diff --git a/docs/content/design/design.md b/docs/content/design/design.md
index 109ea876b800..e9fa1fe426ee 100644
--- a/docs/content/design/design.md
+++ b/docs/content/design/design.md
@@ -39,7 +39,7 @@ This separation allows each node to only care about what it is best at. By separ
 
 The following diagram shows how queries and data flow through this architecture, and which nodes (and external dependencies, discussed below) are involved:
 
-<img src="../img/druid-dataflow-3.png" width="800"/>
+<img src="../../img/druid-dataflow-3.png" width="800"/>
 
 All nodes can be run in some highly available fashion, either as symmetric peers in a share-nothing cluster or as hot-swap failover nodes.
 
@@ -51,7 +51,7 @@ Aside from these nodes, there are 3 external dependencies to the system:
 
 The following diagram illustrates the cluster's management layer, showing how certain nodes and dependencies help manage the cluster by tracking and exchanging metadata:
 
-<img src="../img/druid-manage-1.png" width="800"/>
+<img src="../../img/druid-manage-1.png" width="800"/>
 
 
 ### Segments and Data Storage
diff --git a/docs/content/design/indexing-service.md b/docs/content/design/indexing-service.md
index 910448df23ca..eac3083bfa06 100644
--- a/docs/content/design/indexing-service.md
+++ b/docs/content/design/indexing-service.md
@@ -13,7 +13,7 @@ Overlords and middle managers may run on the same node or across multiple nodes
 Indexing Service Overview
 -------------------------
 
-![Indexing Service](../img/indexing_service.png "Indexing Service")
+![Indexing Service](../../img/indexing_service.png "Indexing Service")
 
 <!--
 Preamble
diff --git a/docs/content/design/realtime.md b/docs/content/design/realtime.md
index 1cd4049f3230..3f0f5903e74f 100644
--- a/docs/content/design/realtime.md
+++ b/docs/content/design/realtime.md
@@ -19,7 +19,7 @@ Segment Propagation
 
 The segment propagation diagram for real-time data ingestion can be seen below:
 
-![Segment Propagation](../img/segmentPropagation.png "Segment Propagation")
+![Segment Propagation](../../img/segmentPropagation.png "Segment Propagation")
 
 You can read about the various components shown in this diagram under the Architecture section (see the menu on the right). Note that some of the names are now outdated.
 
diff --git a/docs/content/design/segments.md b/docs/content/design/segments.md
index 298d19438cf7..e97618de8366 100644
--- a/docs/content/design/segments.md
+++ b/docs/content/design/segments.md
@@ -29,7 +29,7 @@ for a query.  There are three basic column types: the timestamp
 column, dimension columns, and metric columns, as illustrated in the
 image below:
 
-![Druid column types](../img/druid-column-types.png "Druid Column Types")
+![Druid column types](../../img/druid-column-types.png "Druid Column Types")
 
 The timestamp and metric columns are simple: behind the scenes each of
 these is an array of integer or floating point values compressed with
diff --git a/docs/content/development/integrating-druid-with-other-technologies.md b/docs/content/development/integrating-druid-with-other-technologies.md
index 9f258377c5bb..399b15801b62 100644
--- a/docs/content/development/integrating-druid-with-other-technologies.md
+++ b/docs/content/development/integrating-druid-with-other-technologies.md
@@ -4,4 +4,4 @@ layout: doc_page
 # Integrating Druid With Other Technologies
 This page discusses how we can integrate druid with other technologies. Event streams can be stored in a distributed queue like Kafka, then it can be streamed to a distributed realtime computation system like Twitter Storm / Samza and then it can be feed into Druid via Tranquility plugin. With Tranquility, Middlemanager & Peons will act as a realtime node and they handle realtime queries, segment handoff and realtime indexing.
 
-<img src="../img/druid-production.png" width="800"/>
\ No newline at end of file
+<img src="../../img/druid-production.png" width="800"/>