]->(e:Business_Function)-[:serves]->(d:Business_Area) RETURN p,c,n,m,v,w,a,b,d,e