Planning against qualitative job insecurity: testing two online exercises

Flexibilization, digitalization and Artificial Intelligence (AI) advancements are drastically transforming the process and content of work (Nyberg et al., 2025). As a result, many workers worry about losing valued work features (i.e. qualitative job insecurity; De Witte et al., 2010). Job insecurity relates to lowered physical and mental well-being and lowered work performance (for meta-analyses, see Cheng and Chan, 2008; Sverke et al., 2019, 2022). While prior research has mostly focused on worries about losing one’s entire job (i.e. quantitative job insecurity), qualitative job insecurity alone appears as detrimental to well-being and work engagement as a combination of the two (Urbanaviciute et al., 2021). As such, insight into how workers can mitigate qualitative job insecurity becomes increasingly important.

While most predictors of quantitative and qualitative job insecurity are beyond the control of individuals (for meta-analyses, see Jiang et al., 2021), career planning may pose a practical solution for individuals to exert influence on their job insecurity experience. For example, Koen and Parker (2020) show that workers who use more career planning experience more control and less quantitative job insecurity. Career planning is often considered the starting point to initiate proactive actions to improve career perceptions and experiences (Parker et al., 2010) and relates to various career success indicators such as career satisfaction (for meta-analyses, see Ng et al., 2005). However, when looking specifically at career planning in relation to quantitative and qualitative job insecurity, there are exceptions for career planning success. Studies show that more career planning can be unrelated or even positively related to quantitative and qualitative job insecurity (El Khawli et al., 2022; Langerak et al., 2022). This may be explained by the external context of the studies, as they were conducted during the COVID-19 pandemic. Relatedly, qualitative studies illustrate how careers often do not unfold according to plan and take shape in ways not anticipated beforehand (Ibarra, 2004; Koen et al., 2016).

The aim of this study is to investigate the idea that career planning only lowers qualitative job insecurity when the type of career planning matches the context of the worker. To this purpose, we first integrate the exploitation-exploration framework (Almahendra and Ambos, 2015) with the career development literature (Bandura, 1991; Ibarra, 2004) and differentiate between exploitation-based career planning (aimed at strengthening skills and knowledge for the current career goal) and exploration-based career planning (aimed at strengthening skills and knowledge for multiple career possibilities). We then hypothesize that both types of career planning may decrease qualitative job insecurity through different mechanisms: career goal awareness and career option awareness. We test these hypotheses in two online intervention studies comparing workers who do an exploitation-based or exploration-based career planning exercise with a control group. To test the impact of workers’ context, we investigate whether individual career context (career commitment, Study 1) and external work context (perceived labor market demand, Study 2) moderate the expected pathway from career planning via increased awareness to job insecurity (see Figures 1 and 2 for the research models).

By investigating qualitative rather than quantitative job insecurity, we add to the limited literature on how to minimize worries about losing valued job features (Van der Heijden et al., 2024). In addition, prior insight into the relation between career planning and career outcomes is largely based on correlational studies involving self-report measures (Ng et al., 2005) and could be influenced by attribution bias (i.e. workers in more favorable career situations could attribute that to their own planning skills). We contribute to the literature by investigating the relationship between career planning and job insecurity using an experimental design. Moreover, we contribute to the theory on how career planning may affect qualitative job insecurity. Lastly, by investigating the context-dependency of career planning interventions, we contribute to uncovering theoretical conditions for career planning effects on qualitative job insecurity.

In the exploitation-exploration framework (Almahendra and Ambos, 2015), coping with dynamic environments is realized through two activities: exploitation (i.e. development of existing knowledge, improving existing features, low risk actions) and exploration (i.e. pursuit of new knowledge, experimenting, risk-taking). While exploitation and exploration have mostly been used for understanding organizational success, the duality of exploitation and exploration is also valid for individuals (Rosing and Zachter, 2017). Career planning is an activity that involves reflection on interests and values and how to realize a situation with a closer match between the ideal and current situation, considering those interests and values (Savickas et al., 1996). Based on the career development literature, we distinguish between two types of career planning that align with the exploitation-exploration framework.

