Behaviour Apps Design - by example

Behavioural design context

A few months ago I was part of a small team that pushed a near completed app onto the shelf marked SUNSET. This happens more often than not in small self-funded start-ups. This event coincided with the completion of my PhD studies. It was an opportune moment to look long and hard at what I had learned from both experiences and apply these as I looked to "go again" with fresh IP. Some of my research work has been presented on this blog and other places.  It provides proof that you can combine existing measures of a health behaviour theory with an emergent behavioural systems design model so that you could empirically predict who will use social networking features of exercise apps and how that may positively affect persistent use of a wearable device. In a nutshell,  it effectively combined so-called "old school" behaviour change theories with nascent models and toolkits ostensibly intended to aid the development of software, that could be used as instruments of behavioural change and support. It's hard to transform this kind of work into a marketable phrase.

Nonetheless, the trick in these cases is to apply the appropriate set of knowledge and skills to a real world problem and operationalise your findings. Enter, ZING stage left..........which looks to encapsulate the learnings from both the research projects and the failed app. Staying as it does in the exercise space, ZING aims to bring an immediacy of advice, plan, mentorship, analysis, education and social support to the trigger moment when a user synchronises their fitbit data. No more guessing or the blind leading the blind here, ZING brings the certainty and expertise of exercise and sports scientists to the user's exercise management state, too often underlined by indecision and confusion.  The app provides context, direction and planning in digestible, affordable and valuable microbytes.

Nudge theory

From a behavioural design perspective, I have applied portions of Persuasive Systems Design (PSD), Nir Eyal's "HOOKED" approach, Nudge Theory and Support for Engineering Persuasive Interactive Applications (SEPIA). The trouble with these models and "theories" is the surfeit of redundancy.  You have to wonder why so many people spend so much time thinking of synonyms for interchangeable concepts and constructs. That aside, I have treated these particular models as toolsets  bound within a substrate of an overarching behaviour change theory; Self Determination Theory, with particular emphasis on the satisfaction of the 3 basic psychological needs of autonomy, competence and relatedness.

The motivation for sustainable exercise is made possible by satisfaction of 3 basic psychological needs: autonomy, competence and relatedness.  Ryan & Deci (2002)

Setting the Scene for HOOKED use

In terms of our motivation, besides problem-solving and creating a possible revenue stream, we need look no further than the following graphs and their indictment of the fitbit model wrt consumer persistence with their device to be inspired.

To establish the business operators for this app, I adopted a modified version of the HOOKED approach in looking to create a habit from a product. For ZING to succeed it has to be habitual enough to negate any investment by the business in marketing spend. The app explained in HOOKED terms:

TRIGGER

The trigger is synchronisation of fitbit data by the user which cues ZING.

ACTION

User actions are simple and hopefully compulsive - a personalised exercise plan ( as short as 7 days as long as 28 days), an instant message with the ZING exercise scientist who creates the plan and supports the user. All things ZING stem from this.

REWARD

To feed the dopamine, ZING serves up to the user fresh advice,  interactive knowledge, dynamic in situ experimentation and optional actual rewards. Research shows us that people seek out small victories. Successful people like to challenge themselves and to compete, even when their efforts yield only small victories. Small victories build new androgen receptors in the areas of the brain responsible for reward and motivation. Any increase in androgen receptors enhances the influence of testosterone, which further increases their confidence and eagerness to tackle challenges. When you achieve a series of small victories, the boost in your confidence can last for months. We tap into this phenomenon through the transparency of our scoring system, the DOJO self-knowledge path anduser-preferenced and optional variable rewards. It is also reflected in our pricing model that allows the user to avoid the industry-standard month by month and annual lock ins for something far more flexible to enable small challenges and small victories.

INVESTMENT

The ZING user accrues a digital-emotional investment in assets such as the professional relationship (ZING exercise scientist), knowledge from the academy (aka DOJO), peer-based social relationships and redeemable coupons.

In our parlance, ZING's raison d'être is to help improve an individual's physical and emotional  well-being enabling them to be self determined in planning and implementing positive, sustainable exercise behaviour.

Nir Eyal goes on to leverage some of the work of BJ Fogg of Captology fame by way of identifying the role of user motivation and ability in app design. This is a true distillation of the problem space. In applying these "principles" to the design of ZING we rifled in on key motivators as drivers of app facilitation..

Users seek pleasure and avoid pain

ZING needs to be fun and easy to use. If it is obtuse, ugly and valueless, little pleasure will be accrued. Away from the digital world, the fact is exercise for fitness often involves some levels of discomfort in order to elicit a physiological adaptation; it's not always easy to avoid pain. By realising improvement and having clinically qualified support ZING goes a long way to ensuring the exercise completed is suitable to the individual and avoid unnecessary pain. The exercise plans are individualised, not generic and are periodised to ensure steady, health adaptation.

