diff --git a/ecs/example/locust/locustfile.py b/ecs/example/locust/locustfile.py
new file mode 100644
index 00000000..6b75cd98
--- /dev/null
+++ b/ecs/example/locust/locustfile.py
@@ -0,0 +1,29 @@
+import random
+
+from locust import HttpLocust, TaskSet, task, between
+
+
+class UserBehaviour(TaskSet):
+    @task(5)
+    def status_2xx(self):
+        self.client.get("/status/200")
+
+    @task(2)
+    def status_3xx(self):
+        self.client.get("/redirect-to", params=dict(
+            url="/status/200",
+            status_code=302,
+        ))
+
+    @task(2)
+    def status_4xx(self):
+        self.client.get("/status/400")
+
+    @task(1)
+    def status_5xx(self):
+        self.client.get("/status/500")
+
+
+class WebsiteUser(HttpLocust):
+    task_set = UserBehaviour
+    wait_time = between(5, 10)
diff --git a/ecs/example/locust/start b/ecs/example/locust/start
new file mode 100755
index 00000000..dab7b0a1
--- /dev/null
+++ b/ecs/example/locust/start
@@ -0,0 +1,10 @@
+#!/bin/sh
+
+set -eux
+
+docker run \
+  -p 8089:8089 \
+  --volume "$PWD/locust/locustfile.py:/mnt/locust/locustfile.py" \
+  -e LOCUSTFILE_PATH=/mnt/locust/locustfile.py \
+  -e TARGET_URL="$(terraform output lb_url)" \
+  locustio/locust
diff --git a/ecs/example/main.tf b/ecs/example/main.tf
index c0022183..a8213283 100644
--- a/ecs/example/main.tf
+++ b/ecs/example/main.tf
@@ -48,7 +48,8 @@ module "worker_task" {
   environment       = local.environment
   task              = "worker"
   image             = "kennethreitz/httpbin:latest"
-  memory_soft_limit = 128
+  memory_soft_limit = 48
+  cpu               = 128
 
   environment_variables = {
     DEBUG            = "True"
@@ -72,7 +73,8 @@ module "web_task" {
   environment       = local.environment
   task              = "web"
   image             = "kennethreitz/httpbin:latest"
-  memory_soft_limit = 128
+  memory_soft_limit = 48
+  cpu               = 128
   ports             = [80]
 
   environment_variables = {
@@ -87,7 +89,7 @@ module "web" {
   name                = "web"
   cluster_arn         = module.cluster.arn
   task_definition_arn = module.web_task.arn
-  desired_count       = 1
+  desired_count       = 2
 
   vpc_id           = module.cluster.vpc_id
   listener_arn     = module.cluster.http_listener_arn
@@ -96,6 +98,31 @@ module "web" {
   healthcheck_path = "/"
 }
 
+module "dashboard" {
+  source = "./../../cloudwatch/dashboard"
+
+  name = "terraform-ecs-example"
+  widgets = [
+    module.web.widgets.responses,
+    module.web.widgets.response_percentages,
+    module.web.widgets.response_time,
+    module.web.widgets.scaling,
+    module.web.widgets.cpu_utilization,
+    module.web.widgets.memory_utilization,
+    module.worker.widgets.scaling,
+    module.worker.widgets.cpu_utilization,
+    module.worker.widgets.memory_utilization,
+  ]
+}
+
 output "hosts_id" {
   value = module.hosts.id
 }
+
+output "dashboard_url" {
+  value = module.dashboard.url
+}
+
+output "lb_url" {
+  value = "http://${module.cluster.load_balancer_domain}"
+}
diff --git a/ecs/main.tf b/ecs/main.tf
index dfb64fd4..4b1bfeed 100644
--- a/ecs/main.tf
+++ b/ecs/main.tf
@@ -23,5 +23,10 @@ resource "aws_ecs_cluster" "cluster" {
   count = var.create ? 1 : 0
 
   name = "${var.project}-${var.environment}"
+
+  setting {
+    name  = "containerInsights"
+    value = "enabled"
+  }
 }
 
diff --git a/ecs/services/web/README.md b/ecs/services/web/README.md
index a09e21bf..0fd7e847 100644
--- a/ecs/services/web/README.md
+++ b/ecs/services/web/README.md
@@ -117,6 +117,10 @@ Creates an ECS service exposed to the internet using an Application Load Balance
 
     Service id
 
+* `metrics`
+
+    Cloudwatch metrics, see [metrics.tf](./metrics.tf)
+
 * `target_group_arn`
 
