Micromobility services before and after a global pandemic: impact on spatio-temporal travel patterns

1. Introduction

On March 2020, the World Health Organization (WHO) declared the novel Coronavirus Disease caused by the SARS-CoV-2 virus (hereafter, COVID-19), as a global pandemic. As an effort to mitigate its rapid spread, lockdowns were imposed causing many changes in travel behavior, not yet exactly quantified, or understood, offering numerous possibilities for exploration: one of them being the impacts on micromobility services. Before the pandemic outbreak, micromobility experienced a burst worldwide, with operative services in over two hundred cities and many investors valuing these companies in the billions of dollars (Wiggers, Citation2019). With the aim to tackle the last mile problem, it has been defined as a type of shared mobility service that allows short-term access to low-speed vehicles (Shaheen et al., Citation2016). Micromobility’s rapid expansion is mainly based on the fact that it is offering a flexible transport option that is able to avoid road congestion, reduce parking space needed, lower noise/air pollution and encourage intermodality with mass transit (Aguilera-García et al., Citation2020).

Due to its benefits, micromobility’s scientific literature is and will continue to grow. The available and revised studies can be divided in two groups. The first one includes all the studies that used GPS trip data to approach different analyses of single shared modes. For example, (Degele et al., Citation2018) clustered shared moped’s users according to their different characteristics, while (Xu et al., Citation2019) explored the variations of temporal signatures during weekdays and weekends for dockless bikes, and (Caspi et al., Citation2020) found scooters’ usage was concentrated mainly in the city center and closely related to high employment zones and areas with cycling infrastructure. The second group of studies focuses on the comparison of more than one type of shared mode. For example, (Zhang et al., Citation2020) compared dockless bike and scooter-sharing services and their findings showed that scooters had a more compact spatial distribution than bikes and that their high demand was associated with points of interests, metros, and residential zones. Similarly, (Lazarus et al., Citation2020; McKenzie, Citation2020) compared docked and dockless bikes; the results from the first study revealed that dockless bikes traveled for longer distances, while in the second study the author compared docked and dockless bikes with scooter services concluding that the docked systems were predominantly used for commuting while dockless scooters suggested a more recreational use.

All the above-mentioned studies were published before the pandemic. The scientific literature that addresses micromobility services after the pandemic is just beginning to be published and available, especially now, as these services are demonstrating to be crucial for improving the cities’ resilience (European Commission, Citation2020). After lockdowns were relaxed, grey literature suggested that micromobility services grew in popularity and that they help increase public transport system’s resilience to pandemics (Ardila-Gomez, Citation2020; Asenjo, Citation2021; Blanchar, Citation2020; Harrabin, Citation2020; Lomas & Dillet, Citation2020; Oltermann, Citation2020; Porcel, Citation2021). However, scientific research supporting/contrasting these statements continue to be scarce and the few studies found are mostly based on online surveys (Awad Núñez et al., Citation2020; Beck & Hensher, Citation2020; Bergantino et al., Citation2021; Campisi et al., Citation2020; Christoforou et al., Citation2021; Meena, Citation2020; Nikiforiadis et al., Citation2020; Pawar et al., Citation2020). The study done by (Dias et al., Citation2021) conducted a systematic literature review on the spread of shared scooters and analyzed the case of Braga in Portugal where the city relied on scooters to maintain social distance. Other studies like (Wang et al., Citation2021) estimate the effects that public transport restrictions may have in increasing private car usage and traffic congestion. The authors recommended that these restrictions should be accompanied with a support for micromobility services and congestion alleviation strategies. Survey-based studies include, for example, the survey conducted by (Awad Núñez et al., Citation2020) which found that even though the demand for micromobility services was marginal before the outbreak, the willingness to use bike-sharing or scooter-sharing services was relatively high during the first months of the pandemic (67.7%). Another survey conducted in Greece showed that bike-sharing was not perceived as safe as walking or using a private car, but it was seen safer than using a taxi or public transit (Nikiforiadis et al., Citation2020). Both studies pointed to the fear of contagion as the main reason behind micromobility’s increased popularity after COVID-19, stating that more sustainable modal choices could be a temporal behavior and not necessarily a permanent one. This latter issue was the focus of the research by (Monterde-I-Bort et al., Citation2022) that distributed a survey around 10 countries to evaluate the extent to which pandemic travel behavior will be retained. They applied the survey in two phases: phase 1- spring 2020 and phase 2-fall 2020 obtaining 636 cases. During lockdowns, the majority of respondents reported a reduced car usage, but also public transport and walking. In phase 2, when the lockdowns were relaxed, they reported going back to using the car and walking as normal but in the case of public transport, pre-pandemic levels were not achieved yet. The least affected mode reported was bike as respondents declared that its usage hardly changed at all. This was also the case in (Štefancová et al., Citation2022) which conducted a survey to 1.048 people in Slovakia and evaluated the impact of COVID-19 in travel behavior. Their results showed that many of the respondents reduced the use of public transport but as an alternative, they started using more bikes and scooters. Similarly, (Fistola et al., Citation2022) obtained almost 200 responses from an online survey distributed around university students in Italy and found that 81,6% of the students were willing to change their travel behavior toward more sustainable modes after the pandemic. The study demonstrated that micromobility services are highly appreciated by university students in study campuses. Their results also suggested that reducing the service’s price (especially those that demand a payment for unlocking the vehicle), offering bonuses to buy electric bikes or scooters and increasing segregated cycling infrastructure were demanded strategic measures to promote micromobility. The study conducted by (Nguyen & Pojani, Citation2022) examined recreational cycling in Hanoi, Vietnam after the pandemic. Their results from a sample of 356 face-to-face surveys collected in April 2021 showed that a quarter of them started cycling recreationally since the first lockdown and nearly half of them cycled regularly (more than four times per week).