Users seek hope and avoid fear

Just because you have a wearable and you get incessant emails exalting your badge worthiness and specialness it doesn't necessarily mean you are being fulfilled, educated and likely to sustain a self determined path. As an aside, The Atlantic had an interesting piece on fitbit abandonment  . As we have seen, fitbit looks to counter that by pushing the devices onto those who have been hiding in caves and not yet been inundated with their TV advertising. 

Let's face it, the fitbit throwaway mentality is factored into their business plans as shown in Figure 1.

In practical terms, users get bored, confused and disconnected through the use of current devices and apps. 

ZING looks to bring hope to these people by giving them a human expert at their fingertips who is there for only as and when needed. Small microbytes of exercise and health content are published and available with optional quizzing and online certification to elevate the user's knowledge and competence; raising hope through self-efficacy. Of course the ZING exercise clinician must be available, prompt, polite,empathetic in consoling the user so as to increase and hope and allay fears.

Users seek acceptance and avoid rejection

Almost everybody likes to belong. The social network aspect of exercise and fitness apps is an oft overplayed hand. It's pervasiveness needs to be subtle and relevant.  ZING provides the user with optional access to like-minded supportive peers and a means to connect, share, support, recognise, co-operate, compete with and learn from these people. Who knows being a ZINGER may be a membership people will aspire to.

Again, Fogg & Eyal make a point of stressing the role absolute simplicity of task design in encouraging habit formation. We try to  pay homage to Fogg's 6 Elements of Simplicity. as follows:

TIME

ZING is built with the aim of affording the user the fewest possible steps en route to satisfying their intentions. We try to achieve this by utilising the Persuasive Systems Design (PSD) principle of Reduction or as we like to put it building a Compressed Path. Without busying the screens, effort should be made to leverage common UI components that provide one click access to contextually related functions.

MONEY

In a world where for the user at least, Freemium is king, people like to be able to sample for free and avoid lock-in like the plague. We have created something of a blended pricing model based on market analyses of people's current preferences against a backdrop of data on gym plan madness and the reality of the gap between an individual's exercise intentions and actual behaviour. ZING is free to download and comes with a free 7 day generic exercise plan. To engage an exercise clinician individual's simply complete an In-app purchase where we proffer only 2 choices 7 or 28 days. We have no interest in conning an outlier of users into locking into a 12 month plan paid annually in the first instance. People float in and out of exercise regimes across a calendar year and our model needs to reflect that and offer convenience and quality while reducing the number of choices; in itself a decent application of Nudge Theory.

EFFORT

The less the better; we offer minimal choices but choices nonetheless.

Why fitbit SUCKS

Figure 1. Who cares about existing users?   

Figure 1. Who cares about existing users?

 

The Nitty Gritty

The screen samples provide a broad sweep of the brush in terms of function coverage. This article will examine 30+% of the current ZING app screen set.

 

Figure 2. Sample ZING mobile app screens (Android)

Figure 2. Sample ZING mobile app screens (Android)

DO TODAY

On open the app presents the user with this screen which tells them simply and clearly what they had planned to do today.

 

Figure 3. Do Today

Figure 3. Do Today

Behaviour Design principles and sources

From PSD

Reduction design - compressing information streams and actions into a single elegant screen design.

Personalisation - the plan has been custom written for the individual by the ZING exercise scientist

Social - the ever present chat-messaging system is accessed by the slide drawer menu overlay 

From SEPIA

We contribute to user understanding of behaviour by providing the 7 day tracker that also complies with the Social Cognitive Model (SCM) dictate of enabling self-regulation and feedback. This contributes to increased awareness and with it greater competence. Their last 7 day’s behaviour is observable. It is also visible to the ZING exercise clinician

The SEPIA recommend option is delivered by affording the user the opportunity to swap different exercise sessions in the plan so that it suits their current condition not the condition when the plan was agreed. From a SDT perspective by empowering the user here to commit, delay or avoid then we are encouraging autonomous decision making.

We enable an alert function via system generated prompt if the # skipped sessions rises to 2 or more. This provides them with control in order to adjust and is consistent with the self-regulatory and autonomy encouraging nature of this screen.

 

ZING & PING

Firstly, a brief explanation. ZING is a dynamic composite measure of an individual's engagement with planned exercise and dimensions of their exercise outcomes. PING is a dynamic composite measure of an individual's engagement with the online relatedness functions of the system and their own self-reported emotional state. These scores are determined by our machine learning algorithms (written-in R); presented as a simple single number and instantly amenable to what-if adjustments but the user through simple slider controls on the TWEAK screen.

Figure 4. ZING score

Figure 4. ZING score

From PSD

Reduction design - compressing information streams and actions into a single elegant screen design.

