diff --git a/cluster-autoscaler/loop/trigger.go b/cluster-autoscaler/loop/trigger.go
index 85d6e14b2237..093773cb9d0f 100644
--- a/cluster-autoscaler/loop/trigger.go
+++ b/cluster-autoscaler/loop/trigger.go
@@ -43,8 +43,12 @@ type scalingTimesGetter interface {
 	LastScaleDownDeleteTime() time.Time
 }
 
-type provisioningRequestProcessTimeGetter interface {
-	LastProvisioningRequestProcessedTime() time.Time
+// provisioningRequestProcessingTimesGetter exposes recent provisioning request processing activity regardless of wether the
+// ProvisioningRequest was marked as accepted or failed. This is because a ProvisioningRequest being processed indicates that
+// there are other ProvisioningRequests that require processing regardless of the outcome of the current one. Thus, the next iteration
+// should be started immediately.
+type provisioningRequestProcessingTimesGetter interface {
+	LastProvisioningRequestProcessTime() time.Time
 }
 
 // LoopTrigger object implements criteria used to start new autoscaling iteration
@@ -52,11 +56,11 @@ type LoopTrigger struct {
 	podObserver                          *UnschedulablePodObserver
 	scanInterval                         time.Duration
 	scalingTimesGetter                   scalingTimesGetter
-	provisioningRequestProcessTimeGetter provisioningRequestProcessTimeGetter
+	provisioningRequestProcessTimeGetter provisioningRequestProcessingTimesGetter
 }
 
 // NewLoopTrigger creates a LoopTrigger object
-func NewLoopTrigger(podObserver *UnschedulablePodObserver, scalingTimesGetter scalingTimesGetter, scanInterval time.Duration, provisioningRequestProcessTimeGetter provisioningRequestProcessTimeGetter) *LoopTrigger {
+func NewLoopTrigger(podObserver *UnschedulablePodObserver, scalingTimesGetter scalingTimesGetter, provisioningRequestProcessTimeGetter provisioningRequestProcessingTimesGetter, scanInterval time.Duration) *LoopTrigger {
 	return &LoopTrigger{
 		podObserver:                          podObserver,
 		scanInterval:                         scanInterval,
@@ -74,7 +78,7 @@ func (t *LoopTrigger) Wait(lastRun time.Time) {
 	// immediately if the previous one was productive.
 	if !t.scalingTimesGetter.LastScaleUpTime().Before(lastRun) ||
 		!t.scalingTimesGetter.LastScaleDownDeleteTime().Before(lastRun) ||
-		!t.provisioningRequestProcessTimeGetter.LastProvisioningRequestProcessedTime().Before(lastRun) {
+		!t.provisioningRequestProcessTimeGetter.LastProvisioningRequestProcessTime().Before(lastRun) {
 		select {
 		case <-t.podObserver.unschedulablePodChan:
 			klog.Info("Autoscaler loop triggered by unschedulable pod appearing")
diff --git a/cluster-autoscaler/main.go b/cluster-autoscaler/main.go
index 3d8b57b890bf..8ce451705315 100644
--- a/cluster-autoscaler/main.go
+++ b/cluster-autoscaler/main.go
@@ -633,7 +633,10 @@ func run(healthCheck *metrics.HealthCheck, debuggingSnapshotter debuggingsnapsho
 	defer cancel()
 	if *frequentLoopsEnabled {
 		podObserver := loop.StartPodObserver(context, kube_util.CreateKubeClient(createAutoscalingOptions().KubeClientOpts))
-		trigger := loop.NewLoopTrigger(podObserver, autoscaler, *scanInterval, injector)
+		// A ProvisioningRequestPodsInjector is used as provisioningRequestProcessingTimesGetter here to obtain the last time a
+		// ProvisioningRequest was processed. This is because the ProvisioningRequestPodsInjector in addition to injecting pods
+		// also marks the ProvisioningRequest as accepted or failed.
+		trigger := loop.NewLoopTrigger(podObserver, autoscaler, injector, *scanInterval)
 		lastRun := time.Now()
 		for {
 			trigger.Wait(lastRun)
diff --git a/cluster-autoscaler/processors/provreq/injector.go b/cluster-autoscaler/processors/provreq/injector.go
index b4718c1ec5d7..94eb0a3f2ea7 100644
--- a/cluster-autoscaler/processors/provreq/injector.go
+++ b/cluster-autoscaler/processors/provreq/injector.go
@@ -24,7 +24,6 @@ import (
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	"k8s.io/autoscaler/cluster-autoscaler/apis/provisioningrequest/autoscaling.x-k8s.io/v1"
 	"k8s.io/autoscaler/cluster-autoscaler/context"
-	// "k8s.io/autoscaler/cluster-autoscaler/processors/pods"
 	"k8s.io/autoscaler/cluster-autoscaler/provisioningrequest"
 	provreqconditions "k8s.io/autoscaler/cluster-autoscaler/provisioningrequest/conditions"
 	provreqpods "k8s.io/autoscaler/cluster-autoscaler/provisioningrequest/pods"
@@ -147,7 +146,7 @@ func NewProvisioningRequestPodsInjector(kubeConfig *rest.Config) (*ProvisioningR
 	return &ProvisioningRequestPodsInjector{client: client, clock: clock.RealClock{}}, nil
 }
 
-// LastProvisioningRequestProcessedTime returns the time when the last provisioning request was processed.
-func (p *ProvisioningRequestPodsInjector) LastProvisioningRequestProcessedTime() time.Time {
+// LastProvisioningRequestProcessTime returns the time when the last provisioning request was processed.
+func (p *ProvisioningRequestPodsInjector) LastProvisioningRequestProcessTime() time.Time {
 	return p.lastProvisioningRequestProcessedTime
 }