Most literature positions career planning as a set of behaviors that resemble exploitation activities. Leading theories state that careers are shaped through a goal-directed process and that workers’ goals are based mostly on their past or current environment and knowledge (self-regulation theory, Bandura, 1991; goal-setting theory, Locke and Latham, 1990; expectancy theory, Vroom, 1964). Pursuing a goal is generally closely related to one’s current self-image and positioned as a relatively low-risk activity (Jiang et al., 2019). By formulating career goals, strategies, and taking actions, people may decrease discrepancies between the current and desired situation (Savickas and Porfeli, 2012). For example, by improving one’s expertise or voicing ambitions. This resembles exploitation, as these are processes aimed at the improvement of existing features and have few risks attached. We refer to this perspective on shaping careers as exploitation-based career planning.

Alternatively, a smaller stream of literature positions career planning as an exploration process. Possible selves theories (Markus and Nurius, 1986; Ibarra, 2004) state that workers develop their career by imagining and trying out multiple roles and potential career goals, as pursuing one goal can be demotivating during setbacks and limit workers’ potential. Workers are guided by various identity narratives about their current selves and, more importantly, potential future selves. The more vivid the images of future selves are, the stronger the motivation towards or away from those selves becomes. For more vivid future selves, workers must find information on these identities, which involves experimenting with different roles and taking risks by challenging the unknown. These activities resemble exploration, where the pursuit of new knowledge, experimenting, and risk-taking are central. We refer to this perspective on shaping careers as exploration-based career planning.

Both exploitation-based and exploration-based career planning help workers give direction to their careers. Exploitation does so by encouraging individuals to set a clear goal and sub-goals, whereas exploration prompts them to recall or discover multiple possible career options. In both cases, workers reflect on the job features they value, which increases their awareness of what their goals or desired options involve. Goal awareness reflects the extent to which workers are conscious of what career goal they aim to pursue. Option awareness reflects the extent to which workers are conscious of multiple career alternatives.

For exploitation-based career planning, we argue that greater goal awareness helps workers form a clearer picture of the job features they want to attain. Sub-goals provide a concrete path toward achieving these features. Having such a path may help workers feel they are actively shaping their desired work situation (Locke and Latham, 1990), reducing worry about losing valued aspects of their job. For exploration-based career planning, we argue that option awareness helps workers identify several ways to secure the job features they value. If one option becomes threatened, they can pursue another that aligns with their interests and values (Ibarra, 2004), which may also reduce worry about losing important job features. As such, career planning helps workers change their perception in a way that helps them realize the work features they desire (Lichtenthaler and Fischbach, 2019).

We test these hypotheses in an online intervention study where workers do an exploitation-based career planning exercise (exploitation group), an exploration-based career planning exercise (exploration group), or are assigned to the control group:

We propose that exploitation-based and exploration-based career planning reduce qualitative job insecurity only when they match the worker’s context. Drawing on the exploitation-exploration framework, individuals facing uncertain futures may benefit from shifting between exploitation and exploration over time – using exploitation in more stable contexts and exploration when the future is unclear. Both forms of career planning may lower qualitative job insecurity through increased goal or option awareness, but only when they fit the worker’s individual and market context. To test this, our intervention studies include two moderators: career commitment (Study 1) and perceived labor market demand (Study 2).

Study 1 examines career commitment (i.e. workers’ emotional attachment to their current line of work; Carson and Bedeian, 1994) as an individual context variable. We pose that attachment to the current line of work may form a source of stability that favors exploitation activities (Almahendra and Ambos, 2015). When workers experience less stability from their line of work, such as early career workers or those dissatisfied with their jobs (Ibarra and Obodaru, 2016), we expect workers may struggle to gain more awareness regarding one career goal through doing an exploitation-based career planning exercise.

Exploitation-based career planning may thus not be effective to lower qualitative job insecurity via increased goal awareness in this case. These workers, low in career commitment, may benefit instead from exploration-based career planning, as this matches with the unclarity of their future career (Almahendra and Ambos, 2015). They may be more willing to explore alternatives, whereas those strongly attached to their current line of work may find it difficult to imagine more than one future work self. Thus, exploration-based career planning may only be effective to lower qualitative job insecurity via increased option awareness for workers low in career commitment.

