This dataset result from an experiment in an urban area near Paris (France) to test whether pollinators present in an urban environment contributed to the production of strawberries (Fragaria × ananassa). We performed flower-visitor observations and pollination experiments on strawberries, Fragaria × ananassa, in an urban area near Paris, France, in order to assess the effects of (i) insect-mediated pollination service and (ii) potential pollination deficit on fruit set, seed set, and fruit quality (size, weight and malformation). Flower-visitor observations revealed that the pollinator community was solely comprised of unmanaged pollinators, despite the presence of apiaries in the surrounding landscape. Based on the pollination experiments, we found that the pollination service mediated by wild insects improved fruit size as a qualitative value of production, but not fruit set. We also found no evidence for pollination deficit in our urban environment. These results suggest that the local community of wild urban pollinators is able to support strawberry crop production, and thus plays an important role in providing high quality, local and sustainable crops in urban areas. The data are related to the scientific paper "Blareau, E., Sy, P., Daoud, K., Requier, F. (under review) Economic costs of the invasive Yellow-legged hornet on honey bees. Insects".Three datasets are available:(1) Flower visitor observations were performed during the whole flowering period, from the 20th of April to the 27th of May 2021 (24 observation days). Flower visitor observations consisted of a 10 minutes observation sessions carried out between 9am and 6pm. We favoured days with temperatures above 12°C, with little cloud cover and no wind although some observation sessions were carried out on cloudy days since several occurred during the flowering period. Each flower was observed several times (i.e. several 10 minute observation sessions per flower) but never on the same day. Subsequent observations of a same flower were carried out at a later date at different times of day, in order to maximise our chances of seeing a diversity of pollinators, since different pollinators are active at different times of day [5,55]. Over the flowering period, we carried out 30 h and 10 min of flower observations (181 time replicates of 10-minute observations) on a total of 88 flowers (each observed on average 2.9 ± 1.1 times). We observed an average of 10.7 ± 7.4 flowers per day. An average of 4.4 ± 2.3 flowers were observed per location. Each flower visitor was counted and identified within the following 8 categories: honey bee (Apis mellifera), bumble bee (Bombus sp.), solitary bee, hoverfly (Syrphidae), other fly (Diptera), ant (Formicidae), thrips (Thysanoptera) or other insect.Data are available as a csv file titled " Flower.visitors.csv” with the following metadata:# METADATA# 'data.frame': 257 obs. of 11 variables:# $ FlowerID : Factor variable ; identity of the flower# $ Honey bee : Numeric variable; number of honey bees observed# $ Bumble bee : Numeric variable; number of bumble bees observed# $ Solitary bee: Numeric variable; number of solitary bees observed# $ Hoverfly : Numeric variable; number of hoverflies observed# $ Other fly : Numeric variable; number of other flies observed# $ Ant : Numeric variable; number of ants observed# $ Beetle : Numeric variable; number of beetles observed# $ Spider : Numeric variable; number of spiders observed# $ Thrip : Numeric variable; number of thrips observed# $ Other insect: Numeric variable; number of other insects observed(2) We performed pollination treatments at each location during the same time period as pollinator observations, to assess pollination services provided by the urban pollinator community. The four treatments are as follows: (i) flowers open to pollinator visits, (ii) flowers open to pollinator visits and cross pollinated by hand, (iii) flowers excluded from pollinator visits (self or wind pollination only), and (iv) flowers cross-pollinated by hand and excluded from pollinator visits. Comparing self/wind and hand pollination measures pollinator dependence. Comparing self/wind and open pollination measures pollination service, i.e. the contribution of insect pollinators to crop production. Comparing open and hand pollination measures pollination deficit, i.e. whether pollinators are able to saturate the flower in pollen, thus allowing it to produce fruit at its highest potential. Comparing open pollination with and without hand pollination indicates whether insect pollination alone is sufficient to maximise fruit yield. The comparison of hand pollination and open pollination with hand pollination indicates whether hand pollination only is enough to maximise fruit production, or weather an input from pollinators is necessary. Plants from the self/wind pollination treatment and the hand pollination treatment were bagged with mesh netting (Alt’Droso Maraichage, 0.8 × 0.8 mm mesh) to prevent pollinators from visiting these flowers. For treatments that required hand pollination, pollen was collected from the study plants. We visited each flower twice within the same day with a paintbrush to ensure flowers of the same treatment received pollen from several other plants. In total, 172 flowers were considered for the pollination experiment, 46 flowers were affected to the open pollination treatment (2.3 ± 1.2 per location), 28 to the hand and open pollination treatment (1.4 ± 0.9 per location), 63 to the self/wind pollination treatment (3.2 ± 1.5 per location), and 35 to the hand pollination treatment (1.8 ± 1.3 per location). Once flowering was over, we measured fruit set by recording whether each flower from each pollination treatment successfully produced fruit or not.Data are available as a csv file titled "Fruit.set.csv” with the following metadata:# METADATA# 'data.frame': 172 obs. of 3 variables:# $ Location : Factor variable ; identity of the location of the plant# $ Treatment: Factor variable ; identity of the pollination treatment with: O for flowers open to pollinator visits; O+H for flowers open to pollinator visits and cross pollinated by hand; E for flowers excluded from pollinator visits (self or wind pollination only), and E+H for flowers cross-pollinated by hand and excluded from pollinator visits# $ Fruit set: Numeric variable; fruit set with O for failure (no fruit) and 1 for success (fruit formed)(3) Fruits were then harvested once they were fully formed (i.e. as soon as fruits had fully reddened), between the 31st of May and the 10th of June 2021. We recorded fruit malformation, by considering a fruit with a clear aggregation of unfertilised achenes as showing a malformation (Fig. S3). We measured fruit weight (Ohaus, Adventurer, precision 0.01 g, capacity 3100 g) and fruit size as the maximum width at the widest point (France métrologie, accuracy 1 mm, capacity 1600 mm) within one day of harvesting. We chose width as the measure for fruit size since it is used to determine the commercial class of fruits [46]. Seed set was then counted once all fruits had been cropped. For maximum precision, strawberry flesh was separated from the seeds before counting, using a small meshed sieve which collected only the seeds.Data are available as a csv file titled "Fruit.quality.csv” with the following metadata:# METADATA# 'data.frame': 125 obs. of 6 variables:# $ Location : Factor variable ; identity of the location of the plant# $ Treatment : Factor variable ; identity of the pollination treatment with: O for flowers open to pollinator visits; O+H for flowers open to pollinator visits and cross pollinated by hand; E for flowers excluded from pollinator visits (self or wind pollination only), and E+H for flowers cross-pollinated by hand and excluded from pollinator visits# $ Fruit malformation: Numeric variable; fruit malformation with O for the absence of malformation and 1 for the presence of malformation# $ Fruit size : Numeric variable; size of the fruit (in cm)# $ Fruit weight : Numeric variable; weight of the fruit (in g)# $ Seed number : Numeric variable; number of seeds