Goal setting theory has a highly empirical foundation. Working on creating clear objectives channels individual focus toward high-impact priorities, identifies effort required, helps build, builds mental toughness, and directs individuals to find more effective operational strategies [11]. This isn’t just theoretical intuition. A meta-analysis of 141 randomised controlled trials showed that setting a goal acts as a catalyst for behavioural transformation, independent of supplementary training modules [12].

Identifying and naming goals is just the starting point. A meta-analysis of 138 experiments and nearly 20,000 subjects demonstrated that actively tracking progress through specific milestones improves goal attainment, even more so when progress was either documented or shared with someone else [13]. The gap between planning to do something and doing it, reduces dramatically when individuals truly formalise the when, where, and how of their actions [14].

Such benefits also go far beyond metrics or standard organisational tasks. For example, students who committed to focused time to writing down and reflect on personal ambitions went on to achieve higher academic marks compared to control groups [15]. In the business world goal-oriented frameworks form the absolute bedrock of modern, evidence-based workplace coaching [16].

A further input to the discussion is the meta-analysis on corporate coaching environments which found huge benefits across individual performance, emotional wellbeing, psychological coping mechanisms, career attitudes, and self-regulation [17]. One thing of real note is that in building goals the evidence also suggests that not one size fits all [18]. Goals need to be personally connected to the individual: “develop goal setting programs that place a strong emphasis on the characteristics, needs, preferences, and goal setting styles of individual athletes.” [19]

From this research, four important behavioural Science principles emerged as the foundation for building the app, how it would work and how the results would be interpreted (See Table3).

Feedbackmustbeactionoriented. It is known that evaluation improves output yet across approximately a third of historical studies, it degraded performance. This can occur when feedback makes the individual hyper focused on personal self-worth rather than tactical execution [20]. To avoid this, the IMI app translates numeric scores straight into steps for action instead of making a judgement or a grade.Loggingandauditingmilestonesispartoftheactualcure. Written, milestones offer a massive advantage over vague, unwritten ambitions [21][22]. Hence, checking in on the app is much more than a data-collection exercise as it keeps the personal objectives active in a user’s mind.Neurologicalrewiringrequirestime! Lally et al. [23] found a new daily habit took on average 66 days to become automatic with a wide individual variation. An eight-week study is therefore considered long enough to identify early change for some participants but not to establish whether new habits are sustained. This is addressed in section 8. Individualandenvironmentalcontextremainsakeyvariable. Professional development sticks more effectively smoothly when there is a positive and encouraging culture. This is especially critical for complex “open skills” such as leadership, which has significant unpredictability [24]. This mirrors the original 2025 study’s findings where organisational stability and manager buy-in was critical for the intervention’s success.

Table 3. Goal-setting principles and how the IMI App applies them.

The study was designed to run for an eight-week period for a single-group, pre-post design, with the pre survey in July 2026, and a post survey in September 2026. Both surveys combined numeric ratings and open-ended questions.

Participants were recruited through the researcher’s professional network and XVenture client and alumni over the last decade. The researcher targeted 30 participants, 50:50 male /female from Countries worldwide. 30 participants enrolled. No two participants worked for the same organisation. The researcher was specifically looking for people who had some interest in personal development. The eligibility criteria included being over 18, owning a smart phone, executive or supervisory positions or elite sport level and be willing to commit to up to ten minutes a week, twice a week with no prior knowledge of the IMI app. Participation was voluntary and unpaid. AppendixB provides the breakdown of the cohort.

Figure 1. IMI app overview.

Measures

The pre survey and post survey presented twelve-item self-rating indices to track IMI changes on a scale of 1 - 10. The additional behavioural items in the two surveys were worded differently: the pre-survey asked about usual practice. (e.g. “I set specific goals for my personal and professional improvement and track them.”) the post-survey asked about behaviour in the past week (e.g. I have taken actions this week aligned with my personal and professional goals.) A 1 - 5 scale was used for this. The pre survey also obtained some contextual data on an individual’s previous experiences of similar digital apps and a question seeking individuals to define peak personal performance. The Week 8 post-survey asked thirteen different questions relating to contemporary workplace behaviour along with four questions evaluating app usability and usage frequency.