     Load balancer target group ARN
@@ -124,3 +128,7 @@ Creates an ECS service exposed to the internet using an Application Load Balance
 * `target_group_name`
 
     Load balancer target group name
+
+* `widgets`
+
+    Cloudwatch dashboard widgets, see [widgets.tf](./widgets.tf)
diff --git a/ecs/services/web/main.tf b/ecs/services/web/main.tf
index c32b12ca..fbef48c3 100644
--- a/ecs/services/web/main.tf
+++ b/ecs/services/web/main.tf
@@ -86,4 +86,3 @@ resource "aws_lb_listener_rule" "service" {
     values = [var.rule_path]
   }
 }
-
diff --git a/ecs/services/web/metrics.tf b/ecs/services/web/metrics.tf
new file mode 100644
index 00000000..d5cebe0a
--- /dev/null
+++ b/ecs/services/web/metrics.tf
@@ -0,0 +1,276 @@
+locals {
+  metrics = {
+    requests                       = module.metric_requests
+    status_2xx_responses           = module.metrics_response_statuses.out_map["2xx"]
+    status_2xx_response_percentage = module.metrics_response_status_percentages.out_map["2xx"]
+    status_3xx_responses           = module.metrics_response_statuses.out_map["3xx"]
+    status_3xx_response_percentage = module.metrics_response_status_percentages.out_map["3xx"]
+    status_4xx_responses           = module.metrics_response_statuses.out_map["4xx"]
+    status_4xx_response_percentage = module.metrics_response_status_percentages.out_map["4xx"]
+    status_5xx_responses           = module.metrics_response_statuses.out_map["5xx"]
+    status_5xx_response_percentage = module.metrics_response_status_percentages.out_map["5xx"]
+    connection_errors              = module.metric_connection_errors
+    connection_error_percentage    = module.metric_connection_error_percentage
+    average_response_time          = module.metrics_response_time.out_map.average
+    p50_response_time              = module.metrics_response_time.out_map.p50
+    p90_response_time              = module.metrics_response_time.out_map.p90
+    p95_response_time              = module.metrics_response_time.out_map.p95
+    p99_response_time              = module.metrics_response_time.out_map.p99
+    max_response_time              = module.metrics_response_time.out_map.max
+    desired_tasks                  = module.metrics_tasks.out_map.desired
+    pending_tasks                  = module.metrics_tasks.out_map.pending
+    running_tasks                  = module.metrics_tasks.out_map.running
+    healthy_tasks                  = module.metric_healthy_tasks
+    average_cpu_reservation        = module.metric_average_cpu_reservation
+    min_cpu_utilization            = module.metrics_cpu_utilization.out_map.min
+    average_cpu_utilization        = module.metrics_cpu_utilization.out_map.average
+    max_cpu_utilization            = module.metrics_cpu_utilization.out_map.max
+    average_memory_reservation     = module.metric_average_memory_reservation
+    min_memory_utilization         = module.metrics_memory_utilization.out_map.min
+    average_memory_utilization     = module.metrics_memory_utilization.out_map.average
+    max_memory_utilization         = module.metrics_memory_utilization.out_map.max
+  }
+}
+
+data "aws_lb_listener" "listener" {
+  count = var.create ? 1 : 0
+
+  arn = var.listener_arn
+}
+
+data "aws_lb" "lb" {
+  count = var.create ? 1 : 0
+
+  arn = data.aws_lb_listener.listener[0].load_balancer_arn
+}
+
+module "cloudwatch_consts" {
+  source = "./../../../cloudwatch/consts"
+}
+
+locals {
+  colors = module.cloudwatch_consts.colors
+}
+
+module "metric_requests" {
+  source = "./../../../cloudwatch/metric"
+
+  namespace = "AWS/ApplicationELB"
+  name      = "RequestCount"
+  label     = "Responses"
+  stat      = "Sum"
+  period    = 60
+
+  dimensions = {
+    LoadBalancer = var.create ? data.aws_lb.lb[0].arn_suffix : ""
+    TargetGroup  = var.create ? aws_lb_target_group.service[0].arn_suffix : ""
+  }
+}
+
+locals {
+  metrics_response_statuses = {
+    "2xx" = { color = local.colors.green }
+    "3xx" = { color = local.colors.blue }
+    "4xx" = { color = local.colors.orange }
+    "5xx" = { color = local.colors.red }
+  }
+}
+
+module "metrics_response_statuses" {
+  source = "./../../../cloudwatch/metric/many"
+
+  vars_map = { for status, variant in local.metrics_response_statuses : status => {
+    namespace = "AWS/ApplicationELB"
+    name      = "HTTPCode_Target_${upper(status)}_Count"
+    label     = "${status} responses"
+    color     = variant.color
+    stat      = "Sum"
+    period    = 60
+
+    dimensions = {
+      LoadBalancer = var.create ? data.aws_lb.lb[0].arn_suffix : ""
+      TargetGroup  = var.create ? aws_lb_target_group.service[0].arn_suffix : ""
+    }
+  } }
+}
+
+module "metrics_response_status_percentages" {
+  source = "./../../../cloudwatch/metric_expression/many"
+
+  vars_map = { for status, variant in local.metrics_response_statuses : status => {
+    expression = "IF(${module.metric_requests.id} == 0, 0, FILL(${module.metrics_response_statuses.out_map[status].id}, 0) / ${module.metric_requests.id} * 100)"
+    label      = "${status} responses"
+    color      = variant.color
+  } }
+}
+
+module "metric_connection_errors" {
