Home/Blog/Leadership & Talent
Leadership & Talent · 13 min read

Change readiness and adoption: how to get people to use a change

The short answer

To get people to adopt a change, assess what changes for each group, measure six readiness beliefs and total change load, equip managers and champions, communicate both ways, train on real tasks before go-live, diagnose resistance by root cause, consult where required, decide go/no-go on criteria agreed early, then track reach, depth and proficiency and reinforce.

Want the tools, not just the method? The The Change Readiness and Adoption Kit kit turns this guide into ready-to-use templates, spreadsheets and printables, with an instant download.

Key takeaways
  • The '70% of changes fail' figure has no valid evidence behind it: define your own success measures.
  • Measure six beliefs by group, repeatedly, and act on the lowest.
  • Map every change landing on the same people; move something when a group is saturated.
  • Resistance is information: respond to the root cause.
  • Agree go/no-go criteria before anyone knows the answer.
  • Adoption means reach, depth and proficiency, reinforced after the project ends.

How do you build change readiness, step by step?

Readiness is built, not announced. The steps below follow the order most changes need, and each has a measurable output.

Two roles carry most of the work. People managers are the channel employees trust most, so give them information before their teams, a weekly 15-minute huddle (what changed, what it means for us, questions and workarounds, one thing to try) and honest answers to hard questions. Champions are respected peers who help colleagues and bring back what people really think; recruit them for credibility rather than enthusiasm, include a thoughtful sceptic or two, and give them about two hours a week and early access. In the worked example, 26 champions covered 286 users, about 11 users per champion.

  • Define the change on one page: what is changing, why now, what is not changing, who is affected, how you will know it worked.
  • Assess impact by group across nine dimensions: process, systems, tools, roles, behaviours, skills, reporting lines, location and pay.
  • Map stakeholders and agree dated, visible actions with your sponsor.
  • Measure readiness by group on six beliefs and map every other change landing on the same people.
  • Equip managers with a weekly 15-minute huddle and recruit champions for credibility.
  • Communicate by audience with a message house; run a short pulse every two weeks and reply with 'you said, we did'.
  • Train on real tasks before go-live and check proficiency.
  • Diagnose resistance by root cause; consult where duties apply; decide go/no-go on criteria agreed early.
  • Track reach, depth and proficiency after go-live, then reinforce and review benefits.

Do 70% of change initiatives really fail?

The claim that around 70% of organisational changes fail is everywhere in management writing. Mark Hughes traced five published sources usually cited for it and concluded, in the Journal of Change Management in 2011, that there is no valid and reliable empirical evidence behind it. In a 2022 follow-up he argued that the 'most change fails' story reflects how change has been written about, and called for change to be judged in context and over time.

What the research does support is more useful. A review of 79 quantitative studies over six decades found that reactions to change depend on pre-change factors and on the change process, perceived benefit or harm and content. Much of that is within a change leader's influence. So stop quoting the number and define success for your own change in your own measures.

What beliefs make people ready for a change?

Armenakis and Harris described five beliefs a change message must build: discrepancy (we need to change), appropriateness (this is the right change), efficacy (we can do it), principal support (leaders back it) and personal valence (it is worth it for me). A validated readiness scale by Holt and colleagues measures closely related beliefs. Another research line, from Weiner and colleagues, produced the 12-item ORIC measure of change commitment and change efficacy, which is freely available.

We add a sixth belief, capacity, because a well-argued change still stalls when people have no room to learn it. Research on change fatigue linked strain from many changes to exhaustion, and exhaustion to lower commitment and higher turnover intentions. Measure the six beliefs by group with one statement each, average the scores, and act on the lowest belief for each group.

How do you measure readiness and change load? A worked example

The example below is from the fictional 420-person insurer that runs through our Change Readiness and Adoption Kit. Each belief is scored 1-5; the readiness index is the average divided by 5. The go threshold was set at 70%.

In February the high-impact groups averaged 54% (weighted by headcount); by the May re-score they reached 73%, after training was moved out of peak weeks, a privacy refresher was moved out of the go-live month and leaders put in writing that overriding the AI tool would never count against anyone. A change load heat map, adding every initiative's monthly load (0-3) by group, showed assessors would hit 7 points in June before the move and 6 after.

Belief (claims assessors, 96 people)FebruaryMay
Need for change3.44.0
Right change2.63.7
Able to do it2.43.6
Leaders back it3.03.9
In it for me2.23.5
Capacity2.03.1
Readiness index (average / 5)52%73%

Why does usefulness matter more than ease of use?