Users were also asked to identify five key improvements from twenty choices, ranking these in order relating to individual impact. They were also asked to provide a written explanation for their choices. The final section of the survey asked participants to describe their largest personal shift and to rate how likely they would be to recommend the app (0 - 10).

Individual entries of pre- and post-survey datasets were paired using corresponding email identifiers. To ensure anonymity, all participants were discussed exclusively via alphanumeric markers (P01-P21). Participant changes were measured from week 1 to 8 and several tools were used to evaluate their significance and effect sizes effect sizes including Cohen’s d for paired configurations using the benchmark framework where 0.2 was a minor effect, 0.5 represented a moderate effect, and 0.8 signified a significant effect [26].

Looking at 12 different metrics at the same time, the more things you test, the easier it is to accidentally find a false alarm. As twelve category scores were tested at the same time, the Holm-Bonferroni correction [27] was applied to manage the risk of false positives across the 12 tests. The analyses focussed on the 21 completed surveys.

In treating the qualitative fields, all commentaries were reviewed thoroughly and linked to corresponding IMI categories and themes.

From the original group of 30, 21 completed both the pre and post surveys, which is a 70% retention rate. This compares favourably with self-reported digital interventions. Fleming and colleagues’ review of digital self-help interventions reported users ranging from 0.5% to 28.6% [28]. A meta-analysis by Torous and colleagues on mobile phone apps on symptoms of depression reported a range between 26.2% and 47.8% [29]. Eysenbach’s work [30] clearly shows that over time the law of attrition is likely to apply for eHealth interventions. A further study of consumer mental health platforms reported a retention rate of 3.3% for users logging into an app 30 days after installation [31]. This study focussed on public downloads rather than a structured research study so should be considered with caution. However, it does highlight the challenge in achieving long-term engagement particularly with self-directed programs. The researchers believe that there were several reasons for the high retention rate: the work that took place to identify the participants, the manageable size of the cohort, the type of participant targeted, specifically with an interest in personal development and the detailed briefing undertaken including reference to the 8-week commitment. Setting out and communicating all the aspects of the 8-week program was critical and was obtained early and despite this, there was still a 30% dropout.

For 17 of the 21 active users, the study was their very first experience of being involved in a mindset or psychological tracking tool.

Based on the participants own estimates, 12 people logged into the application roughly once a week, 5 used the app twice or more each week, while 4 used it less than once a week.

The 12 core IMI questions demonstrated good internal consistency, demonstrating that the items in the IMI are inter-related. (Cronbach’s alpha of 0.81 at baseline and 0.87 by Week 8) [32].

The aggregate IMI metric, calculated as the mean across all twelve measured categories, increased from 7.28 to 7.86, an increase of 0.58 points (8.0%), which was statistically significant. Looked at individually, 18 of the 21 participants registered growth. Within this group, 9 users improved by 0.70 points or higher, 4 improved from 0.40 and 0.69 points, 5 recorded minor progress, and 3 individuals experienced a decline. Figure 2 presents the scores from pre to post survey.

Each line is one participant, coloured by size of change. The heavy line shows the group mean.

Figure 2. Individual composite IMI scores at Week 0 and Week 8 (n = 21).

The data suggests that participants with lower initial baselines achieved the most improvement though the patterns should be interpreted cautiously. Participants with low initial scores would be expected to move closer to the average on re-testing and without a control group this can’t be separated from any effect of the app. Those starting above the median averaged 0.42 points. It’s worth noting that the single largest individual improvement (+1.92 points) was by a participant who recorded the lowest starting score in the study.