+  source = "./../../../cloudwatch/metric"
+
+  namespace = "AWS/ApplicationELB"
+  name      = "TargetConnectionErrorCount"
+  label     = "Connection errors"
+  color     = local.colors.purple
+  stat      = "Sum"
+  period    = 60
+
+  dimensions = {
+    LoadBalancer = var.create ? data.aws_lb.lb[0].arn_suffix : ""
+    TargetGroup  = var.create ? aws_lb_target_group.service[0].arn_suffix : ""
+  }
+}
+
+module "metric_connection_error_percentage" {
+  source = "./../../../cloudwatch/metric_expression"
+
+  expression = "IF(${module.metric_requests.id} == 0, 0, FILL(${module.metric_connection_errors.id}, 0) / ${module.metric_requests.id} * 100)"
+  label      = "Connection errors"
+  color      = module.metric_connection_errors.color
+}
+
+locals {
+  metrics_response_time_variants = {
+    average = { stat = "Average", color = local.colors.red }
+    p50     = { stat = "p50", color = local.colors.red }
+    p90     = { stat = "p90", color = local.colors.orange }
+    p95     = { stat = "p95", color = local.colors.orange }
+    p99     = { stat = "p99", color = local.colors.light_red }
+    max     = { stat = "Maximum", color = local.colors.light_orange }
+  }
+}
+
+module "metrics_response_time" {
+  source = "./../../../cloudwatch/metric/many"
+
+  vars_map = { for k, variant in local.metrics_response_time_variants : k => {
+    namespace = "AWS/ApplicationELB"
+    name      = "TargetResponseTime"
+    label     = "${variant.stat} response time"
+    color     = variant.color
+    stat      = variant.stat
+    period    = 60
+
+    dimensions = {
+      LoadBalancer = var.create ? data.aws_lb.lb[0].arn_suffix : ""
+      TargetGroup  = var.create ? aws_lb_target_group.service[0].arn_suffix : ""
+    }
+  } }
+}
+
+locals {
+  metrics_tasks_variants = {
+    desired = { state = "Desired", color = local.colors.grey }
+    pending = { state = "Pending", color = local.colors.orange }
+    running = { state = "Running", color = local.colors.light_green }
+  }
+}
+
+module "metrics_tasks" {
+  source = "./../../../cloudwatch/metric/many"
+
+  vars_map = { for k, variant in local.metrics_tasks_variants : k => {
+    namespace = "ECS/ContainerInsights"
+    name      = "${variant.state}TaskCount"
+    label     = "${variant.state} task count"
+    color     = variant.color
+    stat      = "Average"
+    period    = 60
+
+    dimensions = {
+      ServiceName = var.name
+      ClusterName = local.cluster_name
+    }
+  } }
+}
+
+module "metric_healthy_tasks" {
+  source = "./../../../cloudwatch/metric"
+
+  namespace = "AWS/ApplicationELB"
+  name      = "HealthyHostCount"
+  label     = "Healthy task count"
+  color     = local.colors.green
+  stat      = "Average"
+  period    = 60
+
+  dimensions = {
+    LoadBalancer = var.create ? data.aws_lb.lb[0].arn_suffix : ""
+    TargetGroup  = var.create ? aws_lb_target_group.service[0].arn_suffix : ""
+  }
+}
+
+module "metric_average_cpu_reservation" {
+  source = "./../../../cloudwatch/metric"
+
+  namespace = "ECS/ContainerInsights"
+  name      = "CpuReserved"
+  label     = "Average CPU reserved"
+  color     = local.colors.grey
+  stat      = "Average"
+  period    = 60
+
+  dimensions = {
+    ServiceName = var.name
+    ClusterName = local.cluster_name
+  }
+}
+
+locals {
+  metrics_utilization_variants = {
+    min     = { stat = "Minimum", color = local.colors.light_orange }
+    average = { stat = "Average", color = local.colors.orange }
+    max     = { stat = "Maximum", color = local.colors.red }
+  }
+}
+
+module "metrics_cpu_utilization" {
+  source = "./../../../cloudwatch/metric/many"
+
+  vars_map = { for k, variant in local.metrics_utilization_variants : k => {
+    namespace = "ECS/ContainerInsights"
+    name      = "CpuUtilized"
+    label     = "${variant.stat} CPU utilized"
+    color     = variant.color
+    stat      = variant.stat
+    period    = 60
+
+    dimensions = {
+      ServiceName = var.name
+      ClusterName = local.cluster_name
+    }
+  } }
+}
+
+module "metric_average_memory_reservation" {
+  source = "./../../../cloudwatch/metric"
+
+  namespace = "ECS/ContainerInsights"
+  name      = "MemoryReserved"
+  label     = "Average memory reserved"
+  color     = local.colors.grey
+  stat      = "Average"
+  period    = 60
+
+  dimensions = {
+    ServiceName = var.name
+    ClusterName = local.cluster_name
+  }
+}
+
+module "metrics_memory_utilization" {
+  source = "./../../../cloudwatch/metric/many"
+
+  vars_map = { for k, variant in local.metrics_utilization_variants : k => {
+    namespace = "ECS/ContainerInsights"
+    name      = "MemoryUtilized"
+    label     = "${variant.stat} memory utilized"
+    color     = variant.color
+    stat      = variant.stat
+    period    = 60
+
+    dimensions = {
+      ServiceName = var.name
+      ClusterName = local.cluster_name
+    }
+  } }
+}
diff --git a/ecs/services/web/outputs.tf b/ecs/services/web/outputs.tf
index 2dae7430..42008b6b 100644
--- a/ecs/services/web/outputs.tf
+++ b/ecs/services/web/outputs.tf
@@ -13,3 +13,12 @@ output "target_group_arn" {
   value       = var.create ? aws_lb_target_group.service[0].arn : null
 }
 
