Abstract

Most relationships do not end in conflict; they fade for lack of contact. This article reviews four bodies of evidence. First, relationships, friendships and bridging ties in particular, weaken unless contact and effort keep them up (Roberts & Dunbar, 2011; Roberts & Dunbar, 2015; Burt, 2002). Second, both the number of relationships people can sustain and their memory for details are limited, and handing information to an external aid reduces that load (Hill & Dunbar, 2003; Murre & Dros, 2015; Risko & Gilbert, 2016). Third, there is a reliable gap between intending to get in touch and actually doing so, and plans tied to a specific moment narrow it substantially (McDaniel & Einstein, 2000; Gollwitzer & Sheeran, 2006). Fourth, people underestimate both the value of reviving a dormant tie and how much the other person appreciates hearing from them (Levin et al., 2011; Liu et al., 2023). On this evidence, Mars combines relationship records, a relationship list ordered by most recent interaction, reminders tied to a time, and pre-meeting briefs with follow-up drafts. We also state the limits of this evidence: it supports the design direction and is not an experimental test of Mars itself.

Keywords: relationship decay · weak ties · dormant ties · cognitive offloading · prospective memory · implementation intentions · relationship management · personal CRM

1. Introduction

The link between social relationships and health is well documented. A meta-analysis of 148 studies with 308,849 participants found about a 50% greater likelihood of survival over the follow-up period for people with stronger social relationships (Holt-Lunstad et al., 2010). The value is not confined to the closest circle: weaker acquaintances are often the source of new information and opportunities at work (Granovetter, 1973; Rajkumar et al., 2022), and everyday interactions with weak ties are associated with greater well-being (Sandstrom & Dunn, 2014).

Yet keeping relationships alive takes sustained attention and time, the two resources people have least of. This article asks why relationships fade, under what conditions a timely follow-up prevents it, and how those answers translate into the design of a relationship assistant.

Method: this is a narrative literature review. We favoured peer-reviewed work accessible through a DOI, and preferred meta-analyses, longitudinal studies and large field experiments. Every factual statement carries its source, and the full reference list with original links follows the article.

2. Relationships decay without contact

In a study of the social networks of 251 women, Roberts and Dunbar found that the closer people felt to someone, the shorter the time since they had last been in touch; their results suggest that friendships cost more to maintain than kin relationships (Roberts & Dunbar, 2011). Following entire personal networks for 18 months across a major life transition, they found that family relationships and friendships changed in strikingly different ways, and that the decline in friendship quality was mitigated only by more effort invested in the relationship: more frequent contact for women, more activities together for men (Roberts & Dunbar, 2015).

The same pattern appears in work networks. In four years of network data on bankers in a large organization, Burt found that bridge relationships, those linking people who are otherwise unconnected, decayed faster than other ties: about nine in ten bridges present one year were gone the next, while decay was slower for more experienced bankers (Burt, 2002). In friendship research, four maintenance behaviours, positivity, supportiveness, openness and interaction, were identified, and both friends’ self-reported maintenance predicted the pair’s satisfaction and commitment (Oswald et al., 2004).

Taken together, relationships fade mainly through prolonged absence of contact rather than through a single conflict; conversely, steady, not necessarily frequent, contact is a precondition for keeping them.

Mars builds on this finding: its Relationships list is sorted by most recent interaction, and each row shows the latest item that matters for that person, so a relationship drifting into a long silence is visible at a glance (Roberts & Dunbar, 2015; Burt, 2002).

3. We cannot remember it all, or attend to everyone

The popular claim that people can keep track of stable relationships with only about 150 others comes from Dunbar’s extrapolation of the relationship between relative neocortex size and group size in primates to modern humans (Dunbar, 1993); in a study of Christmas card exchange, maximum network size averaged 153.5 people (Hill & Dunbar, 2003). The number is contested: a re-analysis with other phylogenetic methods produced 95% confidence intervals of 4–520 and 2–336, concluding that specifying any one number is futile (Lindenfors et al., 2021). Whatever the exact limit, the time and attention people can give to relationships are finite.

Memory itself fades. A modern replication of Ebbinghaus’ classic experiment reproduced his forgetting curve, in which forgetting is steepest soon after learning and then levels off (Murre & Dros, 2015). News heard at a meeting, a family member someone mentioned, their plans: unless recorded, they soon blur.

Cognitive science calls the use of external aids to reduce the cognitive demand of a task cognitive offloading (Risko & Gilbert, 2016). Writing a to-do and setting a reminder are examples. The catch is that manual record-keeping is itself a burden: most people do not write notes after every conversation.

This is why Mars does not ask you to keep notes: it extracts what you learn about people from your conversations and chat screenshots, files it into groups such as Basic Info, Family, Business & Team, and Mindset & Preferences, and links every fact back to the original words (Risko & Gilbert, 2016).