Personalisation - the scoring is individually calculated

Social

     1.    Learning

     2.    Comparison

         1.    Against the MEAN for the individual’s profile-type based on Age-Gender-BMI

         2.    Against ZING friends

     3.    Recognition via the simple displayed numbers for the scoring as well as LIKES-COMMENTS-social media shares

From NUDGE

Social conformity & anchoring by the score comparisons.

Priming for sustained positive exercise behaviour; it really sets the scene for enactment of the exercise plan and interact with the clinician and peers. A real call to arms.

Positive visual framing - the user’s scores are always larger and clearer than that of their peers and friends even if in real terms they currently have a lower score

From SEPIA

We contribute to user understanding of behaviour by providing the 7 day tracker.

We reveal their last 7 day’s exercise and positive online social behaviour making it observable. This is also visible to the ZING exercise clinician

We provide discovery; the user sees the correlation between their exercise and online sociability and their scoring.

TWEAK

This screen brings the power of the ZING machine learning backend to life in many ways putting the power of behavioural instances and their ramifications squarely with the individual. By adjusting the major (but not all) levers that influence their ZING score, the user can safely evaluate the effects. For example if skipping planned exercise sessions or  reducing or increasing exercise intensity. An animation plays to expand or contract the displayed score accordingly to the direction and magnitude of changes o the slide controls. This functionality is also available for PING scores on a separate screen.

 

Figure 5. TWEAK your scores

Figure 5. TWEAK your scores

From PSD

Reduction design - compressing information streams and actions into a single elegant screen design.

Personalisation - the scoring is individually calculated 

Simulation and rehearsal - classic what-if power at the fingertip

Suggestions - not shown here but movement of each slide control will trigger a callout bubble with text explaining the role of the measure in calculating the score and how the user can improve it. This is to be transformed with an ai implementation in 1.0 of the app

From NUDGE

Feedback and self regulation to facilitate self-efficacy and autonomy.

Priming for sustained positive exercise behaviour by providing the function to visualise positive outcomes from changes.

Positive visual framing - the user’s score uses a simple animation to grow when positive changes are made to ZING factors.

Sympathy having effects of change mirrored in a benign manner is sympathetic to the user’s situation about continuing to use a device and a mobile instant pro to help them manage sustainable exercise

From SEPIA

We contribute to user understanding of behaviour and their ability to affect it (positively or negatively).

We provide discovery; the user sees the correlation between their exercise and online sociability behaviour and efforts and their scoring.

We provide Learnability through Experimentation.

DOJO

A Japanese word which means "place of the way". 

In the West more simply a training hall. For ZING users it is both; a microbyte-style online learning place which offers optional progression of knowledge via quizzing and achievement of belts.

 

Figure 6. DOJO

Figure 6. DOJO

From PSD

Reduction design - compressing information streams and actions into a single elegant screen design.

Personalisation - the DOJO belts are awarded for individual accomplishment

Tunnelling - ZING guides user through interactive tips to show the path to belt accomplishment

Self-monitoring - the screen demonstrates the current BELT status and changes if the user accomplishes enough quiz completion to elevate this level

Rewards - the user’s improvement in knowledge is rewarded with BELT conferment 

Social role - here the app takes on the support role of tutor, augmenting the primary social role delivered by the ZING exercise scientists

From NUDGE

Framing effect - we keep the number of content carousel choices in the DOJO to a bare minimum (3).

Priming for sustained positive exercise behaviour by providing the function to visualise positive outcomes from changes.

Positive visual framing - the user’s belt grade visual  

Sympathy having the content proffered in small digestible chunks of interesting and focused content  is sympathetic to the user’s time and attention constraints

Loss aversion ZING is designed to encourage considerable investment from the user which is retained as assets - DOJO Belts, social relationships, professional relationship, reward coupons,data analysis. DOJO reflects part of this investment and seeing their BELT status is a constant reminder of that investment.

From SEPIA

We provide Engagement and Control. Users have the option to engage with learning as a means to help shape positive exercise behaviour through knowledge.

The control is afforded the user by the optioned facility to do as little or as much learning as they want. They may be able to ascertain the association between their knowledge attainment and improved levels of self determination.

NOTICES

Or what the hell is going on in my little ZING world that may help me? This screen ties together core elements of SDT to help facilitate autonomy, competence and relatedness in exercise behaviour.

 

Figure 7. NOTICES

Figure 7. NOTICES

From PSD we apply

Reduction design - compressing information streams and actions into a single elegant screen design.

Personalisation - each alert is based on the individual’s needs, performance and relationships.

Self-monitoring - the screen provides up-to-the-minute notification of achievement, performance repercussions and social activity.

