Beyond SMART

Adaptive goal-setting in the context of healthcare QI and transformation

Satyan Chari
February 25, 2025

Introduction

Of the many rituals that saturate project management activities across sectors, SMART goals are perhaps the most ubiquitous. In healthcare, the SMART framework can often be found baked into everything from business planning processes, to performance reviews, and unsurprisingly, quality improvement work.

In their guidance on the quality improvement process for ambulatory care settings [1], the US Agency for Healthcare Research and Quality recommends SMART as a way of setting smaller, incremental objectives within the context of larger projects. They provide an example of a SMART objective, suggesting something tangible like a "1-year goal of a two percent increase" in a given outcome metric.

The Institute for Healthcare Improvement's materials on Improvement Science also channel a similar perspective, emphasising clarity, precision, time-limitation, achievability and measurability across various documents that talk about goal- and aim-setting. [2,3]

It is a seductive logic. And yet, something has never quite added up for me over the years. Many projects with the most meticulously crafted SMART objectives have failed to produced lasting change (some never even made it out of the gate). Equally, examples abound of programs that germinated in a corridor conversation, incubated for years within a small clinical team, only to snowball into nationally impactful initiatives under favourable conditions (with nary a SMART goal in sight). If the ‘one right way’ of approaching these QI programs (centred around SMART goals) has a such a mixed record of success, how do we interpret this?

For the first part of my career, I wondered whether the fault was personal. Maybe we, those of us who conceived and led improvement programs, hadn’t been methodical enough, disciplined enough, or even expert enough or perhaps the fault lay with the recipients of said programs. Maybe they weren’t receptive or a tad too self-serving in their interests.

Over time, I (along with so many others) have come to appreciate a core flaw in this premise. The fault is not in the implementation or the people but rather in the tool. Methods that sit within a formal, precise, goal-setting tradition (SMART and a multitude of variants like ABC, PURE, CLEAR and SLIM) falter in programs of complex change because those challenges are adaptive rather than technical.

Photo by Stanislav Filipov on Unsplash

An Unexamined Inheritance

The SMART mnemonic should require no introduction, but for completeness: it stands for Specific, Measurable, Attainable, Relevant and Time-Limited. SMART has a been a key feature of the organisational and project management toolbox for four decades.

Commonly attributed to George T. Doran's 1981 paper "There's a S.M.A.R.T. Way to Write Management's Goals and Objectives" [4], the framework drew in influences from empirical research on goal-setting theory (Locke and Latham) and popular thinking in quality management and management science circles, such as Peter Drucker’s work on management by objectives (MBO).

Whatever path SMART took into healthcare, it arrived to find deep resonance with Evidence-Based Medicine (EBM) and the clinical quality movement, due to a common focus on structured, measurable, and effective practice. This likely smoothed the way to the widespread adoption of SMART goals we see today.

But SMART is just a tool. No more. no less. The deeper more important question relates to the role of formal goal-setting activities in the context of quality improvement and more extensive change work in healthcare. MBO is a useful proxy to think through this because the types of objectives that MBO demands are exactly what SMART typically delivers.

A quick clarification first. While the average reader will not see much difference between the terms ‘goal’ and ‘objective’, Drucker proposed a clear delineation. He reserved the term ‘goals’ for the more far reaching stuff (the ‘what’ and ‘why’) while preferring the term ‘objectives’ for the well-defined, short term steps to get to the final destination (the ‘how’). In this sense, what we refer to as SMART goals would equate to Drucker’s idea of objectives. For the purposes of clarity, the rest of this piece stays with Drucker’s convention.

Drucker’s focus on objectives had a lot to do with context of operational challenges in large US corporations in his day, which were often large industrial manufacturing companies with deep technical engineering foundations. As the struggle with process inefficiency and waste was one of the biggest drivers of quality management activity in the post-war era, and a laser focus on objective measurement was valid centrepiece for this effort.

In stable manufacturing environments with predictable processes, Drucker’s prescription of specific, measurable objectives made a lot of sense, even as contemporaries like W Edwards Deming, while sharing considerable agreement with Drucker on many tenets of organisational management, were already becoming critical of MBO for its propensity to drift towards an emphasise on goals over planning, a neglect of the environment, a tendency towards rigidity, an increasingly short term focus, and an overemphasis on quantitative metrics [5].

Whatever the case may have been in the 70-80s, healthcare in 2025 is nothing like the manufacturing environments of Drucker and Deming’s era. In particular, most of the improvement, innovation and transformation challenges we are left with today are fundamentally complex, as are the social and technological systems that modern care relies on. Complexity overlaid with complexity.

These new complexities can be parsed in a multitude of ways, but to keep things simple, we can just examine the distinction between technical problems and adaptive ones, and cast our minds to the palette of QI problems we routinely engage with.