Individual metrics showed that eleven of the twelve categories improved, with the largest shift in Reflection & Feedback, which improved by 1.29 points (21%) and saw 14 individual users improve. Social Skills & Communication improved by 0.76 points with no reduction for any participant. Other positives included: Self-Awareness, Self-Management, and Learning & Development. Metrics for Decision-Making, Resilience, and Empathy each improved by around 0.6 points. Growth Mindset was the only category that did not improve with a starting mean of 9.10 out of 10, this may be explained by a ceiling effect. After Holm correction, social skills and communication remain statistically significant while the changes in other categories including reflection and feedback should be seen as indicative (Table 4 and Figure 3).

Table 4. IMI dimension scores at Week 0 and Week 8 (n = 21).

Figure 3. Mean rating on each IMI dimension at Week 0 and Week 8, ordered by size of change.

At the beginning of the study, the pre-survey showed that goal setting was not common practice for most individuals, with less than half (48%) actively building and tracking personal objective and less than 40% had anything in place when they felt stressed, overwhelmed or stuck. By Week 8, when participants were asked about noticeable changes in behaviour over the last week, with 20 of the 21 participants (95%) reported taking real action aligned to personal and professional goals and 86% reported noticing and managing their emotional reactions more effectively. e.g. Goal setting, Emotional regulation and stress management had improved between 40 and 50% and their self assessment of their own performance had improved by 14% (See Table5).

Figure 4 shows agreement with each pair of related but not identical statements at week 0 and week 8.

Week 8 focussed on previous week behaviours. Week 0 focussed on typical behaviours.

Figure 4. Agreement with related statements at Week 0 and Week 8 (n = 21).

Table 5. Examples of participant reflections at Week 8.

Other outputs from the post survey provided a similar picture, with 90% of respondents having deeper confidence in executing their duties while 86% reported feeling driven and engaged. 86% characterised their overall state of well-being as solid, and another 86% felt sharp and focused. The lowest score emerged around sleep and health, where only 57% felt rested and energised. This links with the smaller improvements seen in the Health, Wellbeing & Balance category of the IMI index.

For most of the active participants the app was seen as a real conduit for self-reflection. 86% agreed that the app helped increase their day-to-day awareness of mindset, emotions, and behaviours, while 81% saw the app helping to identify areas needing development.

71% also believed the PIMS goal-mapping provided a highly practical solution needed to take real-world action.

Participants view on the built-in AI coach, were more ambivalent; 5 individuals didn’t use the AI functionality at all, and for those who did interact with it, a 50/50 split emerged between those who found it helpful and those who remained neutral. Post survey ratings of the IMI app are provided in Figure 5.

Figure 5. Participant ratings of the IMI app post survey.

Participants were asked to choose five behavioural improvements that supported their 8-week journey. For many a deeper sense of self-awareness topped the list (67%), followed closely by structured daily habits and routines (62%), and greater interest in their own personal development (48%). Other leading indicators included improved confidence in strategic decision-making and more effective communication skills which each got 38% approval (Figure 6). Overall, satisfaction with the app was high: 18of the 21 active users (86%) would be recommending the tool at a 7 or higher out of 10, with seven of those scores 9 out of 10.

More frequent use was not associated with larger gains that participants described. Those who reported an average use of the app about once a week (n = 12) reported improvements by an average of 0.73 points, those using the app less than once a week (n = 4) by 0.56 and those using it twice or more a week (n = 5) by 0.24. The groups are small and usage is self-estimated so further research will be useful to identify the optimal participant use.

Figure 6. Improvements participants chose as most reflecting their experience.

The qualitative, open-ended responses collected at the end of the study add context to the quantitative findings. When analysing the users’ testimonies, four distinct themes materialised (Table 6).

Table 6. Main themes in participants’ comments.

The 2025 research program blended four months of digital learning, one-hour monthly recorded behavioural interviews, monthly leaders’ reviews and a range of surveys throughout. The app use took a fraction of that time for both administrators and participants: ten minutes for each session at a time to suit, in half the time (8 weeks v 4 months) delivering many similar benefits and no live facilitator to arrange. Cost effectiveness wasn’t measured explicitly, but this will be tested in future research.