Study 2 examines perceived labor market demand (i.e. extent to which workers believe their occupation is currently in demand) as an external context variable. We pose that sufficient demand for one’s line of work may form a source of stability that favors exploitation activities (Almahendra and Ambos, 2015). We expect that exploitation-based career planning only lowers qualitative job insecurity through more goal awareness, when workers perceive sufficient demand for their occupation to fulfill their set goals. Without this perceived demand, workers may lack the belief that they are capable to achieve their goals, which is a prerequisite to lower career-related worry (Kleine et al., 2023).

For workers who perceive low demand for their occupation, exploration-based career planning may form a better alternative to lower qualitative job insecurity, as this matches with the unclarity of their external work context (Almahendra and Ambos, 2015). Being aware of multiple options could provide alternative routes to maintaining valued job features despite fewer job opportunities. When perceived demand is high, workers may feel less inclined to seriously consider the options they imagined during the exercise, weakening the effect.

Two online intervention studies were conducted to examine whether exploitation-based career planning and exploration-based career planning can lower qualitative job insecurity via increased career (goal and option) awareness (H1a and H2a). In addition, Study 1 tested a first-stage moderation of career commitment (H1b and H2b) and Study 2 tested a second-stage moderation of perceived labor market (H1c and H2c). Both studies involved exploratory analyses to investigate the role of acting on the intentions formulated during the career planning exercises and if results sustained two and six weeks later.

At T1, demographics were measured and participants were randomly assigned to one of three groups to do an exploitation exercise, an exploration exercise, or no exercise (control group). Study 1 also measured career commitment and Study 2 measured perceived labor market demand. Following the intervention, career goal awareness, career option awareness, and qualitative job insecurity were measured. At T2 (two weeks later) and T3 (six weeks later), career goal awareness, career option awareness, and qualitative job insecurity were measured again. At T2, respondents from the intervention groups also reported the extent to which they acted upon their intentions.

For the exploitation exercise (Appendix A), participants formulated one attainable, positive, and specific career goal that they wanted to achieve in the next ten years. Next, they were instructed to make a career plan following an arrow structure with sub-goals. Lastly, participants wrote down six actions they plan to initiate in the next two weeks to work towards their sub-goals. The exercise was based on the career adaptability intervention from Koen and colleagues (2012) and is in line with traditional career development literature based on goal-oriented motivation theories (Savickas et al., 1996).

For the exploration exercise (Appendix B), participants imagined and described three possible and positive future work selves. Next, participants wrote down what they may need, in terms of materials, skills, or abilities, to become more like each possible future work self. Lastly, participants wrote down six actions they plan to initiate in the next two weeks to further discover these possible selves. The exercise was based on the future work self-salience prompt from Strauss and colleagues (2012) and is in line with possible selves theories.

Both exercises took 10–15 min and participants received an overview of their answers.

The measures used a 7-point Likert scale (“strongly disagree” and “strongly agree”). See Appendix C for all items and Tables 1 and 2 for Cronbach’s alphas. For the analyses, scale scores were calculated for all variables as the average of the scale items. Career commitment was measured with four career identity items from the Career Commitment Measure (Carson and Bedeian, 1994). Perceived labor market demand was measured with four items from Wanberg et al. (2002). Career goal awareness was measured with six items from Gould’s (1979) career planning scale. Career option awareness was measured with five items from Germeijs and De Boeck’s (2003) career indecision scale, which were reverse-scored to calculate career option awareness. Qualitative job insecurity was measured with four items from Langerak et al. (2022), rephrased from questions into statements. Exploitation/exploration actions were measured two weeks after the intervention by asking to what extent respondents had fulfilled their intended actions.

Power analyses (Gpower version 3.1.9.7), including two groups and one moderator, indicated 50 participants per group (150 in total) to have 90% power to detect a medium effect (f² = 0.15) with an alpha level of 5% (critical F = 2.70). Inclusion criteria were fluency in Dutch and working at least 20 h per week. We recruited participants ^[1] in November and December 2021: 121 participants via social networks of six research assistants and 135 more via Prolific. The Prolific sample was paid, and the other sample could win a museum pass. We combined both samples as relationships between the study variables did not significantly differ between the samples (Box’s M = 63.36, p = 0.28).

