AbstractPhenological changes have been observed globally for marine, freshwater and terrestrial species, and are an important element of the global biological ‘fingerprint’ of climate change. Differences in rates of change could desynchronize seasonal species interactions within a food web, threatening ecosystem functioning. Quantification of this risk is hampered by the rarity of long‐term data for multiple interacting species from the same ecosystem and by the diversity of possible phenological metrics, which vary in their ecological relevance to food web interactions. We compare phenological change for phytoplankton (chlorophyll a), zooplankton (Daphnia) and fish (perch, Perca fluviatilis) in two basins of Windermere over 40 years and determine whether change has differed among trophic levels, while explicitly accounting for among‐metric differences in rates of change. Though rates of change differed markedly among the nine metrics used, seasonal events shifted earlier for all metrics and trophic levels: zooplankton advanced most, and fish least, rapidly. Evidence of altered synchrony was found in both lake basins, when combining information from all phenological metrics. However, comparisons based on single metrics did not consistently detect this signal. A multimetric approach showed that across trophic levels, earlier phenological events have been associated with increasing water temperature. However, for phytoplankton and zooplankton, phenological change was also associated with changes in resource availability. Lower silicate, and higher phosphorus, concentrations were associated with earlier phytoplankton growth, and earlier phytoplankton growth was associated with earlier zooplankton growth. The developing trophic mismatch detected between the dominant fish species in Windermere and important zooplankton food resources may ultimately affect fish survival and portend significant impacts upon ecosystem functioning. We advocate that future studies on phenological synchrony combine data from multiple phenological metrics, to increase confidence in assessments of change and likely ecological consequences.