HARD Goals Framework for Leaders: Four Lenses in Practice

The HARD goals framework for leaders uses four lenses: Heartfelt, Animated, Required and Difficult. Mark Murphy presents it in HARD Goals and Leadership IQ materials as a way to create more compelling goals. Its attraction is understandable: a technically precise target can still leave a leader unsure why it matters or what they are willing to change.

Use the framework to improve the goal conversation, then investigate whether the work makes sense. A leader's conviction does not establish another person's agreement, available capacity or a reliable method. This guide develops all four lenses and follows an original fictional goal from an appealing aspiration to a mixed, fully accounted result.

Coachful is our #1 recommendation for organizing this coaching workflow through an original worksheet, an agreed goal and actual private follow-up. This is our editorial recommendation for practical delivery, without a claim that the software or the HARD acronym guarantees leadership performance.

Separate the named model from the evidence for goals

Murphy's own explanation associates Heartfelt with emotional connection, Animated with a clearly pictured outcome, Required with urgency, and Difficult with challenge beyond familiar comfort. Attribute that proposal to him. Do not present the acronym as a validated diagnostic score, reproduce a proprietary assessment, or infer a neuroscience mechanism from a leader's enthusiasm.

Broader goal-setting research asks different questions. Locke and Latham's 2002 review describes benefits of specific, difficult goals alongside conditions involving ability, commitment, feedback and task complexity. For unfamiliar complex tasks, learning goals can be more useful than simply demanding a higher outcome. That review is not an independent comparison establishing that HARD outperforms other acronyms.

Likewise, in a 2009 HBS interview, coauthor Max Bazerman explains concerns about goal pursuit neglecting other valued work and unmeasured consequences. This is a reason to examine the surrounding system, not a reason to conclude that every challenging goal is harmful.

The applications below are original coaching practices. They add feasibility, consent, authority and review questions around the model rather than claiming those checks are additional letters in Murphy's acronym.

Heartfelt: connect importance with the people affected

Emotional connection asks whether this goal matters personally. In practice, help the leader name a concrete reason: a customer should not have to repeat the same information, a new colleague should know where to get help, or an unreliable process should stop consuming everyone's attention. An important reason is more useful than a demand to sound inspired.

Then widen the view. A director might find an ambitious launch exciting while employees worry about repeated overtime. Both experiences belong in the discussion. Ask whose benefit is being described, whose effort it requires and what the leader would protect if the target conflicted with existing responsibilities.

Separate personal meaning from a demand for personal disclosure. A colleague can perform a necessary duty without sharing a moving story about why it matters. A quiet objection about quality or staffing is not evidence of insufficient commitment. The leader should be able to explain the purpose and hear disagreement without rating people's emotional attachment.

An original practical output is a two-part purpose statement: “This matters because…” and “We will not pursue it by…”. For example: “We want new coordinators to find the right escalation route. We will not count a fast answer as useful if it gives them authority they do not have.” The second sentence makes the stated purpose harder to abandon when the metric becomes inconvenient.

If the reason is a routine obligation, say so. Correcting a known error can deserve priority without a personal breakthrough. HARD is optional; a framework should not make reliable maintenance appear unworthy of a leader's attention.

Animated: describe a future interaction, then test its assumptions

The model's pictured outcome can be translated into a concrete future scene. Describe a person, a situation and an action that would be different. “Our handover is excellent” is difficult to investigate. “A coordinator sees an exceptional request, finds the current contact and refers it without guessing approval” gives the team something to examine.

Offer writing, speaking, a sketch or a demonstration. Do not require mental imagery, closed eyes or a particular emotional response. The point of the coaching application is shared understanding, and the client should be able to choose a suitable way of describing it.

Mark the assumptions within the scene. Does the coordinator have access to the card? Is the contact actually staffed? Does the proposed action respect the role's decision rights? A persuasive scenario can hide all three problems if it is treated as an inevitable future rather than a draft.

Ask a colleague to explain what the proposed future permits. If they interpret “resolve the exception” as permission to approve something the leader intended them only to refer, the apparent shared vision is not shared. Record that disagreement before turning the wording into a training instruction.

The practical output is a brief scenario plus two lists: what is already known, and what must be checked. Keep those lists separate. Describing a better experience is not the same event as observing someone use it independently.

Required: give the timing an honest reason

Urgency concerns why action matters now. In a leadership conversation, distinguish the actual source of a date: an external obligation, an operational change, a customer's request or the leader's own planning preference. These sources give different room to change timing or scope.

A rota change may be fixed while the leader's preferred publication date is adjustable. That does not make preparation pointless. It means the leader must state what will operate if the new material is not ready. “We need it by Friday” should lead to a fallback decision, rather than pressure to call an unfinished product complete.