+output "metrics" {
+  description = "Cloudwatch metrics, see [metrics.tf](./metrics.tf)"
+  value       = local.metrics
+}
+
+output "widgets" {
+  description = "Cloudwatch dashboard widgets, see [widgets.tf](./widgets.tf)"
+  value       = local.widgets
+}
diff --git a/ecs/services/web/widgets.tf b/ecs/services/web/widgets.tf
new file mode 100644
index 00000000..2961171d
--- /dev/null
+++ b/ecs/services/web/widgets.tf
@@ -0,0 +1,99 @@
+locals {
+  widgets = {
+    responses            = module.widget_responses
+    response_percentages = module.widget_response_percentages
+    response_time        = module.widget_response_time
+    scaling              = module.widget_scaling
+    cpu_utilization      = module.widget_cpu_utilization
+    memory_utilization   = module.widget_memory_utilization
+  }
+}
+
+module "widget_responses" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title   = "${var.name} service responses"
+  stacked = true
+  left_metrics = [
+    local.metrics.connection_errors,
+    local.metrics.status_5xx_responses,
+    local.metrics.status_4xx_responses,
+    local.metrics.status_3xx_responses,
+    local.metrics.status_2xx_responses,
+  ]
+}
+
+module "widget_response_percentages" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title   = "${var.name} service response percentages"
+  stacked = true
+  left_metrics = [
+    local.metrics.connection_error_percentage,
+    local.metrics.status_5xx_response_percentage,
+    local.metrics.status_4xx_response_percentage,
+    local.metrics.status_3xx_response_percentage,
+    local.metrics.status_2xx_response_percentage,
+  ]
+  left_range = [0, 100]
+  hidden_metrics = [
+    local.metrics.requests,
+    local.metrics.status_2xx_responses,
+    local.metrics.status_3xx_responses,
+    local.metrics.status_4xx_responses,
+    local.metrics.status_5xx_responses,
+    local.metrics.connection_errors,
+  ]
+}
+
+module "widget_response_time" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title   = "${var.name} service response times"
+  stacked = true
+  left_metrics = [
+    local.metrics.p50_response_time,
+    local.metrics.p95_response_time,
+    local.metrics.p99_response_time,
+    local.metrics.max_response_time,
+  ]
+}
+
+module "widget_scaling" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title = "${var.name} service scaling"
+  left_metrics = [
+    local.metrics.desired_tasks,
+    local.metrics.pending_tasks,
+    local.metrics.running_tasks,
+    local.metrics.healthy_tasks,
+  ]
+  left_range = [0, null]
+}
+
+module "widget_cpu_utilization" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title = "${var.name} service CPU utilization"
+  left_metrics = [
+    local.metrics.average_cpu_reservation,
+    local.metrics.min_cpu_utilization,
+    local.metrics.average_cpu_utilization,
+    local.metrics.max_cpu_utilization,
+  ]
+  left_range = [0, null]
+}
+
+module "widget_memory_utilization" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title = "${var.name} service memory utilization"
+  left_metrics = [
+    local.metrics.average_memory_reservation,
+    local.metrics.min_memory_utilization,
+    local.metrics.average_memory_utilization,
+    local.metrics.max_memory_utilization,
+  ]
+  left_range = [0, null]
+}
diff --git a/ecs/services/worker/README.md b/ecs/services/worker/README.md
index 188b66f4..d4413e7a 100644
--- a/ecs/services/worker/README.md
+++ b/ecs/services/worker/README.md
@@ -52,3 +52,11 @@ Creates an ECS service for background workers
 * `id`
 