352 participants started the first survey. Of these participants, 256 finished the first survey, completed the exercise seriously ^[2], and met the inclusion criteria (72.7%). These responses were distributed over the exploitation group (n = 78, 30.5%), the exploration group (n = 73, 28.5%) and the control group (n = 105, 41.0%). Mean age was 35.2 years (SD = 11.7) and 50.8% was female. The sample varied in education level (8.6% high school, 13.7% vocational education, 44.9% bachelor’s degree, 31.3% master’s degree, 1.6% doctorate degree). Regarding contract type, 62.5% had a permanent contract, 24.6% had a temporary contract, 5.1% had a flexible contract, and 7.8% were self-employed. Response rates for the follow-up surveys were: N_{exploitation T2} = 57 (73.1%), N_{exploitation T3} = 53 (67.9%), N _{exploration T2} = 63 (86.3%), N _{exploration T3} = 50 (68.5%), N_{control T2} = 85 (81.0%), N_{control T3} = 71 (67.6%). A chi-square test showed lower T2 attrition in the control group, χ²(2) = 6.33, p = 0.04. T3 attrition was equal across groups, χ²(2) = 0.58 p = 0.75.

Participants were recruited via Prolific in May and June 2023 ^[3] with the same inclusion criteria as Study 1. Participants from Study 1 could not participate again. 276 started the first survey. Of these participants, 212 participants finished the first survey, completed the exercise seriously ^[2], and met the inclusion criteria (76.8%). These responses were distributed over the exploitation group (n = 70, 33.0%), exploration group (n = 62, 29.2%) and control group (n = 80, 37.7%). Mean age was 31.4 years (SD = 7.9) and 36.8% was female. The sample varied in education level (6.1% high school, 10.8% vocational education, 46.2% bachelor’s degree, 34.9% master’s degree, 1.9% doctorate degree) and contract type (60.4% permanent contract, 24.1% temporary contract, 6.1% flexible contract, 9.4% self-employed). Response for follow-up surveys was N_{exploitation T2} = 56 (80.0%), N_{exploitation T3} = 52 (74.3%), N_{exploration T2} = 51 (82.3%), N_{exploration T3} = 46 (74.2%), N_{control T2} = 70 (87.5%), N_{control T3} = 63 (78.8%). Chi-square tests showed that attrition did not differ significantly across groups, χ²(2) = 1.98, p = 0.37 (T2), χ²(2) = 2.20, p = 0.36 (T3).

Tables 1 and 2 show descriptives, correlations and alphas of the Study 1 and Study 2 variables. To test the indirect effects (H1a and H2a) and the first-stage moderation of career commitment (H1b and H2b), we used Model 7 from PROCESS 3.3 (Hayes, 2017) in SPSS. We used Model 14 to test the indirect effects (H1a and H2a) and the second-stage moderation of perceived labor market demand (H1c and H2c). We created dummy variables to compare groups and mean-centered all independent variables. Results are presented in Figure 3 (exploitation-based career planning) and Figure 4 (exploration-based career planning).

Exploitation-based career planning: In Study 1, the exploitation group showed higher career goal awareness than the control group (b = 0.55, SE = 0.16, p < 0.01, 95% CI: [0.24, 0.85]), and workers with higher career goal awareness experienced less qualitative job insecurity (b = −0.56, SE = 0.08, p < 0.01, 95% CI: [−0.72, −0.39]). Supporting H1a, the negative indirect relationship from exploitation to job insecurity via career goal awareness was significant (b = −0.31, SE = 0.09, p < 0.01, 95% CI: [−0.50, −0.14]). However, Study 2 did not replicate these findings as the results showed no difference in goal awareness between groups (b = 0.03, SE = 0.18, p = 0.89, 95% CI: [−0.33, 0.38]; H1a not supported). There was a negative relationship between career goal awareness and qualitative job insecurity (b = −0.41, SE = 0.09, p < 0.01, 95% CI: [−0.59, −0.23]).