4. The gap between meaning to get in touch and doing it

“Let’s catch up sometime” is a prospective memory task: remembering to do something at a future moment. The multiprocess framework holds that retrieving such intentions relies either on effortful monitoring or on a cue that appears at the right moment; when the cue coincides with the moment to act, people are more likely to remember (McDaniel & Einstein, 2000).

Research on implementation intentions gives more direct evidence. Plans that link an anticipated situation to a response (“whenever situation X arises, I will do Y”) hand control of the response to that situational cue, which then triggers it when it is encountered (Gollwitzer, 1999), and a meta-analysis of 94 independent tests reported a medium-to-large effect (d = 0.65) (Gollwitzer & Sheeran, 2006).

Simply meaning to follow up is therefore not enough. It works better to tie the follow-up to a specific time or situation and to be prompted at that moment.

Mars follows this finding: plans and times mentioned in your chats become events on your calendar, with reminders before meetings, deadlines and birthdays you saved, and a reminder can carry a prompt that Mars starts running at the set time, so the intention arrives together with its moment (McDaniel & Einstein, 2000; Gollwitzer & Sheeran, 2006).

5. Reconnecting is worth more than we think

Network research has long held that ties must be maintained to keep their value. Levin and colleagues showed the opposite for dormant ties, former contacts now out of touch: Executive MBA students who consulted dormant contacts about an important work project got outcomes that compared favourably with those from their current ties, and reconnecting formerly strong ties brought all four benefits usually split between weak ties (efficiency and novelty) and strong ties (trust and shared perspective) (Levin et al., 2011).

For large-scale causal evidence, Rajkumar and colleagues analysed randomized experiments that varied the prevalence of weak ties in the networks of over 20 million people over five years. The effect of tie strength on job transmission followed an inverted U: moderately weak ties, measured by mutual connections, created the most job mobility (Rajkumar et al., 2022), refining the weak-tie theory (Granovetter, 1973).

What holds people back is often a misjudgment. In a series of preregistered experiments, Liu and colleagues documented a robust underestimation of how much people appreciate being reached out to, and the gap was larger when the contact came as more of a surprise, including between more socially distant people (Liu et al., 2023). Within a conversation, people who ask more questions, particularly follow-up questions, are better liked by their partners (Huang et al., 2017).

On this evidence, Mars gathers a person’s details, recent insights, open reminders and past conversations into one brief before a meeting or call, and drafts a follow-up that fits your tone and what you actually agreed, with follow-up questions when you are not sure what to ask; sending is always your decision (Liu et al., 2023; Huang et al., 2017).

6. From evidence to design

These findings point to one conclusion: a tool that truly helps people keep relationships must remember details for them, show which relationships are fading, prompt at the right moment and lower the cost of reaching out again. Mars meets each requirement as follows:

1. Limited memory, unsustainable manual notes → relationship records without data entryMars extracts what you learn about people from your conversations and chat screenshots, files it into groups such as Basic Info, Family, Business & Team, and Mindset & Preferences, and links every fact back to the original words. This is cognitive offloading that takes no extra effort (Risko & Gilbert, 2016; Murre & Dros, 2015).
2. Relationships decay as contact falls → a Relationships list ordered by most recent interactionThe Relationships list is sorted by most recent interaction, and each row shows the latest item that matters for that person. One glance shows who you have not been in touch with for a long time, so you can reach out before the relationship visibly fades (Roberts & Dunbar, 2015; Burt, 2002).
3. Intentions need a moment → events and reminders tied to a timePlans and times mentioned in your chats become events on your calendar, with reminders before meetings, deadlines and birthdays you saved; a reminder can also carry a prompt that Mars starts running at the set time. This turns “let’s catch up sometime” into an implementation intention: at a given moment, do a given thing (McDaniel & Einstein, 2000; Gollwitzer & Sheeran, 2006).
4. Reaching out is worth more than we think → pre-meeting briefs and follow-up draftsBefore a meeting or call, Mars gathers the person’s details, recent insights, open reminders and past conversations into one brief; when you follow up, it drafts a message that fits your tone and what you actually agreed, and suggests follow-up questions when you are not sure what to ask. It lowers the cost of starting the conversation, while sending is always your decision (Levin et al., 2011; Liu et al., 2023; Huang et al., 2017).

7. Limitations

First, the studies cited here test general regularities of human relationships, not Mars itself. They support our design direction but do not prove that using Mars produces the same effects; the product’s effect needs its own evaluation.

Second, several studies draw on specific populations, such as 251 women (Roberts & Dunbar, 2011), bankers in one organization (Burt, 2002), Executive MBA students (Levin et al., 2011) or users of one professional networking platform (Rajkumar et al., 2022), so generalizing to other cultures and settings calls for care. Some findings are correlational and do not by themselves establish causation (Holt-Lunstad et al., 2010; Sandstrom & Dunn, 2014).