Although all these surveys offer useful insights, we aim to explore studies that used another type of data source (GPS records) and analyzed COVID-19 impacts as done in (Li et al., Citation2020; Liu et al., Citation2021; Molloy et al., Citation2021; Teixeira & Lopes, Citation2020). The study by (Teixeira & Lopes, Citation2020) addressed how New York’s subway and bike-sharing systems were affected by COVID-19 outbreak. Their findings revealed that bikes were more resilient with a less significant ridership drop (71% vs 90% ridership drop), an increased average trip duration (from 13 min to 19 min) and an observed modal transfer from some subway users to the bike sharing system. Their study provided evidence on the role of the bike-sharing system and cycling in general, to ameliorate the effects of the pandemic, as it quickly provided an alternative transport option reinforcing the transport offer in areas with higher demand at a fraction of the cost of new roads or public transport infrastructure. (Li et al., Citation2020) evaluated the changes of micromobility services’ behavior before and during COVID-19 pandemic period in Zurich, Switzerland. They considered four types of shared services: 1) docked bike, 2) docked e-bike, 3) dockless e-bike and 4) dockless e-scooter. Their results showed that micromobility trips were longer in distance and duration after the pandemic and that activities related to home, park and groceries were more relevant, while leisure and shopping decreased in importance. (Song et al., Citation2022) used trip data to apply graph-based techniques and spatial correlation models to analyze the changes of the bike-sharing behavior after the pandemic in Singapore. Their results showed that bike ridership climbed by 150% after lockdowns were relaxed and they found a positive spatial autocorrelation between bike ridership with centrality measures, high residential densities and mixed land use. These studies published after the pandemic, tried to address impacts, the changes in modal choices, spatio-temporal distribution of people or shared mobility trips, the effects of lockdowns/telework and online studies and the extent to which these changes could be temporal or become permanent in a post-COVID transition.