Exploration-based career planning: Study 1 results showed no differences in career option awareness between the exploration and control group (b = 0.14, SE = 0.17, p = 0.41, 95% CI: [−0.19, 0.47]; H2a not supported). There was, as expected, a negative relationship between career option awareness and qualitative job insecurity (b = −0.64, SE = 0.09, p < 0.01, 95% CI: [−0.81, −0.47]). Study 2 results also showed no difference in career option awareness between the exploration and control group (b = −0.06, SE = 0.17, p = 0.74, 95% CI: [−0.40, 0.29]; H2a not supported). There was again a negative relationship between career option awareness and qualitative job insecurity (b = −0.54, SE = 0.10, p < 0.01, 95% CI: [−0.75, −0.34]).

Exploitation-based career planning: Study 1 results showed a first-stage moderation of career commitment in the opposite direction than expected at the 90% confidence level (b = −0.22, SE = 0.12, p = 0.06, 90% CI: [−0.41, −0.02]; H1b not supported): The exploitation exercise increased career goal awareness for workers with an average (b = 0.56, SE = 0.16, p < 0.01, 95% CI: [0.25, 0.86]) or low career commitment (i.e. M-1SD; b = 0.85, SE = 0.22, p < 0.01, 95% CI: [0.41, 1.28]) but did not show a significant effect on career goal awareness for participants highly committed to their career path (i.e. M+1SD; b = 0.27, SE = 0.22, p = 0.22, 95% CI: [−0.17, 0.70]). Career commitment had a positive direct relationship with career goal awareness (b = 0.45, SE = 0.09, p < 0.01, 95% CI: [0.28, 0.62]).

Exploration-based career planning: Study 1 results indicated no first-stage moderation of career commitment (b = −0.00, SE = 0.14, p = 0.98, 95% CI: [−0.27,0.27]; H2b not supported). There was a positive direct relationship between career commitment and career option awareness (b = 0.25, SE = 0.09, p < 0.01, 95% CI: [0.07, 0.43]).

Exploitation-based career planning: Study 2 results indicated no second-stage moderation of perceived labor market demand (b = 0.03, SE = 0.07, p = 0.17, 95% CI: [−0.11, 0.17]; H1c not supported). There was a direct negative relationship between perceived labor market demand and qualitative job insecurity (b = −0.31, SE = 0.08, p < 0.01, 95% CI: [−0.47, −0.15]).

Exploration-based career planning: Study 2 results showed no second-stage moderation of perceived labor market demand (b = 0.08, SE = 0.07, p = 0.30, 95% CI: [−0.07, 0.22]; H2c not supported). There was no significant direct relationship between perceived labor market demand and qualitative job insecurity (b = 0.01, SE = 0.08, p = 0.93, 95% CI: [−0.15, 0.17]).

We explored whether acting on intentions formulated during the exploitation and exploration exercises may influence effectiveness. Because the control groups did not formulate actions, actions could not be analyzed as a moderator between the intervention and career goal/option awareness. Instead, we conducted analyses using the intervention groups only.

Exploration actions: For Study 1, we applied PROCESS Model 7 to test whether exploration actions after the intervention predicted T2 career option awareness and qualitative job insecurity, controlling for T1 career option awareness. We found an indirect effect of exploration actions on qualitative job insecurity via career option awareness, regardless of career commitment (b = −0.17, SE = 0.10, p < 0.01, 95% CI: [−0.39, −0.01]). We found no direct effect of exploration actions on qualitative job insecurity (b = −0.09, SE = 0.16, p = 0.54, 95% CI: [−0.42, 0.22]), which indicates full mediation by career option awareness.

For Study 2, we applied PROCESS Model 14 using exploration actions as predictor and subsequent (T2) career option awareness as dependent, while controlling for prior (T1) career option awareness. Results showed no relationship between exploration actions and career option awareness when controlling prior career option awareness. We also found no direct effect of exploration actions on job insecurity nor a moderation of perceived labor market demand.

Exploitation actions: We applied the same procedures to test the role of exploitation actions, but included career goal awareness instead of career option awareness. For Study 1, career goal awareness directly after the intervention was unaltered by exploitation actions, and career goal awareness at T2 did not add predictive value beyond the career goal awareness measured directly after the intervention.