Examine the consequence of waiting without exaggeration. Will a service be unavailable, will an existing process continue, or will a useful experiment simply occur later? Write the answer with its uncertainty. Avoid inventing a business emergency to make an ordinary improvement feel important.

Also ask what work can stop. Urgency does not create extra hours or remove an existing promise to a customer. If two priorities conflict, the person with authority needs to choose. Do not demand that employees solve the conflict through unpaid receiving work or silent reductions in quality.

The practical output is a timing statement: the real event, the chosen preparation date, the consequence of missing it and the person who controls the fallback. A deadline without those distinctions is a calendar entry, not a complete reason to act.

Difficult: identify what must be learned or changed

Challenge can involve unfamiliar skills, coordination, uncertainty or volume. These are different problems. A person who lacks a reliable method needs learning and support. A person who knows the method but has no available time needs a capacity decision. More motivational language does not resolve either condition.

Ask what makes this particular goal hard and what evidence would reduce uncertainty. If a document must work for inexperienced users, the challenge may be discovering wording they interpret correctly. Counting more published pages would not answer that question.

Choose a bounded test with protected duties, an effort allowance and a stopping condition. The test should be able to produce an inconvenient answer. If the team will announce success whatever happens, the activity is a demonstration for approval rather than an investigation.

Difficulty should not be increased merely to make the acronym fit. A necessary safety check, a recovery period or a reliable recurring task can be worth doing at an ordinary level of difficulty. For proposals where no reliable route is known, use the stretch-goal guide to examine the extra uncertainty and downside.

The practical output is a learning question and a boundary: “Can two coordinators interpret the current escalation instruction without hints?” and “If either selects an unauthorized action, the card stays a draft.” This protects the purpose even when the speed result looks attractive.

Work through a complete fictional leadership goal

Fictional case: Rhea leads six coordinators. Davi is her coach, and Sel owns the current exception procedure. Ari and Mina are two coordinators who can participate in a small test. A rota change takes effect on 2 June 2036. Rhea initially wants all six coordinators to use a new contact card independently by Friday 30 May.

On Monday 12 May, Rhea and Davi use a 30-minute coaching session: ten minutes exploring the four lenses, fifteen writing the goal and five defining the first checks. Rhea's completed planning record is below. It is an original discussion record, not Murphy's assessment.

The purpose and scenario give the goal meaning, but neither proves the card will work. The six-person outcome remains distinct from the two-person learning test.

Obtain real participation and check the first version

Rhea spends 30 minutes drafting and sending the card to Sel, and 15 composing invitations that explain the synthetic task, the ten-minute test and its limits. Ari and Mina each spend five minutes reading and replying. Ari accepts Monday 19 May at 10:00. Mina declines that time because of existing shift work and offers Tuesday 20 May at 14:00 instead.

Rhea spends ten minutes reading both answers and confirming those exact appointments. She does not describe Mina's first refusal as a lack of commitment. It gives her a real constraint to plan around.

Sel spends 15 minutes reading the draft and returning a factual correction. Its opening phrase, “Approve the exception,” wrongly assigns authority to the coordinator. Rhea spends ten minutes reading that correction and changing the phrase to “Refer the exception to Sel.” Sel's factual check does not establish that a new user will interpret the entire card correctly.

Rhea facilitates each ten-minute test separately, using 20 minutes in total. Each coordinator's ten minutes includes reading the scenario, answering and explaining the interpretation. No hint is supplied. Ari names Sel and gives the referral reason in 60 seconds. Mina answers in 70 seconds but says a coordinator can approve the request after contacting Sel, misreading a later phrase, “Then sort it out.”

The first result is one of two meeting all criteria. Both answers meet the 90-second condition, but Mina's answer fails the authority condition. Rhea cannot call that two successful performances simply because the clock looks good.

Revise the actual instruction and keep the result limited

Rhea spends 15 minutes recording the results and asking Davi whether her quality rule is too demanding. Davi spends ten minutes in total reading the question, returning the following comment in their individual client conversation and reading Rhea's acknowledgment:

“Your stated purpose is correct referral, not speed alone. Mina answered quickly but described an approval the card does not grant. What exact instruction should replace the ambiguous phrase, and what would a clean retest need to preserve?”

Rhea spends ten minutes reading and replying that she will retain the quality condition and change the wording. She then spends 25 minutes revising the card and sending the revision to Sel. Its completed action instruction reads:

“For an exception outside your current authority, refer the request to Sel and state the decision needed. Do not approve it yourself. Continue only after the authorized decision is returned through the existing procedure. If Sel is unavailable, use the existing procedure's named backup route.”