Third, figures such as “150 people” are often over-cited. We use it only to illustrate that relationship capacity is limited, not as a design parameter (Lindenfors et al., 2021).

Fourth, more reminders are not always better. Mars builds its reminders around the people, plans and times you yourself mention in conversation, so that limited attention goes to the relationships that matter.

8. Conclusion

Relationships usually fade quietly: no conflict, just a long silence. The research shows that limits of memory and attention, the gap between intention and action, and an underestimate of the value of reconnecting together drive this decline, and that external memory, cues tied to the right moment and a lower cost of reaching out are the matching remedies. Mars relationship records, the relationship list, reminders and follow-up drafts are designed on this evidence.

References

Listed in APA 7th edition style, each with its original DOI link.

1. Burt, R. S. (2002). Bridge decay. Social Networks, 24(4), 333–363. https://doi.org/10.1016/S0378-8733(02)00017-5
2. Dunbar, R. I. M. (1993). Coevolution of neocortical size, group size and language in humans. Behavioral and Brain Sciences, 16(4), 681–694. https://doi.org/10.1017/S0140525X00032325
3. Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54(7), 493–503. https://doi.org/10.1037/0003-066X.54.7.493
4. Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. https://doi.org/10.1016/S0065-2601(06)38002-1
5. Granovetter, M. S. (1973). The strength of weak ties. American Journal of Sociology, 78(6), 1360–1380. https://doi.org/10.1086/225469
6. Hill, R. A., & Dunbar, R. I. M. (2003). Social network size in humans. Human Nature, 14(1), 53–72. https://doi.org/10.1007/s12110-003-1016-y
7. Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7(7), e1000316. https://doi.org/10.1371/journal.pmed.1000316
8. Huang, K., Yeomans, M., Brooks, A. W., Minson, J., & Gino, F. (2017). It doesn’t hurt to ask: Question-asking increases liking. Journal of Personality and Social Psychology, 113(3), 430–452. https://doi.org/10.1037/pspi0000097
9. Levin, D. Z., Walter, J., & Murnighan, J. K. (2011). Dormant ties: The value of reconnecting. Organization Science, 22(4), 923–939. https://doi.org/10.1287/orsc.1100.0576
10. Lindenfors, P., Wartel, A., & Lind, J. (2021). ‘Dunbar’s number’ deconstructed. Biology Letters, 17(5), 20210158. https://doi.org/10.1098/rsbl.2021.0158
11. Liu, P. J., Rim, S., Min, L., & Min, K. E. (2023). The surprise of reaching out: Appreciated more than we think. Journal of Personality and Social Psychology, 124(4), 754–771. https://doi.org/10.1037/pspi0000402
12. McDaniel, M. A., & Einstein, G. O. (2000). Strategic and automatic processes in prospective memory retrieval: A multiprocess framework. Applied Cognitive Psychology, 14(7), S127–S144. https://doi.org/10.1002/acp.775
13. Murre, J. M. J., & Dros, J. (2015). Replication and analysis of Ebbinghaus’ forgetting curve. PLoS ONE, 10(7), e0120644. https://doi.org/10.1371/journal.pone.0120644
14. Oswald, D. L., Clark, E. M., & Kelly, C. M. (2004). Friendship maintenance: An analysis of individual and dyad behaviors. Journal of Social and Clinical Psychology, 23(3), 413–441. https://doi.org/10.1521/jscp.23.3.413.35460
15. Rajkumar, K., Saint-Jacques, G., Bojinov, I., Brynjolfsson, E., & Aral, S. (2022). A causal test of the strength of weak ties. Science, 377(6612), 1304–1310. https://doi.org/10.1126/science.abl4476
16. Risko, E. F., & Gilbert, S. J. (2016). Cognitive offloading. Trends in Cognitive Sciences, 20(9), 676–688. https://doi.org/10.1016/j.tics.2016.07.002
17. Roberts, S. G. B., & Dunbar, R. I. M. (2011). Communication in social networks: Effects of kinship, network size, and emotional closeness. Personal Relationships, 18(3), 439–452. https://doi.org/10.1111/j.1475-6811.2010.01310.x
18. Roberts, S. G. B., & Dunbar, R. I. M. (2015). Managing relationship decay: Network, gender, and contextual effects. Human Nature, 26(4), 426–450. https://doi.org/10.1007/s12110-015-9242-7
19. Sandstrom, G. M., & Dunn, E. W. (2014). Social interactions and well-being: The surprising power of weak ties. Personality and Social Psychology Bulletin, 40(7), 910–922. https://doi.org/10.1177/0146167214529799