For Study 2, no relationships were found between exploitation actions and career goal awareness and between career goal awareness and qualitative job insecurity, when controlling prior career goal awareness. We did find a direct relationship between exploitation actions and qualitative job insecurity (b = −0.39, SE = 0.12, p < 0.01, 95% CI: [−0.64, −0.14]). We found no moderation of perceived labor market demand.

The main results found for Study 1 and Study 2 sustained when modeling job insecurity T2 and job insecurity T3 in the models rather than job insecurity T1 (see Appendix D).

Excluding the convenience sample from the analyses resulted in similar findings, albeit with larger effect sizes (see Appendix D).

Flexibilization, digitalization, and AI advancements make insight into mitigating qualitative job insecurity increasingly important. Through two intervention studies, we examined whether and how career planning can lower feelings of qualitative job insecurity and if career planning effectiveness depends on the match with workers’ context. Based on an integration of the exploitation-exploration framework (Almahendra and Ambos, 2015) and career planning literature, we identified two career planning approaches: Traditional approaches grounded in goal-directed motivation theories (Bandura, 1991; Locke and Latham, 1990; Vroom, 1964) that largely emphasize exploitation-based career planning aimed at one desired future, and possible selves theories (Markus and Nurius, 1986; Ibarra, 2004) that point to exploration-based career planning aimed at imagining and developing multiple future selves. We argued that exploitation-based career planning would only be effective during more stable contexts via strengthening goal awareness and that exploration-based career planning would only be effective during unstable contexts via strengthening option awareness. These hypotheses were investigated through testing moderating effects of career commitment and perceived labor market demand.

Against expectations, we did not find that exploitation-based career planning was effective during stable contexts and ineffective during unstable contexts. Instead, Study 1 showed that exploitation-based career planning can lower qualitative job insecurity via increased goal awareness regardless of context. However, we found no such intervention effect in Study 2. The timing of Study 1 during the COVID-19 pandemic may be a post-hoc explanation for these inconsistent results. The pandemic could have functioned as a career shock that made workers more receptive to career planning exercises (Akkermans et al., 2020). Prior research indicates career shocks can function as “wake up calls” that stimulate reflection and creative thinking (Chen et al., 2021; Leong et al., 2024) ^[4]. Theoretically, this could have increased workers’ concern and curiosity about their vocational future and possibilities (Savickas and Porfeli, 2012), which are considered essential for effective career adaptation activities. As respondents were rewarded for participating, Study 2 participants could have done the exercises mainly based on extrinsic motivation rather than such concern and curiosity.

In addition, against the expectation that exploration-based career planning would be effective during stable contexts and ineffective during unstable contexts, we found no intervention effects for the exploration-based career planning exercise at all. Thus, although workers actively reflected upon their career alternatives (we checked this in the responses), this did not result in more awareness of career alternatives. This could be because participants did not possess the necessary information to perceive more options than pre-intervention. This post-hoc explanation is substantiated by Study 1’s exploratory finding that workers who reported doing the exploration actions did experience more options awareness. Possibly, cognitive effort alone is insufficient, and actions are required to gain more information. This would be in line with possible selves theories’ proposition that workers should actively explore their options in practice (Markus and Nurius, 1986; Ibarra, 2004). However, this exploratory finding could also be the result of self-selection bias. Participants who acted on their formulated intentions may differ systematically from those who did not, which could explain the different outcomes. For example, more proactive individuals may be more likely to continue the survey and score higher on awareness than less proactive individuals (Jiang, 2017).

While the interventions did not reliably increase goal and option awareness, we did consistently find the expected negative relationships between career goal/option awareness and qualitative job insecurity, which sustained over six weeks. As such, whereas prior job insecurity research predominantly focused on quantitative job insecurity, this study illustrates two mechanisms through which qualitative job insecurity may be mitigated. This could mean that goal-directed – “exploitation” – approaches (Bandura, 1991; Locke and Latham, 1990; Vroom, 1964) and possible selves – “exploration” – approaches (Markus and Nurius, 1986; Ibarra, 2004) can be applied not only to understand career development and quantitative job insecurity but to further our understanding of qualitative job insecurity as well. As these preliminary findings are correlational, more research is required to investigate how to manipulate levels of career goal/option awareness.