Following on this line of research, the objective of this paper is to answer the following question: how was micromobility affected by the pandemic? We aim to measure the extent to which the services’ usage levels dropped, how did they recover and the key aspects to understand their recovery in relation to urban activities and land uses. To that end, taking the city of Madrid as case study, we evaluate the changes in travel behavior comparing two scenarios: pre-COVID time for 2019 and the first two months of 2020 and COVID time for the rest of 2020. We analyze generated and attracted micromobility trips throughout the day and during morning and afternoon peak hours (capturing commuting behavior to/from home) and explore its relationship with different land uses to assess which activities gained/lost importance. Our paper contributes to the literature in several ways. Firstly, to our knowledge, this is one of the few studies that analyze and compare three different micromobility services’ usage patterns (BiciMAD bike, mopeds and scooters) before and after the pandemic using trip data (GPS records). Unlike many of the recent publications, our study will focus not on users’ responses, but rather on their trips by analyzing GPS records collected by different micromobility services. Secondly, with most studies considering daily behavior, our study’s temporal resolution offers larger granularity as it evaluates how the travel patterns change throughout the day, considering the morning and afternoon peak hours. Thirdly, we relate micromobility trips with the distribution of land uses, using spatial regression models. This makes it possible to know if the pandemic has altered the spatial patterns of the use of micromobility systems. The selection of Madrid as a case study is also of special interest, as it is known as one of Europe’s most important living labs for shared mobility and one of the metropolitan areas that was most affected by COVID-19 pandemic. Our research relates to the sustainable transport topic, as it involves analyzing micromobility services. Bike-sharing programs and more recently, micromobility services, have proven to positively influence people to adopt a more sustainable travel behavior (López-Carreiro, Citation2021; Munkácsy, Citation2017; Romanillos, Citation2018; Velázquez Romera, Citation2019). With the current climate crisis, achieving sustainable mobility is at the heart of many global and regional public policies (European Commission, Citation2020; United Nations, Citation2015), with the reduction of car usage as one of the main goals to reduce CO2 emissions. By covering travel demands for those low or unserved areas by public transport systems, micromobility is facilitating the needed multimodality and intermodality to substitute door-to-door car trips. Each person now more than ever before, has the option, to not own a car but instead, use public transport and different shared vehicles to satisfy her/his travel demands. However, due to its recent nature, micromobility services are not fully integrated in the traditional modeling and planning processes. We consider as effort-worthy the intention to understand these new forms of mobility in order to integrate them accordingly in the planning processes. Therefore, this is an effort-worthy topic of interest, as before COVID-19 city planners and transport authorities were already struggling to assess micromobility impacts on travel behavior and existing infrastructure. Nowadays, it is more necessary to explore how the pandemic has changed travel patterns and urban dynamics, as it will be essential for decision-making processes and formulating future regulations (Romanillos et al., Citation2021). The remainder of the article is organized as follows. The introduction of the case of study is offered in Section 2. In Section 3 we described the data and methods used. Section 4 summarizes the main results and discussion and finally in Section 5 we offer conclusions.

2. Case of study

We consider Madrid as our case of study. Madrid has a multiple and varied shared mobility supply, great diversity of land use and high densities of population and employment with more than 6 million people in the Metropolitan Region, and half located in the Municipality of Madrid (Instituto Nacional de Estadística, Citation2018). As many cities worldwide, Madrid imposed a lockdown from March to June 2020 in order to reduce social interaction (Ministerio de la Presidencia, Citation2020; see Figure 1). In general, (Romanillos et al. Citation2021) identified five phases with different lockdown rules in Madrid (see Table 1). This measure resulted in approximately an 80% public transport ridership drop during the pandemic-peak months (March, April and May). From June on, lockdowns were relaxed, and services began to slowly recover with the re-opening strategy launched in May (called “Transition Plan to the New Normal”) (Glodeanu et al., Citation2021). Most micromobility operators in Madrid started to conduct hygiene/disinfection duties to their fleet of vehicles and inform customers about it (as so did the public transport systems), as well as offering free face masks and hand sanitizer, in order to increase safety.

Figure 1. Timeframes for the research databases and Madrid’s timeline related to COVID-19 mobility restrictions.

Source: own elaboration.

