You have connected two streams of data to a FeatureJoiner and are joining them using a common key attribute. After writing out the data, you notice your output looks incorrect with no errors or warnings. Where might you look first to troubleshoot?

• A. Open the written data in its native application.
• B. The Translation Log
• C. Check the FeatureJoiner output port feature counts. You might have chosen the wrong Join Mode or connected the wrong port.
• D. FME Options

Answer: (C)

The thing to check first is the way the join results are flowing through the transformer. The FeatureJoiner has multiple output paths, and the counts on those output ports show exactly how many features are going where. This quick snapshot tells you whether any matches occurred and whether you wired to the correct port. If you’ve chosen a join mode that doesn’t produce the expected matches (for example, mixing one-to-one with a data source that should be one-to-many) or you connected to a non-joined output, the feature counts will reveal it right away—often showing most features on an “unmatched” port and few or none on the “joined” port. That’s your first clue to fix.

From there, you can confirm by inspecting a few features on the joined port to verify keys and values, or adjust the Join Mode and key fields accordingly. Opening the data in a native application or tweaking FME Options won’t pinpoint the misconfiguration as directly as the per-port counts will.

So, checking the FeatureJoiner output port feature counts is the best first step because it directly exposes how the data is being routed after the join and whether the configuration matches the data realities.