Sel spends ten minutes reading the revision and approving its use for another synthetic test, without authorizing replacement of the existing procedure. Rhea spends five minutes reading that permission.

Ari and Mina each spend five minutes reading the retest request and accepting separate ten-minute slots on Friday 23 May, at 10:00 and 14:00 respectively. Rhea spends five minutes in total writing and sending the retest request, then reading and confirming both replies, followed by 20 facilitating the two tests. The same criteria and no-hint rule apply.

Ari answers correctly in 55 seconds and Mina in 65. Each names Sel, explains the need for referral and avoids an unauthorized approval. Both therefore meet the declared conditions on this retest. They are the same participants who saw the earlier version, so the result is not an independent test with two new users or evidence that the wording alone caused improvement.

At their second 30-minute coaching appointment on 30 May, Rhea and Davi spend ten minutes checking the results, ten comparing them with the original six-person goal and ten deciding the current status. Two coordinators have demonstrated the task; four have not been tested. The original all-six target has not been met.

Rhea spends ten additional minutes completing the held-release status record. The card remains a draft, the existing procedure stays in force for the rota change, and no further test or rollout is approved within this case. Rhea still cares about the goal, but does not use that commitment to rewrite its denominator.

Check whether the challenge fits the work allowance

This account covers all described coaching, drafting, permission, message, test and closing work by the five named participants. It does not estimate unrelated shift duties. Shared meetings count for each person taking part.

Total described effort is 435 person-minutes against 490 allowed, leaving 55. The unused allowance is not automatic permission for more tests: additional people, actual time and Sel's authority would still need agreement.

Contrast a routine contact-number correction. If Sel already knows the correct number and can update the authorized procedure through its normal process, inventing a difficult leadership challenge may add bureaucracy. Use the least elaborate method that addresses the actual task.

Prepare and review the goal in Coachful

Create an original worksheet under Forms & worksheets for purpose, future interaction, timing, assumptions, boundaries and the next learning question. Let the leader distinguish an aspiration from an authorized action. Do not label it a licensed HARD assessment or calculate an emotional commitment score from the answers.

Agree the practical goal and record the work that supports it. Rhea can record the actual tests and results in the client goal activity, keeping four untested coordinators visible. Completing a task such as “revise the draft” does not establish that the six-person goal has been achieved.

Worksheet Coach review is private to the authorized coach team. Davi's internal assessment is not client feedback delivery. His actual returned comment belongs in Rhea's individual client conversation, reached through her message route, with the intended audience checked. The ledger includes both sending and receiving that exchange.

Keep the coaching record distinct from an agreed operational update. Unshared journal pages are not ordinary coach-readable notes, while explicit client sharing through supported specialty flows can make particular pages available. Review any automatic recap's wording and audience before sending; session-type settings can override the general recap setting, and a later change cannot recall an already sent message.

For practical target wording, compare the SMART goals guide. Use the Coachful features to organize the chosen process, without mistaking a recorded goal, positive reflection or completion indicator for independent evidence of team performance.

Troubleshoot the goal conversation

• The leader feels strongly but others disagree. Ask what they disagree about: purpose, method, cost, authority or timing. A different view can improve the plan; do not test loyalty through visible enthusiasm.
• The future scene keeps changing. Save the current version and the unresolved choice. Agree which interpretation the first test will examine rather than pretending everyone already means the same thing.
• The date cannot move. Review scope and the authorized fallback. A genuine external event still does not prove an untested method is ready.
• The target is missed. Preserve the original result and denominator. A revised next goal can be sensible without retroactively making the old target successful.
• The goal works only when the leader supplies hints. Record that assistance. Decide whether the purpose requires supported use or independent use, then test the relevant condition honestly.

Frequently asked questions

Who created HARD goals?

Mark Murphy presents the framework in his book and Leadership IQ materials. The four lenses are Heartfelt, Animated, Required and Difficult. The case and planning record in this guide are original applications, not his assessment or source examples.

Are HARD goals proven better than SMART goals?

No universal superiority is established here. The originator's explanation and broader research on goal setting are different kinds of evidence. A meaningful ambition still needs clear criteria, resources, feedback and appropriate decisions.

Must a leader visualize the outcome?

No particular mental imagery technique is required in this coaching application. A written scenario, spoken account or demonstration can make the intended change understandable while keeping assumptions open to testing.

Should every leadership goal be difficult?

No. Reliable maintenance, ordinary duties and bounded learning can be valuable. Choose difficulty because it serves a useful purpose, not because a routine goal would otherwise fail to fit an acronym.

Can HARD replace an operational plan?

No. The four lenses can improve the conversation about an ambition. Actual ownership, authority, receiving work, quality conditions, evidence and continuation decisions still need to be established.