In comparing the two studies outcomes, there is one important caveat. The methods for evaluation differed significantly. In the first study in 2025 there were researcher-scored qualitative interviews and in this study, user self-assessments were the data source. As a result, comparisons are structurally inappropriate even though both approaches provide a similar conclusion, in that most participants experienced improvements in both their professional environments and their personal daily routines. Table 7 provides a comparison between the original program and the latest study using the app.

Table 7. The original program and the IMI App compared.

The outcomes align with established literature on goal-setting behaviours. The largest improvement seen was in the Reflection & Feedback category. This is the category most directly prompted by the app’s assess and adjust loop. Changes mentioned by participants, such as improved self-awareness, more effective daily habits, and a proactive growth attitude, align well with goal-setting theory. The change on goal setting from week 1 to week 8 was dramatic. At the start of the study, fewer than half of the participants reported goal setting and tracking. By week 8 almost all participants completing the program reported taking goal related actions in the previous week. When reviewing actual weekly data captured in the app, it demonstrated this was occurring for most users (n = 21) every week.

Several participants explained that formalising and recording plans acted as a mental anchor that ensured goals were front of mind throughout the week, even when they weren’t using the app. This is consistent from research that shows recording of goals and progress towards goals influences behaviour in between actual monitoring [13]. This raises the possibility that short weekly check-ins may be enough for some people particularly those with demanding schedules. It was outside the scope of this study to test this. But should be considered for a larger sample.

One other thing worth drawing to attention here. The researcher has been applying Rock and Davachi AGES principles of learning in building learning programs for over a decade. Whilst a detailed review of the participants actual learning processes hasn’t been analysed for this study, they appear to fit well with the model: Attention, Generation, emotion, Spacing [5].

The initial 2025 investigation indicated that improvements in mindset thrives best within a highly stable culture supported by a dedicated, encouraging manager. Although this study did not recruit participants through a centralised organisation structure, and not two of them worked in the same place, the findings and comments regarding this aspect were very similar at the individual level. Participants who reported challenging or stressful situations were typically among the least active users over the study period. One of the three individuals who registered a reduction in overall scores noted that the study period collided specifically with personal extended summer holidays. Despite the dip, that same participant noted that the actions they set up at the beginning of the study remained valuable and intended to use them in the future.

This is aligned to research that confirms professional development translates into changing behaviour far more successfully in supportive environments [24]. For organisations looking to integrate the study’s approach, there are some practical considerations: identify individuals who have real appetite for improvement, will commit to the time required and have the capability to deal with the ups and downs of life reasonably well; plan and optimise the timing of the rollout and ensure genuine and visible advocacy from leaders. It may well be that during challenging operational periods employees will need some form of support rather than just leaving them to navigate the app in isolation.

Participants feedback on their experience was both practical and actionable. For example, one individual suggested making automated push reminders to display active goals on a device home screen, rather than having to manually copy and paste action plans into a mobile notepad. As referenced earlier, participant engagement with the built-in AI coach was sporadic with five participants skipping the feature. This suggested that onboarding procedures at the beginning of the study could have been more effective in introducing this functionality. As the lowest scoring subjects across the study related to sleep quality and physical vitality future app development should introduce actions relating to wellbeing-focus and this could be highlighted in the core PIMS menu.

A Cronbach’s alpha of 0.81 at baseline moving up to 0.87 by Week 8. Both these values suggest that the measurement scale has good internal consistency reliability, and its items are cohesive and dependable over the course of the study. Ratings tended to move together. However, a factor analysis would need to take place to demonstrate that the twelve categories measure a single construct. This would require another study over a larger sample. A score above 0.80 is universally classified as good, and higher than the standard research benchmark of 0.70.

30 participants took part in the study.

Male Female split. 50:50

From 6 countries: Australia, New Zealand, USA, UK, France, Netherlands

Participant professional backgrounds

Business C suite x 12

Business Supervisory x 7

Elite Sport x 5

Education x 3

Clinical x 3