“Digital transformation” can sit on a leadership agenda for years without becoming a decision. It is broad enough to include the website, client service, internal systems, data, AI, security, skills and operating processes. Once everything is relevant, every proposed first step appears premature.
The source article compared this to deciding to “get healthy” and then standing in a gym without knowing which machine to use. The intention is sound. The unit of action is missing.
A firm does not need to reduce its ambition. It needs to choose a bounded problem, learn enough to act safely and define the decision that follows. That first move may lead to delivery, deeper discovery or a decision to stop. Progress begins when uncertainty is converted into evidence.
Replace a transformation noun with a problem statement
“Transform our client experience” provides direction and little basis for choosing work. A usable problem statement describes who is affected, what happens, why it matters and what is known.
For example:
Prospective clients for two priority services frequently reach the service pages and leave without starting an enquiry. We do not know whether the cause is relevance, reassurance, usability, acquisition quality or another factor.
This statement resists a premature website rebuild. It suggests research, analytics review and comparison with other routes. It also leaves room for the evidence to show that the website is not the main constraint.
A strong first brief contains:
- a specific group or service;
- observable behaviour or operating friction;
- commercial, client, risk or capacity consequence;
- evidence already available;
- important unknowns;
- an owner able to make the next decision; and
- a boundary that protects the work from absorbing every adjacent issue.
The scope will evolve as the team learns. Expansion should be a recorded choice rather than a reflex whenever somebody mentions a related problem.
Three useful doors
Many mid-market B2B service firms can start their diagnosis through one of three doors. These are prompts, not a complete taxonomy or a forced choice between unrelated concerns.
Door one: client and buyer experience
Start here when client or buyer evidence points to avoidable friction: unclear differentiation, difficult enquiries, inconsistent onboarding, inaccessible interactions, weak digital service or repeated complaints.
Use first-party evidence. Interview clients and prospects, review behavioural analytics, observe priority tasks and speak to the employees who recover failed journeys. A dated-looking website may be a concern; it is not proof of lost revenue by itself.
Our article on why your digital experience may be worse than you think offers a deeper diagnostic for this door.
Door two: platform and operational constraint
Start here when systems create measured delay, manual work, risk or dependence. Examples include routine content changes requiring specialist support, repeated integration incidents, a product approaching end of support or a control that cannot be evidenced reliably.
Map the process and cost before selecting a replacement. The source article’s phrase “a platform problem wearing a content costume” is useful: an apparently slow content team may be constrained by templates, permissions or publishing architecture. The reverse can also be true. A new CMS will not solve slow approvals or unclear ownership.
How to assess whether your digital platform is holding back innovation explains how to distinguish the platform from its surrounding operating model.
Door three: AI and data readiness
Start here when employees are already using AI without a clear framework, when a commercially important use case deserves investigation or when leadership cannot judge which claims are relevant to the firm.
Begin with the work, information and professional consequence. Identify approved and unapproved use, the data involved, required human judgement, security and legal obligations, and the evidence needed to assess value. Some use cases require specialist technical capability. Others are mainly service, process and governance problems.
Our explanation of what using AI in your business actually means separates ordinary use, workflow change and more advanced implementation.
Choose through consequence and evidence
Most firms will identify problems behind all three doors. Pick the first investigation using four questions.
Where is the clearest material consequence?
Look for customer harm, revenue friction, control weakness, excessive operating effort or a strategic constraint. Materiality should be supported by evidence at the firm rather than a general story about the market.
Where can the organisation learn something decision-relevant?
A useful first step resolves an uncertainty that currently blocks action. Research may establish why buyers abandon a journey. A technical spike may show whether a difficult integration is viable. A controlled AI evaluation may reveal whether employees can detect material errors.
Where is action safe and proportionate?
Small scope does not make a project low risk. A narrowly deployed tool can still expose confidential information or affect a vulnerable client. Involve the relevant legal, compliance, security and professional specialists before testing where the context requires it.
Where is there an accountable sponsor?
The sponsor should have authority to secure participation, resolve dependencies and make the next investment decision. Enthusiasm without authority produces a pilot that nobody can scale. Authority without time produces delayed decisions.
Scorecards can support the conversation. They should not conceal judgement. Record why one problem comes first and what condition would cause the firm to change that sequence.
Make the first step small enough to finish and large enough to teach
The source recommended a three-to-four-month initiative. That can be a useful discipline, though duration should follow the question, risk and dependency. A first step might be a two-week evidence review, a six-week service test or a longer pilot where controls and realistic usage require it.
Define completion as evidence and a decision, not simply a delivered artefact. Examples include:
- a tested diagnosis and an agreed priority journey;
- a proof of concept that resolves one technical uncertainty;
- a revised onboarding step with baseline and post-change measures;
- an AI use case evaluated for value, failure modes and governance; or
- a decision against further investment, supported by what the team learned.
Avoid choosing a tiny feature only because it is easy. The work should connect to a consequential problem and produce evidence that leadership can use. Equally, avoid allowing the first initiative to carry the entire transformation promise. A bounded project cannot prove that every later investment will succeed.
Align enough leadership to act responsibly
Professional services firms often need consultation across partners and functions. Universal enthusiasm is rarely a realistic condition for beginning discovery. The firm still needs appropriate authority and participation from people affected by the work.
Three conditions help:
- A named sponsor with a real decision. The sponsor owns the outcome and the route back to leadership.
- A defined boundary and affected group. People understand what will change, what remains outside scope and how concerns can be raised.
- A dated evidence review. Leadership knows when it will see findings and which choice it will be asked to make.
A scope that “doesn’t threaten anyone” is the wrong test. Useful change may affect roles, budgets or established practice. The better test is whether effects are understood, proportionate and governed.
Use disagreement to improve the brief. A finance partner concerned about hiring capacity may be comparing the initiative with a different urgent use of funds. Put the alternatives and consequences into the decision rather than treating scepticism as resistance to digital work.
Create a rhythm after the first decision
A completed first initiative can change the leadership question from whether to invest to what the evidence supports next. It can also reveal that the organisation’s original priority was wrong.
Capture the learning before momentum becomes another slogan. Review outcomes, harms, adoption, assumptions and capability. Decide what to continue, revise, stop or investigate. Allocate capacity and set the next review.
Distinction uses WHNN®, What and How, for the Now and the Next, as a quarterly structure for this conversation. It is our framework and part of our commercial approach, rather than the only valid method. What WHNN looks like in practice explains it in more detail.
The principle is wider than the framework: revisit priorities as evidence and context change, while keeping a record of commitments that should not be reopened casually.
The first step is therefore neither a grand strategy nor action for its own sake. Choose one material problem, make the unknowns explicit and fund the smallest responsible piece of work that can change a real decision. That gives the firm somewhere to stand before it chooses the next machine in the gym.