A bit set of Philips bits for a electronics screwdriver set.

Photo by Mika Baumeister on Unsplash

When Is Precision a Limitation?

Below is a comparison between the key features of technical and adaptive problems.

Dimension	Technical Problems	Adaptive Problems
Definition	Clear and well-defined	Complex, ambiguous, and often ill-defined
Solutions	Known and readily available	New solutions require learning and experimentation
Expertise	Predictably solved by identified experts or authorities	Require participation from and collaboration with those affected
Change	Involve implementing existing procedures	May require changes in values, beliefs, behaviours, and mindsets
Implementation	Often straightforward, use rational problem-solving as needed	Evolves over time, requires on going adjustments, guesses and innovation
Acceptance	People will readily accept technical solutions	People often resist changes technical solutions
Problem Ownership	Experts or authorities can own the problem solving process	The people facing the problem must own the problem solving process

It is not difficult to see why SMART objectives are a natural fit for projects that map to the column in the middle.

Examples of such technical problems in healthcare might include an upgrade to payroll reconciliation system, establishing server side cyber security protocols, or constructing a new wing for a paediatric ward. I say “might” because these too can hide adaptive gremlins with the potential to derail projects if not managed well.

On the other hand, healthcare has a large amount of clearly adaptive problems. In essence, anything that can be described as ‘wicked’ would fit this description - challenges like enhancing safety culture, building organisational resilience, lifting patient satisfaction, resolving capacity pressures or reducing rates of harm sustainably.

If there were proven technical solutions for this second set of problems they would have been solved already.

Yet our tools and templates, our project planning processes, our targets, the demands of our masters and even the language we use, can cajole, bully and hoodwink us into dealing with complexity through an overly simplistic lens.

This “fallacy of misplaced concreteness” as philosopher and mathematician Alfred North Whitehead put it [6], (whether true fallacy or intentional blindness in our case), is one reason why so many fundamentally adaptive challenges in healthcare are erroneously prosecuted with technical toolkits and against structured rationalist and measurement-centric goals.

Photo by Samuel Cruz on Unsplash

Rethinking goal-setting as adaptive activity

So where does this leave us? Are goals completely redundant? Not at all. While the familiar rituals of setting precise objectives within complex healthcare change work (and elsewhere) might be of questionable value (in all but a few focused cases). The setting of thoughtful goals can play a pivotal role in transformation.

Consider this somewhat trivial example. There are two people starting fitness journeys. The first develops the perfect SMART objective: "I will increase my bench press by 10% every two weeks by following a specific progressive loading program." The second has a less precise but possibly more meaningful goal: "I want to feel strong enough to keep up with my kids and set a healthy example for them." Which objective/goal is more likely to motivate either person to get out of bed at 5 AM every morning in the middle of winter?

Of course, in the above example, both the objective and the goal are complementary. Admittedly you could adopt both in the above example. Yet in the case of more complex adaptive challenges, the ‘how’ can be far more difficult to define. In these instances, SMART formulations via carefully defined objectives (in a ‘Druckerian’ sense), can distract from the actual task at hand. For adaptive problems, goals work far more effectively when they serve to galvanise a group with the momentum needed to creatively find novel answers without constraining action in problematic way.

The apocryphal tale of John F Kennedy pressing the NASA janitor on what he does there (“I put people on the Moon, Mr President”) is often shared as an exemplar of a positive organisational culture.

To me, that vignette is much more an example of the galvanising effect of an audacious but persuasive goal within a supportive context. The goal of putting a man on the moon was preposterously aspirational (hence the phrase - “moonshot”) but it tapped deeply into the core purpose of what it might have meant to work at NASA - to innovate and keep the US at the forefront of space technology. The backdrop of the cold war and the Soviet bogeyman did not hurt. This goes beyond the mere technical challenge and into the realm of something more sociological.

I know many QI professionals will be eyeing the exits right about now. But if there are people and groups involved, then we do ourselves a great disservice by not thinking about the sociological characteristics of successful improvement. I encourage you to stick around and consider goals from just two sociological angles:

Photo by Ethan Hu on Unsplash

Randall Collins through his research on ‘interaction rituals’, illuminated (among other things) how goals gain their social power. He proposed that when individuals engage intensely with group objectives—whether in peer interactions or team meetings—they generate emotional energy that transforms abstract ideas, objects and goals into meaningful shared symbols. These moments of heightened collective interaction contribute to crucial emotional and cognitive alignment. These shared experiences powerfully imbue meaning to these objects in a way that goes beyond the material and into symbolic territory - meaningful to the group and thereby enhancing shared identity. In effect, merely defining a rational goal, without attending to the important emotional and ritual experiences that generate group commitment and adherence, is never enough to promote collective action that QI and transformation work often presumes.