In Fred Davis's original technology acceptance studies, how useful people found a system was more strongly linked to their use of it than how easy they found it. The unified model that followed, UTAUT, brought eight acceptance models together around four determinants, including performance expectancy and effort expectancy, and explained about 70% of the variance in intention to use.

For a rollout, that means demonstrations and messages should show the new way working on the person's own tasks, with their own kind of cases. 'It is intuitive' persuades fewer people than 'this removes the side spreadsheet you keep for follow-ups'.

How should you handle resistance to change?

Treat resistance as information. Ford, Ford and D'Amelio argued that change agents contribute to resistance through their own actions and inactions, and that resistance can be a resource. Diagnose the root cause before you respond: awareness gaps need information, skills gaps need practice, capacity problems need time, fear needs facts and commitments you can keep, fairness concerns need a fair process, and sometimes the people raising the concern are right and the plan should change.

Write concerns down in the person's own words, pick the root cause that fits best, ask whether they could be right, and respond to the cause with an owner and a date. Then check back within a week. A log reviewed weekly shows patterns a single conversation hides: in the worked example, two of ten logged concerns were about job security and the AI tool, and the most important one ('If I override the AI, will I be marked down?') was answered by a written commitment from the sponsor, not by more training.

  • Awareness gap ('nobody told us'): targeted information and a huddle.
  • Not convinced ('wrong fix'): evidence and a seat in design.
  • Personal loss ('I lose my say'): name the loss; offer a new role.
  • Skills or confidence ('the screens are slower'): practice time and champions at desks.
  • Capacity or load ('we are flat out'): move other work; protect time.
  • Trust in leaders ('last time they promised'): admit the past; keep small promises.
  • Fairness or process ('why them first?'): publish the rules; allow swaps.
  • Job security fear ('will AI take my job?'): an honest answer and written commitments.
Measure where you are

Reading helps; measuring tells you what to work on. These AI-graded assessments on AssessAll pair with this topic:

Do you have to consult employees about workplace change?

Often, yes, but the duty depends on where people work and what covers them. The position below is correct at November 2026; check the official source and your adviser.

  • Australia: enterprise agreements must include a consultation term on major workplace change and roster or hours changes; award terms require notice, discussion, written information and consideration of views.
  • New Zealand: good faith requires access to relevant information and a chance to comment before a decision likely to affect continued employment.
  • UK: the ICE Regulations apply to undertakings with 50+ employees; TUPE requires informing and consulting before transfers; fire-and-rehire restrictions under the Employment Rights Act 2025 are due from March 2027.
  • EU: Directive 2002/14/EC requires consultation on decisions likely to lead to substantial changes in work organisation; German works councils co-determine monitoring technology; the AI Act requires employers to inform workers before deploying high-risk AI at work, from 2 December 2027 for Annex III systems.
  • Canada: federally regulated unionised employers give 120 days' notice of significant technological change; British Columbia requires 60 days' notice and good-faith work on an adjustment plan.
  • US: unionised employers must bargain in good faith over wages, hours and other terms, and over the effects of decisions they need not bargain; some cities and states regulate AI in employment decisions.

How do you decide whether you are ready to go live?

Agree go/no-go criteria with the sponsor at least six weeks before go-live, mixing people criteria (training completion, task check pass rate, readiness by group) with technical ones (defects, data migration, rollback). Mark which are critical. Agree the rule: any critical miss means no-go; non-critical misses within a limit you set mean go with conditions, each with an owner and date. Then score honestly and let the verdict follow the rule, not the announced date.

In the worked example, eleven of twelve criteria were met a week before go-live: training completion for target users 96%, task check pass rate 79%, readiness in high-impact groups 73%, and champion coverage 11 users per champion. Open high-severity defects stood at 7 against a threshold of 5, a non-critical miss. The verdict, calculated by formula, was GO WITH CONDITIONS: defects down to five or fewer within ten days, with tested workarounds, and extra floor walkers for two weeks. A two-week slip would have pushed go-live into a peak period, which was the mistake an earlier pilot had made.

After go-live, run hypercare: floor walkers where high-impact groups sit, champions available, and a 15-minute daily stand-up covering top issues, workarounds spotted, good news and the day's message to staff. Exit hypercare on criteria such as tickets per day, open defects and active users, not on the calendar.

How do you measure adoption after go-live?

Track three measures weekly against targets you set: active users (reach and speed), utilisation (how much of the work is done the new way end to end) and proficiency (people passing a short task check). Add workarounds found and, for AI tools, the share of suggestions people change, treated as information rather than a target. Keep measuring monthly to month 6, reinforce with recognition by name, remove old routes and use team if-then plans, which had a medium-to-large effect on goal attainment in a meta-analysis of 94 tests.