The duality of exploitation and exploration can be observed within the career development literature, with goal-directed motivation theories (Bandura, 1991; Locke and Latham, 1990; Vroom, 1964) emphasizing preparing for one desired future, and possible selves theories (Markus and Nurius, 1986; Ibarra, 2004) emphasizing imagining and developing multiple future selves. Based on our findings, we cannot provide a conclusive answer regarding whether contextual match is of importance for career planning effectiveness. On the one hand, the lack of moderation effects may imply that the proposed suitability with stable (for exploitation) and unstable (for exploration) contexts may not apply to careers in the same way as it does for organizations. On the other hand, the null findings regarding the contextual match hypotheses may also be explained by intervention characteristics that created the general absence of intervention effects. The correlational findings show potential for both exploitation and exploration within careers, as both more career goal awareness and more career option awareness relate consistently to lower qualitative job insecurity.

To foster careers in which workers are less worried about losing valued work features, we make three recommendations. First, it is important that career orientation programs and career counselors help workers not only with clarifying their career goal but also with creating awareness about multiple career possibilities. Second, it is important to provide support with career planning efforts during challenging contexts. Prior research shows that contextual challenges can limit or stop job crafting attempts (Lazazzara et al., 2020), while at the same time, our research indicates that career planning may be especially impactful when workers experience challenging times. Third, when offering career planning exercises, it may be essential that workers feel concern about their vocational future, curiosity about their multiple possibilities, and are able to actively explore possibilities outside the training setting. Future research is required to assess whether if these criteria are met, the career planning exercises of this research (see Appendices A and B) may have reliable results after all.

Our research has several limitations. First, as we expected that exploration-based career planning and exploitation-based career planning would be effective in different contexts, we did not investigate how to combine exploitation and exploration in a hybrid career planning exercise (Papachroni and Heracleous, 2020). A hybrid task could involve making a goal-oriented plan based on current experience and knowledge (exploitation) while also developing alternative scenarios for potential obstacles (exploration). Future research could investigate to what extent a hybrid task can enable workers to plan their careers in a more ambidextrous manner and whether this may lower qualitative job insecurity.

Second, this study approached career planning as a form of promotion-focused job crafting, as it was hypothesized to create resources (more goal and option awareness) and realize gains (in security regarding job features). Meta-analytic evidence shows that promotion-focused job crafting approaches are especially effective for workers with promotion-focused personalities (Lichtenthaler and Fischbach, 2019). However, prevention-focused job crafting can also be used to avoid potential job feature losses and may align better with prevention-focused personalities. Future research could investigate workers’ personalities, the type of goals or scenarios workers plan (i.e. promotion or prevention focused) and whether this influences the effectiveness of the intervention.

Third, in addition to the moderators investigated in this study, other factors may determine the effectiveness of the career planning exercises. For instance, rather than the emotional attachment to one’s current career path, it may be more important that workers are open to alternative career paths, as is better reflected in protean career orientation (Briscoe and Hall, 2006). In addition, rather than the general perceived demand for one’s line of work, the expectation that one can maintain one’s own work may be more important, pointing to the role of employability.

Lastly, future research may examine whether our interventions are effective in different samples (e.g. graduating students) or when using other modes of delivery (e.g. face-to-face). Such research could use pre-registration and more extensive manipulation checks. For example, our research left open the possibility that the exploitation-based career planning exercise affects option awareness rather than goal awareness (and vice versa). However, the correlations (see Tables 1 and 2) do not show any indication of this possibility. In fact, doing the exploitation exercise was negatively related to T2 option awareness in Study 2: The goal-oriented focus of the intervention may have shifted awareness away from considering options. Nevertheless, we advise future researchers to design and pre-register their intervention studies with manipulation checks in place to have stronger evidence for the argued mechanisms.

This study showed that the duality of exploitation and exploration can be observed in career planning approaches and provides preliminary support for the importance of awareness of both goal and options for reducing feelings of qualitative job insecurity. We encourage future research to further investigate career planning exercises aimed at fostering such awareness to contribute to careers in which workers are less worried about losing valued work features.

The supplementary material for this article can be found online.