     Service id
+
+* `metrics`
+
+    Cloudwatch metrics, see [metrics.tf](./metrics.tf)
+
+* `widgets`
+
+    Cloudwatch dashboard widgets, see [widgets.tf](./widgets.tf)
diff --git a/ecs/services/worker/main.tf b/ecs/services/worker/main.tf
index 0a349946..0bce2830 100644
--- a/ecs/services/worker/main.tf
+++ b/ecs/services/worker/main.tf
@@ -1,3 +1,13 @@
+locals {
+  cluster_name = var.create ? substr(data.aws_arn.cluster[0].resource, length("cluster/"), -1) : ""
+}
+
+data "aws_arn" "cluster" {
+  count = var.create ? 1 : 0
+
+  arn = var.cluster_arn
+}
+
 resource "aws_ecs_service" "service" {
   count = var.create ? 1 : 0
 
diff --git a/ecs/services/worker/metrics.tf b/ecs/services/worker/metrics.tf
new file mode 100644
index 00000000..d566868d
--- /dev/null
+++ b/ecs/services/worker/metrics.tf
@@ -0,0 +1,125 @@
+locals {
+  metrics = {
+    desired_tasks              = module.metrics_tasks.out_map.desired
+    pending_tasks              = module.metrics_tasks.out_map.pending
+    running_tasks              = module.metrics_tasks.out_map.running
+    average_cpu_reservation    = module.metric_average_cpu_reservation
+    min_cpu_utilization        = module.metrics_cpu_utilization.out_map.min
+    average_cpu_utilization    = module.metrics_cpu_utilization.out_map.average
+    max_cpu_utilization        = module.metrics_cpu_utilization.out_map.max
+    average_memory_reservation = module.metric_average_memory_reservation
+    min_memory_utilization     = module.metrics_memory_utilization.out_map.min
+    average_memory_utilization = module.metrics_memory_utilization.out_map.average
+    max_memory_utilization     = module.metrics_memory_utilization.out_map.max
+  }
+}
+
+module "cloudwatch_consts" {
+  source = "./../../../cloudwatch/consts"
+}
+
+locals {
+  colors = module.cloudwatch_consts.colors
+}
+
+locals {
+  metrics_tasks_variants = {
+    desired = { state = "Desired", color = local.colors.grey }
+    pending = { state = "Pending", color = local.colors.orange }
+    running = { state = "Running", color = local.colors.green }
+  }
+}
+
+module "metrics_tasks" {
+  source = "./../../../cloudwatch/metric/many"
+
+  vars_map = { for k, variant in local.metrics_tasks_variants : k => {
+    namespace = "ECS/ContainerInsights"
+    name      = "${variant.state}TaskCount"
+    label     = "${variant.state} task count"
+    color     = variant.color
+    stat      = "Average"
+    period    = 60
+
+    dimensions = {
+      ServiceName = var.name
+      ClusterName = local.cluster_name
+    }
+  } }
+}
+
+module "metric_average_cpu_reservation" {
+  source = "./../../../cloudwatch/metric"
+
+  namespace = "ECS/ContainerInsights"
+  name      = "CpuReserved"
+  label     = "Average CPU reserved"
+  color     = local.colors.grey
+  stat      = "Average"
+  period    = 60
+
+  dimensions = {
+    ServiceName = var.name
+    ClusterName = local.cluster_name
+  }
+}
+
+locals {
+  metrics_utilization_variants = {
+    min     = { stat = "Minimum", color = local.colors.light_orange }
+    average = { stat = "Average", color = local.colors.orange }
+    max     = { stat = "Maximum", color = local.colors.red }
+  }
+}
+
+module "metrics_cpu_utilization" {
