diff --git a/tools/analysis/data-aggregator/function-occurrences-onwrite.sh b/tools/analysis/data-aggregator/function-occurrences-onwrite.sh
index c4009295..6f2d2669 100755
--- a/tools/analysis/data-aggregator/function-occurrences-onwrite.sh
+++ b/tools/analysis/data-aggregator/function-occurrences-onwrite.sh
@@ -8,6 +8,9 @@ corrupted, and completely disregards storage duration (e.g. time between write
 and subsequent read).  The metric is calculated from the results of a normal
 "inject-on-read" campaign, only using the FI results from memory accesses that
 directly follow a "write" access.
+
+Only works if write equivalence classes are also imported into the database
+(i.e. import-trace is not run with --no-write-ecs).
 EOT
 }
 MUST_FILTER=1
diff --git a/tools/analysis/data-aggregator/global-coverage-onwrite.sh b/tools/analysis/data-aggregator/global-coverage-onwrite.sh
index ba43d368..8c56f9a3 100755
--- a/tools/analysis/data-aggregator/global-coverage-onwrite.sh
+++ b/tools/analysis/data-aggregator/global-coverage-onwrite.sh
@@ -8,6 +8,9 @@ completely disregards storage duration (e.g. time between write and subsequent
 read).  The metric is calculated from the results of a normal "inject-on-read"
 campaign, only using the FI results from memory accesses that directly follow a
 "write" access.
+
+Only works if write equivalence classes are also imported into the database
+(i.e. import-trace is not run with --no-write-ecs).
 EOT
 }
 source $(dirname $0)/fail-analysis-common.inc.sh
diff --git a/tools/analysis/data-aggregator/global-occurrences-onwrite.sh b/tools/analysis/data-aggregator/global-occurrences-onwrite.sh
index ee06d4bd..8f3eae39 100755
--- a/tools/analysis/data-aggregator/global-occurrences-onwrite.sh
+++ b/tools/analysis/data-aggregator/global-occurrences-onwrite.sh
@@ -8,6 +8,9 @@ corrupted, and completely disregards storage duration (e.g. time between write
 and subsequent read).  The metric is calculated from the results of a normal
 "inject-on-read" campaign, only using the FI results from memory accesses that
 directly follow a "write" access.
+
+Only works if write equivalence classes are also imported into the database
+(i.e. import-trace is not run with --no-write-ecs).
 EOT
 }
 source $(dirname $0)/fail-analysis-common.inc.sh
diff --git a/tools/analysis/data-aggregator/symbol-coverage-onwrite.sh b/tools/analysis/data-aggregator/symbol-coverage-onwrite.sh
index 31820e95..d65d470b 100755
--- a/tools/analysis/data-aggregator/symbol-coverage-onwrite.sh
+++ b/tools/analysis/data-aggregator/symbol-coverage-onwrite.sh
@@ -1,5 +1,7 @@
 #!/bin/bash
 
+# TODO implement show_description
+
 source $(dirname $0)/fail-analysis-common.inc.sh
 
 $MYSQL << EOT
diff --git a/tools/analysis/data-aggregator/symbol-occurrences-onwrite.sh b/tools/analysis/data-aggregator/symbol-occurrences-onwrite.sh
index d1716a28..2954548b 100755
--- a/tools/analysis/data-aggregator/symbol-occurrences-onwrite.sh
+++ b/tools/analysis/data-aggregator/symbol-occurrences-onwrite.sh
@@ -8,6 +8,9 @@ corrupted, and completely disregards storage duration (e.g. time between write
 and subsequent read).  The metric is calculated from the results of a normal
 "inject-on-read" campaign, only using the FI results from memory accesses that
 directly follow a "write" access.
+
+Only works if write equivalence classes are also imported into the database
+(i.e. import-trace is not run with --no-write-ecs).
 EOT
 }
 source $(dirname $0)/fail-analysis-common.inc.sh
diff --git a/tools/analysis/data-aggregator/translation-unit-occurrences-onwrite.sh b/tools/analysis/data-aggregator/translation-unit-occurrences-onwrite.sh
index 2d491834..4ca91406 100755
--- a/tools/analysis/data-aggregator/translation-unit-occurrences-onwrite.sh
+++ b/tools/analysis/data-aggregator/translation-unit-occurrences-onwrite.sh
@@ -1,5 +1,7 @@
 #!/bin/bash
 
+# TODO implement show_description
+
 source $(dirname $0)/fail-analysis-common.inc.sh
 
 # This implementation makes sure the write -- not the subsequent memory access