In hydromorphologically-degraded lowland streams, large-wood reintroductions are often used to re-establish instream physical structure, which might also increase biodiversity. However, the success rate of this approach varies in terms of positive macroinvertebrate assemblage responses. To obtain better insight into macroinvertebrate–wood relationships, we studied macroinvertebrate assemblage composition and its associated ecological and functional traits in 3 lowland streams in The Netherlands where wood was reintroduced. We used a before–after control–impact design in which we studied stream sections in 3 y: 1 y before and 2 y after large wood was added to some stream sections but not others. We recorded changes in physical structure expressed as substrate diversity, complexity, patchiness, and stability and then compared these parameters within and among the control and treated sections in each stream. We also sampled macroinvertebrates to determine whether the assemblage composition changed because of the wood addition. Finally, we assessed whether changes in macroinvertebrate assemblage could be related to taxa preferences for substrate type and flow and to their functional traits related to mode of locomotion and feeding type. Habitat heterogeneity increased after the wood additions and was relatively stable between years. Macroinvertebrate assemblages changed relative to the control sections in the 2 y after introduction, with 50 to 58% of the taxa increasing or decreasing significantly in abundance. Despite the changes in substrate composition and habitat heterogeneity, most of the functional relationships we expected between macroinvertebrates and large wood were either not apparent or site specific. The only characteristic shared by the macroinvertebrates that consistently increased in response to wood additions was a high affinity for hard substrates. In 1 stream we also observed an increase in taxa with a preference for high-flow velocity and a grazer–scraper feeding mode. These findings suggest that an increase in the surface area of stable, hard substrate was the main underlying ecological effect of reintroducing large wood to the stream channel of sand-bed lowland streams, at least in the short term, and that this change only affected a specific part of the macroinvertebrate assemblage. Changes in assemblage composition occurred primarily during the 1st y after the wood additions and decreased between the 1st and 2nd y, so colonization in this early successional stage seems to be limited to the species pool present in the immediate surroundings.