Display full size

Previous to the pandemic, the city was known to be a shared mobility living lab, which allowed its residents to be familiar with the emerging transport options, especially micromobility services (Aguilera-García et al., Citation2020). In 2019, the shared fleet was estimated in more than 20.000 vehicles (Arias-Molinares & García-Palomares, Citation2020; Bernardo, Citation2019; Granda & Sobrino, Citation2019). These services are usually supported by mobile applications where their clients register and locate the vehicles. In the case of Madrid, all micromobility services offer electric vehicles and can be station-based or dockless models. In this paper, we focus on three micromobility services (shared modes) operated by four different companies: BiciMAD, which is Madrid’s public and station-based bike-sharing system, and three private and dockless micromobility companies (see Figure 2 and Table 2). To access the anonymized trip databases, collaboration agreements were stablished with two of the most important micromobility operators in Madrid (Movo and Muving). In the case of BiciMAD, the data was publicly shared through their open data portal. Station-based services like BiciMAD, have designated locations where users pick and leave the vehicles at, while dockless services, like Movo and Muving, offer more flexibility as the vehicles can be picked/returned at any location within a geographic area (also known as geofence).

Figure 2. Micromobility services analyzed. From left to right: 1) Station-based bike-sharing (BiciMAD bikes), 2 and 3) dockless moped-style scooter-sharing (Movo and Muving mopeds) and 4) dockless scooter-sharing (Movo scooters).

Source: own elaboration.

Display full size

One of the aspects that most differentiates these services is their area of coverage (geofence), which is closely related to their different models. In the case of BiciMAD, being a station-based model, it covers essentially the city’s core center area. In the case of dockless services, the geofence is much larger reaching other peripheral areas and being more homogeneously distributed in the city which creates a relatively less intense usage in each location. This study performs the spatio-temporal analyses of each service considering these different geofences, which are all inside Madrid’s Municipality.

3. Data and methods

4. Results and discussion

In Tables 4 and 5, we have summarized the main descriptive characteristics for the complete and shortened datasets analyzed. As it can be seen, considering the entire dataset, there was a general micromobility ridership drop of around 10% due to the pandemic outbreak, which is quite low considering the downfall of trips in Madrid’s public transport system (estimated in 80% during peak months). This result suggests that, as was found in other studies (Dias et al., Citation2021; Teixeira & Lopes, Citation2020), micromobility is more pandemic-resilient. Some services as the case of BiciMAD and Muving show an even better performance when considering only the last semesters of the 2019 and 2020, as the shortened datasets eliminate the weight of service closures during the peak months of the pandemic. Findings also show that BiciMAD is the most important micromobility service as it accounts for 78,1% of the total micromobility trips, followed by mopeds (Movo and Muving) with 19,6%.

When considering only the last semesters (Table 5), we observe that the downfall in ridership is even lower (around 3% lower than considering the entire dataset). Interestingly, when comparing BiciMAD’s performance in 2019 and 2020, it is the only service with no drops, but rather a 5,8% rise in ridership. In other words, our results are yelling that BiciMAD was the only micromobility service that show higher usage after COVID-19, being the most resilient one to the pandemic. This result is consistent with findings from (Song et al., Citation2022) that also demonstrated that bikes’ usage climbed 150% in relation to pre pandemic levels. We speculate that the sustained usage of the bike-sharing system after the pandemic, may be related to the fact that it is one of the cheapest shared mobility services in the city of Madrid, which allowed many residents to move in a safer (by allowing social distance in comparison with crowded public transport systems) and flexible way after COVID-19. Additionally, based on the last BiciMAD’s survey conducted in 2019, most of its subscribers were males, with 35 to 49 years old, employees with high education level that earn low-medium monthly income, yet 26,5% of them earn less than 1.300 €/month (relatively low income) (Arias-Molinares et al., Citation2021). On this survey, results also showed that the main trip purpose when using BiciMAD was commuting, with 46% of cyclists using it for work/study motives at least once a week or more (of which 22% use it on a daily basis) (Arias-Molinares et al., Citation2021). All these facts may indicate that the bike-sharing system could have been an attractive alternative transport option for those essential workers with no cars and low income that were avoiding crowded public transport systems (fear of contagion). On the other hand, scooters seem to be the most affected service with a downfall of 84%. Moreover, in general, average trip times, distances and speeds had minimal variations in 2020 with respect to 2019 (10 seconds, 100 meters and 0,2 points less respectively).