More info: Collins, R. (2004). Interaction ritual chains. Princeton university press.

Hartmut Rosa's theory of social resonance adds another dimension. According to Rosa, change emerges from ‘resonant’ interactions between people, through mutual responsiveness, rather than by responding to ‘instrumental rationality’ (the paradigm within which SMART objectives operate). People transform and are transformed through social interactions and it is through this process that they explore, challenge, and refine their understanding of what goals mean for their work and identity.

More info: Rosa, H. (2019). Resonance: A sociology of our relationship to the world.

The above is barely a peek into the relevant material (most of which I am quite new to as well). Karl Weick’s work on sense-making in organisations, Emile Durkheim’s work on ‘collective effervescence’ all add unique layers to the notion that goals can operate as powerful social objects that catalyse action in ways that familiar formal goal-setting methods do not adequately capture. The works of Weber (Entzauburung) and Durkheim (Anomie) are also highly relevant when thinking about the other side of the equation - the sometimes negative social effects of imposed goals and targets.

Taken together, these perspectives suggest that the power of goals in organisational change lies not primarily in their objectivity and measurability, but in their capacity to generate collective energy, create resonant relationships, and facilitate collective action.

SPARK goals?

Ok so what follows is entirely speculative - maybe it only serves to show how easy it is to throw catchy mnemonics like SMART together. Maybe it’s a reminder to remain especially skeptical of ideas that come in too neat of a wrapping. This likely misses more things than it manages to include (even from the few ideas touched on above). Yet, for the sake of memorability, the search of the ever elusive “Take Home Message”, or possibly to offer a little structure to those struggling with goal setting for adaptive challenges, I offer you SPARK.

Component	Description
Sensitive	A good goal must be sensitive to the context in which the work will be undertaken, the resources available, the energy to pursue the goal, known challenges and the potential for unknown challenges along the way. Context sensitivity is by far the most important element of any program and it's no different in goal setting. However, there is no shortcut to developing an appreciation of context. The more time and effort you front-load in developing this appreciation the more attuned you will be to the future possibilities for that system.
Purpose-aligned	A good goal must align to the purpose of the system for which it is developed and as individuals see their own purpose within that system. What is purpose you ask? Answers may be found by reading Stafford Beer’s writings and/or The Batman Chronicles.
Activating	A good goal should draw energy towards it ('integrate conflict' and 'harmonise' as Mary Parker Follett might have said). It should promote a shared enthusiasm so that enough people are activated (to work out how and where they must contribute . Goals must also create a productive level of tension, challenging the status quo while remaining attainable.
Relatable	This is very similar to the ‘R’ in SMART but perhaps comes from a more visceral space “What is in it for me and for what I care about?”
Knowledge-enhancing	When we tackle adaptive challenges, the only guaranteed outcome is the promise of rich learning. If goals constrain our view to a narrow set of possible outcomes then we quite literally reset back to zero if our projects do not experience the prescribed metric of success. By widening the field of view, adaptive goals can unlock critical learnings and experience, putting us in a better position to succeed in time.

Conclusions

While the allure of SMART objectives remains strong, particularly within healthcare's structured patterns of organisational improvement activity, its efficacy in addressing complex, adaptive challenges is demonstrably limited. Often, developing projects on this substrate can put QI work at odds with the nuanced realities of contemporary healthcare service delivery. Over-reliance on precise, measurable objectives can predispose us to neglect the crucial social and emotional dimensions of change. We can miss the very ways in which adaptive goals can catalyse the forward movement we seek.

Until next time - may the ‘stream be with you!

CHANGES TO THE HUMAN STREAM!

The Human Stream has primarily been a email newsletter in 2024. While this has been good, a multi-platform presence (across Linkedin and Beehiiv) has been complicated to support.

While the newsletter will continue be sent via email to those subscribed here, we have transitioned to a Linkedin first model from February 2025. What this means is that subscribing to the Linkedin version of the newsletter will offer a more rewarding community experience and will be the main focus of the newsletter going forward. Do consider doing so yourself.

www.thehumanstream.com will continue to host all back issues of the newsletter.

If there are changes will let you know.

Come along for Healthcare Human Factors 2025 in Melbourne!

The first Monash University Human Factors and Patient Safety Conference will be held March 27 and 28, 2025 at Monash College in Docklands, Melbourne, Australia.

AI Use Declaration: This content is written by a human. A large language model (LLM) was used to support background research, enhancing and proofing of content. All quotes are verified and directly taken from source documents.

The Human Stream is a monthly newsletter for clinician improvers, safety and quality professionals, governance teams and healthcare leaders. The Human Stream compiles insights, topic overviews and practical tips from contemporary safety and systems sciences, all in an easy-to-read, information-rich package, conveniently delivered to your mailbox!