+  source = "./../../../cloudwatch/metric/many"
+
+  vars_map = { for k, variant in local.metrics_utilization_variants : k => {
+    namespace = "ECS/ContainerInsights"
+    name      = "CpuUtilized"
+    label     = "${variant.stat} CPU utilized"
+    color     = variant.color
+    stat      = variant.stat
+    period    = 60
+
+    dimensions = {
+      ServiceName = var.name
+      ClusterName = local.cluster_name
+    }
+  } }
+}
+
+module "metric_average_memory_reservation" {
+  source = "./../../../cloudwatch/metric"
+
+  namespace = "ECS/ContainerInsights"
+  name      = "MemoryReserved"
+  label     = "Average memory reserved"
+  color     = local.colors.grey
+  stat      = "Average"
+  period    = 60
+
+  dimensions = {
+    ServiceName = var.name
+    ClusterName = local.cluster_name
+  }
+}
+
+module "metrics_memory_utilization" {
+  source = "./../../../cloudwatch/metric/many"
+
+  vars_map = { for k, variant in local.metrics_utilization_variants : k => {
+    namespace = "ECS/ContainerInsights"
+    name      = "MemoryUtilized"
+    label     = "${variant.stat} memory utilized"
+    color     = variant.color
+    stat      = variant.stat
+    period    = 60
+
+    dimensions = {
+      ServiceName = var.name
+      ClusterName = local.cluster_name
+    }
+  } }
+}
diff --git a/ecs/services/worker/outputs.tf b/ecs/services/worker/outputs.tf
index 4cd50a6f..bd3b7656 100644
--- a/ecs/services/worker/outputs.tf
+++ b/ecs/services/worker/outputs.tf
@@ -3,3 +3,12 @@ output "id" {
   value       = var.create ? aws_ecs_service.service[0].id : null
 }
 
+output "metrics" {
+  description = "Cloudwatch metrics, see [metrics.tf](./metrics.tf)"
+  value       = local.metrics
+}
+
+output "widgets" {
+  description = "Cloudwatch dashboard widgets, see [widgets.tf](./widgets.tf)"
+  value       = local.widgets
+}
diff --git a/ecs/services/worker/widgets.tf b/ecs/services/worker/widgets.tf
new file mode 100644
index 00000000..3ac84172
--- /dev/null
+++ b/ecs/services/worker/widgets.tf
@@ -0,0 +1,45 @@
+locals {
+  widgets = {
+    scaling            = module.widget_scaling
+    cpu_utilization    = module.widget_cpu_utilization
+    memory_utilization = module.widget_memory_utilization
+  }
+}
+
+module "widget_scaling" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title = "${var.name} service scaling"
+  left_metrics = [
+    local.metrics.desired_tasks,
+    local.metrics.pending_tasks,
+    local.metrics.running_tasks,
+  ]
+  left_range = [0, null]
+}
+
+module "widget_cpu_utilization" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title = "${var.name} service CPU utilization"
+  left_metrics = [
+    local.metrics.average_cpu_reservation,
+    local.metrics.min_cpu_utilization,
+    local.metrics.average_cpu_utilization,
+    local.metrics.max_cpu_utilization,
+  ]
+  left_range = [0, null]
+}
+
+module "widget_memory_utilization" {
+  source = "./../../../cloudwatch/metric_widget"
+
+  title = "${var.name} service memory utilization"
+  left_metrics = [
+    local.metrics.average_memory_reservation,
+    local.metrics.min_memory_utilization,
+    local.metrics.average_memory_utilization,
+    local.metrics.max_memory_utilization,
+  ]
+  left_range = [0, null]
+}