Rewards - the user’s performance is recognised with the BLING function

Social role - here the app takes on the support role of a PA keeping them informed and up-to-date

Suggestions - the RHR (resting heart rate alert) which is simultaneously shared with the user’s ZING clinician includes explanatory text about what it may mean and what to do about it

Reminders - ZING applies the Zeigarnik & Goal Gradient Effects to remind the user when their current PLAN will expire that their BLING reward coupons have an expiry for redemption ( when clicked ) 

From NUDGE

Framing effect - we keep the number of content carousel choices in the DOJO to a bare minimum (3).

Priming for positive actions related to their health (RHR), education (new DOJO content)  and social interactions (noting which friends are online)

Positive visual framing - new DOJO content, the health alert, lapsing plan and BLING coupon redemption are one click actionable.

Sympathy having the content proffered in small digestible chunks of interesting and focused content  is sympathetic to the user’s time and attention constraints.

Loss aversion ZING BLING reward coupon redemptions have a short expiry period.

From SEPIA

We provide Alerts and Prevention.

The alerts are self explanatory.

The prevention comes through the plan renewal date notice and the health threatening warning about an excessive resting heart rate.

Along with PSD and SEPIA Morton & Lawson make use of gamification and the Hooked Habit Forming Model from Nir Eyal.

Solid examples of the HOOK for the app are shown on the NOTICES screen.

In terms of the Zeigarnik & Goal Gradient Effects we provide a "cliffhanger" ending by:

- reminding them when their current plan is about to expire

- notifying them when friend uploads new exercise data

From the HOOKED approach

Variable Rewards

A fundamental of habitual use ( e.g. checking email) is the need to alleviate the craving for that reward. Variable rewards are when you positively reinforce a behaviour at an non-fixed (ie less predictable) schedule. By varying when you deliver the reward for a certain behaviour and how big that reward is, you can quickly reinforce that behaviour and make it very strong and resistant to extinction (aka it becomes a habit or routine). Getting a reward increases dopamine levels in your brain, which motivates you to do the thing which got you the reward (rats with missing dopamine receptors struggle to build habits).

But dopamine isn’t just pleasure, it’s about anticipation of pleasure. When we know how the game works (this lever press won’t give us a reward, but the next one will) our novelty seeking brains get bored. But when the rewards are unpredictable, we stay on edge. Studies have shown that unpredictable rewards cause greater increases of dopamine, which may be why the behaviour that lead to the reward gets so strongly reinforced.

Some directly incorporate gambling elements, like JetPack Joyride, a popular iOS game which allowed you to collect special bonuses that could be cashed in at the end of a round  literal spin the wheel bonus points during a particular round of gameplay, which can be cashed in for wheel spins. It turns out this is not a new idea – Super Mario Bros 2 was using the same tactic 18 years prior.

According to Eyal there are 3 types to consider:

The HUNT

Actual physical rewards. The ZING user can redeem their BLING coupons at our web site for towel-swim cap-running socks etc

The TRIBE

People seek recognition via the ZING DOJO grades & ZING and PING scores. Our relatedness functions of LIKE-SHARE-Comments help our people anticipate social validation. Rewards of the tribe such as these may keep users coming back wanting more. Bandura, chief author of the Social Cognitive Model (SCM) learned that people who observe others being rewarded for behaviour 'x" are more likely to alter their actions to gain congruence better when there is homophily and the observed individual is a "role model".  In ZING - we offer up notices of friends' DOJO levels; goals reached, ZING & PING improves providing a social mirror for behaviours worth replicating by others.

The SELF

We need to reflect back their mastery, competence and completion and do so via

PLANS done

ZING and PING up

DOJOs belts earned

Our JOURNEY section on the PROFILE screen which depicts the progress in an individual's core physical measurements over time.

Conclusion

There is an awful lot of  white noise abounding in the app design space in general and the application of behavioural science principles in particular. It appears that too often the case in this land of the blind, the one-eyed man is king; bits and piece snake oil salesmen with sketchy backgrounds in Nudge Economics and UX are touting themselves as behaviouralists. Sorry but no, you are not.  To design and BUILD true behaviour change apps requires behavioural scientists who can design and code augmented with qualified skill sets and knowledge in the target industry problem domains. 

Morton and Lawson back their capabilities by tackling solution building; we don't just borrow your watch to tell you the time and leave a large fluffy report and a whopping great invoice for services rendered. 

ZING isn't the be all and end all, far from it but it shows we back ourselves and don't shy way from operationalising research.

Dr Daryl Foy

Dr Daryl Foy is a Behavioural Scientist who specialises in the design of effective health behaviour change apps based on evidence including his own validated models for optimising persistent use. He consults to industry on how-to integrate persuasive design into LEAN product development as well as conversational UI. He can be contacted at dlfoy@mortonlawson.com