More specifically for the case of BiciMAD and supporting the findings from (Li et al., Citation2020; McKenzie & Adams, Citation2020; Molloy et al., Citation2021; Teixeira & Lopes, Citation2020), average trip time rose 1,2 minutes, average trip distance rose 200 meters while the speed was maintained at 11,3 km/hr from 2019 to 2020. As was mentioned earlier, with the fear of contagion, and with bikes allowing social distancing, BiciMAD users could possibly be traveling for longer distances replacing public transport options (more crowded metros or buses). In the case of scooter-sharing on the other hand, trip time decreased one minute from pre-COVID to COVID times, average trip distance decreased 200 m and average speed rose 20 seconds. Scooter users traveled less distance but at a faster pace which could be related to a “traffic-calming effect” after the outbreak, as automobile traffic flows were reduced with the introduction of telework and mobility restrictions. And finally in the case of mopeds, all the indicators dropped in the case of Movo (70 seconds, 300 meters and 0,9 points less respectively) and were maintained in the case of Muving (with the exception of speed that lowered 0,3 points). Overall, results show that trip times distances and speeds are very similar for all modes and are maintained similar after the outbreak. Muving mopeds are the ones that travel longer distances and at a faster pace, followed by Movo’s mopeds, BiciMAD’s bikes and lastly Movo’s scooters.

4.1. Temporal patterns

Figure 3 shows pre and COVID time temporal behavior for micromobility trips aggregated by month of the year. In the normal situation prior to the COVID-19 outbreak, BiciMAD had high trip counts on the months when most residents are in town with daily routines (not vacations) and temperatures are warmer (spring and fall). These peaks are shown in the months of March and June (spring season) and then again in September and October (fall season). On the contrary, low trip counts are observed on those months toward the summer season, when usually Madrilenians travel out of town for the Holy Week holidays (spring), summer or Christmas vacations, and especially in summer or winter when the temperatures are extreme. In the case of mopeds, there is a slight difference with respect to BiciMAD’s pattern around the months of summer season (July and August), as Muving and Movo seem to better retain their trip counts during vacations. Moreover, with regards to scooters, the pattern is completely different from the other two modes, as trip counts are mostly low and tend to increase, mostly toward the summer season (July, August and September). These temporal patterns may be suggesting an interesting finding: that BiciMAD’s users are mostly residents of Madrid, that moped’s users seen to be a mix of residents and tourists while scooter’s users could mostly be visitors, supporting findings from (Hosseinzadeh et al., Citation2021; McKenzie, Citation2020).

Figure 3. Pre and COVID time micromobility trip counts by month of the year (using complete database).

Source: own elaboration.

Display full size

Regarding the second scenario (COVID-19 time), the yearly pattern drastically changed due to all the services being closed for mobility restrictions during the pandemic peak period (March and April). However, once the services re-opened in May, the monthly patterns seen to recover and showed a “back-to-normal” tendency, which in the case of BiciMAD was better than the normal scenario surpassing its trip record in July, after the lockdown measures were relaxed. The month of July 2020 shows an essential change produced by the COVID outbreak: in July 2019 trip counts were low reflecting the usual patterns of Madrilenians that go on summer vacations, while in 2020, with the tourism decay and most people spending their vacations at their place of residence, the service had almost the same number of trips (313.745 trips) that the month of June 2019 (316.769 trips) which was the best trip record of 2019. For BiciMAD service, July 2020 represented the “boom” of bikes with a rise in ridership of 220% with respect to July 2019. Not only that, but in September the service reached its best record with (319.497 monthly trips) which was 7% higher with respect to the same month in 2019.

Exploring the weekly behavior, Figure 4 shows the temporal signatures of each service across hours and days of the week. In general, the plots clearly show that the most pandemic-resilient service was BiciMAD, maintaining and even surpassing its previous trip counts after the pandemic. In the normal scenario the signatures show that the services were mostly used for commuting morning hours on Tuesdays (in the case of BiciMAD and scooters) and Thursdays (for mopeds). Preferred modes for the morning commuting hours seem to be bicycles and mopeds while scooters are preferred for the afternoon hours. In COVID times, the signatures show less trips at morning peak hours, as many people stayed at home with teleworking measures. On the pre-COVID scenario, we can observe that the most profitable days for micromobility services were Wednesdays (BiciMAD), while in the case of mopeds and scooters, the weekly pattern flattens and seems to have a similar preference from Wednesdays to Fridays (for mopeds and scooters). On weekends, Saturdays are more profitable for all modes and for all the scenarios, especially in the afternoon and night hours.