In the worked example, week 1 came in at 41% active users, 35% utilisation and 30% proficiency, because people were still finishing work in the old system. By week 8 the figures were 90%, 90% and 77%; proficiency reached 80% in week 9; by week 12 they were 95%, 96% and 86%, and workarounds found each week had fallen from 14 to 1. Personal inboxes were switched off in week 10, after utilisation passed 90%, so the old route could not quietly come back. These are a fictional organisation's numbers, there to show how the measures behave, not a forecast.

What are the most common change management mistakes?

  • Quoting the 70% failure figure instead of defining your own success measures.
  • Assessing the system's impact instead of the impact on people.
  • Measuring readiness once at kick-off and never again.
  • Choosing champions by enthusiasm alone.
  • Answering every concern with more communication.
  • Announcing first and consulting second.
  • Letting the go-live date make the decision.
  • Standing the change team down at go-live.
Bodhih Pro Kit

Run the whole change from one kit

The Change Readiness and Adoption Kit gives you the workbooks, forms, message library, casebook and a step-by-step e-book that follows one rollout from pilot to month twelve. Not affiliated with or endorsed by Prosci, Kotter International or any regulator named.

Sources

  1. University of Brighton (Journal of Change Management): Hughes, M. Do 70 per cent of all organizational change initiatives really fail? Journal of Change Management 11(4), 451-464
  2. Fair Work Ombudsman (Australia): Consultation and cooperation in the workplace: best practice guide
  3. Justice Laws Website (Government of Canada): Canada Labour Code, Part I, Technological Change (ss. 51-55)
  4. EUR-Lex: Directive 2002/14/EC establishing a general framework for informing and consulting employees
  5. European Commission AI Act Service Desk: AI Act, Article 26: Obligations of deployers of high-risk AI systems (consolidated text as at 27 July 2026)
  6. Emerald (Journal of Organizational Change Management): Armenakis, A.A. and Harris, S.G. Crafting a change message to create transformational readiness. JOCM 15(2), 169-183
  7. CORE, The Open University (Implementation Science 9, 7): Shea, C.M. et al. Organizational readiness for implementing change: a psychometric assessment of a new measure
  8. University College London (construct database): Davis, F.D. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly 13(3), 319-340
  9. Harvard University DASH repository (Administrative Science Quarterly 44(2)): Edmondson, A. Psychological Safety and Learning Behavior in Work Teams
  10. Shanghai University of Finance and Economics (abstract of Academy of Management Review 33(2)): Ford, J.D. et al. Resistance to change: The rest of the story
  11. Taylor & Francis (Work & Stress 25(4)): Bernerth, J.B. et al. Change fatigue: Development and initial validation of a new measure
  12. University of Konstanz (Advances in Experimental Social Psychology 38, 69-119): Gollwitzer, P.M. and Sheeran, P. Implementation intentions and goal achievement: A meta-analysis of effects and processes
Common questions

Questions people ask next

What is change readiness?

Change readiness is how prepared people are to make a specific change: they see the need, believe it is the right change, feel able to do it, see leaders backing it, see what is in it for them and have the capacity. The kit measures these six beliefs by group and repeats the measure before go-live.

Do 70% of change initiatives really fail?

That figure is widely repeated, but a 2011 study in the Journal of Change Management traced its published sources and found no valid, reliable evidence for it. The kit asks you to define success for your own change in your own measures instead.

How do I measure adoption after go-live?

Track three things weekly: active users (reach and speed), utilisation (how much of the work is done the new way) and proficiency (people passing a task check), plus workarounds. The Adoption tracker sheet does this against targets you set.

Do I have to consult employees before introducing a new system or AI tool?

It depends on where people work and what covers them. Australian awards and enterprise agreements require consultation on major changes and roster changes; Canada's federal Labour Code requires 120 days' notice of some technological changes to a union; German works councils co-determine monitoring technology. The dated crosswalk lists the rules and official sources; your adviser decides what applies.

How long does it take to use the kit?

The quick start takes 15 minutes and the first hour gets your Settings, self-assessment and impact list done. The worked example follows a 34-week plan from kick-off to benefits review, which you can shorten for smaller changes.

Can I do this without a change management consultant?

Yes for the method: the kit is written for a non-specialist lead and includes a filled worked example. Bring in HR or an employment adviser for consultation and bargaining decisions.

Is it based on a proprietary change model?

No. It describes well-known models in its own words and uses generic stages and published research. Not affiliated with or endorsed by Prosci, Kotter International or any regulator named.

Does it cover rolling out AI tools?

Yes. Chapter 10 covers honest answers on job security, treating AI overrides as information, AI literacy training and psychological safety, and the crosswalk includes the EU AI Act's worker information duty for high-risk systems.