Figure 4. Temporal signatures for Pre (left) and COVID time(right) aggregated to hours and day of the week for micromobility services (using shortened database).

Source: own elaboration. Notably, each y-axis is different as the trip volume differs between services.

Display full size

When considering the hourly behavior, Figure 5 shows that in general, all three modes have three peaks: the smallest one in the morning commuting hours (from 06 to 10 hrs), a middle one for lunch hours (13 to 15 hrs) and the highest one for the afternoon (from 17 to 22 hrs). The only service that follows a slightly different pattern is scooter-sharing, as trip counts increase mostly from lunch on (14 to 21 hrs) with less trips on morning commuting hours and the highest counts on the afternoon (around 19 hrs), which supports the hypothesis of recreational use. Additionally, regarding the morning commute hours, BiciMAD is used mostly from 06 to 08 hours, while dockless services (mopeds and scooters) are preferred for later hours (from 08 to 10 hr). The same occurs in the afternoon, with BiciMAD showing higher trip counts from 16 to 18 hrs, while the dockless show their peaks later from 18 to 20 hrs.

Figure 5. Pre and COVID time micromobility trip counts aggregated by hours of the day (using shortened database).

Source: own elaboration.

Display full size

Considering the hourly behavior after COVID-19, we can observe that the curves follow the same temporal trend with three peaks, but these peaks seem to have been displaced into earlier hours, especially in the case of BiciMAD service (for lunch at 12 hrs and for afternoon commute at 17 hrs). With the consolidation of online studying and telework, the most noticeable difference is observed at morning peak hours, with trips at 07 hrs being considerably lower for bikes and at 8-9 hrs for mopeds and scooters. This supports findings from similar research by (Li et al., Citation2020) which found that the only peak that was maintained after COVID-19 was the afternoon commute as the number of commuting trips decreased due to telework measures. Mid-morning trips (from 10 to 12 hrs) also rose, which could be related to people teleworking and having more flexibility to carry errands in this period, similar to findings in (Molloy et al., Citation2021) that noticed a higher amount of trips occurring over midday, hinting that these trips were not conducted for commuting purpose.

5. Conclusion

Micromobility services have demonstrated to be a crucial part of the solution to achieve sustainability (European Commission, Citation2020; Fulton, Citation2018; Lazarus et al., Citation2020; Teixeira & Lopes, Citation2020). These recently introduced services are offered as an alternative to the private car, promising a shift to a more sustainable travel pattern by reducing car-dependency and thus, the environment pollution. With the COVID-19 pandemic, the need for this modal shift has grown and so does the need to understand the changed urban dynamics. We have processed and studied real data from three micromobility operators, which allowed us to compare the last semesters of 2019 and 2020. This exploration helped us to begin to measure and understand the impact of a global pandemic in one of the most affected regions of Europe (Madrid). Our descriptive analysis yields that after COVID-19, BiciMAD actually gained and surpassed its trip records (September 2020). We also found that, as in (Li et al., Citation2020; Molloy et al., Citation2021; Teixeira & Lopes, Citation2020), average trip time and distance for bike-sharing increased with COVID-19. Multiple regression results showed that residential and commercial areas became more relevant activities after the pandemic, while workplace locations (office and industrial) and education facilities had a downfall, especially in the early morning commutes. Our results are similar to previous studies (Song et al., Citation2022; Zhu et al., Citation2020) that have found a close link between micromobility usage and high residential densities. This is closely related to the fact that a great extent of micromobility users were confined at their homes and began to telework or to study online. This caused a ridership drop of 80% in public transport. However, our findings show that micromobility trips were down, in general, only by 10%, which is a relatively better performance. Consequently, one conclusion for our study is that the pandemic has highlighted the importance of micromobility services as they seem to be more pandemic-resilient than mass transit, also demonstrated by (Campisi et al., Citation2020; Dias et al., Citation2021). Interestingly, for the case of Madrid parks were not significant in the models, in contrast to other studies like (Li et al., Citation2020) in which they found park activity, mainly done in open areas, increased their attractiveness after COVID-19. Data also suggests that, as mentioned by previous studies, fear of contagion could have influenced customers to adopt alternative shared modes for their last mile trips (Arias-Molinares et al., Citation2021; Li et al., Citation2020). In general, our results are consistent with the literature revised, as micromobility trips’ time and distance increased after the pandemic and some uses associated with residential and commercial areas were more relevant that others related to activities like work, study, leisure or cultural points of interest.

These results could be of help to understand the impacts brought by the pandemic and which places became more attractive at a certain time of the day, to implement better-informed policies that could help improve urban mobility. Micromobility services had shown to be crucial for maintaining social distance while at the same time being a flexible transport option, thus the positive outcomes brought by the pandemic (increased bike trips) should be maintained or improved by designing and planning better cycling infrastructure and parking facilities for shared mobility. As some studies suggest, the lack of car traffic during the lockdown undoubtedly made cycling more attractive with higher number of trips and bicycle sales (Molloy et al., Citation2021). We have entered the era of pandemic-resilient cities as an indicator of sustainability (POLIS., Citation2021). Hence, cities should consider micromobility services as a crucial part in the mobility ecosystem and as a solution to limit future contagions. As circumstances become normal again, many residents will go back to their pre-pandemic mobility patterns. With the fear of contagion still influencing, some users may decide to start using their private automobiles. Some studies have already shown that as lockdown measures were relaxed, a higher share of kilometers was performed using motorized transport but also active mobility (walking and cycling), with public transport recovering at a slower pace (more gradually) (Molloy et al., Citation2021; Wang et al., Citation2021). As (Wang et al., Citation2021) estimated, a 50% restriction of public transport capacity could derive in a 142% increase of car usage. Hence, these measures should always be accompanied with other complementary strategies that promote micromobility services, like adequate parking, attractive pricing schemes and traffic calming measures. Authorities and transport planners should avoid the unsustainable modal shift to private cars by promoting shared mobility and safety measures in public transport. Strategies to promote cycling include the addition of new cycling infrastructure (e.g., transitioning from pop-up cycling lanes to permanent ones), designing and implementing new mobility hubs and reallocating large swaths of street space from the car to active modes. Such policies oriented to connect and increase cycling paths and lanes, to make intersections and circulation for cyclists safe, to implement storage/parking infrastructure and to improve urban design for cycling are all elements to be considered and broadly adopted as they involve a relatively low-cost implementation with high flexibility. The increased preference for cycling as a result of the pandemic should make authorities and transport planners reflect on long-term policies to maintain this sustainable shift. Telework and online studies demonstrated to play an important role to stop the spread of COVID-19 in workplaces and public transport, hence an important issue will be to reflect on the extent to which these measures should be encouraged after the pandemic and its implications for transport policy as stated in (Molloy et al., Citation2021). On the other hand, micromobility operators could capitalize the opportunity that the pandemic has given them. Recommendations for micromobility operators include maintaining hygiene/disinfection measures adopted during COVID-19, offering competitive prices and discounts to customers (including gamification strategies) and increasing the vehicle supply and coverage areas. Increasing coverage could be especially interesting for operators in those areas with high student densities as many studies have supported that young people tend to be more willing to use micromobility services (Degele et al., Citation2018; Fistola et al., Citation2022). Limitations in this study include the fact that cleaning the datasets required certain assumptions to be made of the underlying platforms, such as no trips over a certain speed, etc. The approach used in this work was overly conservative likely erroneously removing actual trips from analysis in order to not include redistribution trips. This was an unfortunate necessity which could be rectified by the data providers tagging trips as either user trips, or administrative trips. Additionally, a second limitation of the study regards that our datasets do not have any information about micromobility subscribers (e.g., unique id for exploring repeated customers and travel patterns or mode from which they are switching to/from) due to privacy issues. This is crucial to better understand the changes in modal choices and travel patterns after the pandemic. Future studies could analyze data for 2021 to see if the good resilient trends are maintained (the so called “bicycle boom” after COVID-19) and to evaluate in a better way the effect of the service recoveries as well as addressing the impacts of new BiciMAD expansions.