A DEI strategy is a written plan that names two or three specific problems found in the organization's own data, sets out lawful changes to fix them, and gives each change an owner, a budget and a measure reviewed on a fixed schedule. It is built in five steps, from gathering data to choosing priorities, designing actions, assigning measures and reviewing progress.

The order matters more than the format. A strategy that starts with actions and looks for data afterward tends to fund whatever was easiest to announce. One that starts with data ends up funding the problem the organization actually has, and leadership can check whether it worked.

What a DEI strategy is, and what it is not

A strategy is narrower than most first drafts. It is not a values statement, a calendar of observances or a list of every good idea. Those may sit alongside it, but the strategy itself is the short list of problems the organization has decided to fix this year and next.

That narrowness is what makes it fundable. A finance director can approve three costed actions with named owners. Twelve aspirations with no owners tend to survive the meeting and then quietly disappear by the next budget cycle.

It also connects to the business case. The business case for DEI argues that one problem is costing money. The strategy takes that problem, and one or two others, and turns them into a plan with dates.

The five steps from data to a working plan

Each step hands something to the next, so skipping one weakens everything after it. The sequence below is the one most practitioners follow, whatever the size of the organization.

Sequence / 5 stepsA DEI strategy moves from evidence to a reviewed planEach step depends on what the previous one produced
  1. 1

    Gather the data

    Pull the workforce, hiring, pay and survey records the organization already keeps.

    Hands on a set of patterns worth explaining
  2. 2

    Choose the priorities

    Pick the two or three patterns with the clearest cost and the clearest cause.

    Hands on a short list of problems
  3. 3

    Design the actions

    Change the process behind each problem, in a way that applies to every employee.

    Hands on actions that can be costed
  4. 4

    Assign measures, owners and budget

    Give each action one number to watch, one person responsible and a priced budget line.

    Hands on a plan leadership can approve
  5. 5

    Review on a calendar

    Report progress quarterly and revise the plan once a year against the data.

The steps look simple on a page. The work is in doing each one properly, so the sections below take them in turn.

Step one is gathering the data the organization already holds

Most organizations hold more useful data than they realize. Private employers with 100 or more employees already file an annual EEO-1 report with the EEOC, which breaks the workforce down by job category, sex and race or ethnicity. That report is often the quickest starting point.

Beyond it, the useful sources are turnover and exit records, the hiring funnel from applicant to offer, pay and promotion history, and engagement survey results by team. Before cutting any of it by demographic group, agree a minimum group size with HR and counsel, so no individual can be identified.

Pay data deserves particular care, because it is often the least visible. Across the practitioners we have trained at the DEI Excellence Institute over the past two years, fewer than a third said their organization published pay ranges when they began work on its strategy. That makes a pay review a common first priority.

Step two is choosing two or three priorities

The data will show more patterns than any plan can address. The test for a priority is whether it has a visible cost, a likely cause the organization controls, and someone senior who cares about fixing it. A pattern with all three is worth a year of effort.

Patterns without a clear cause go on a watch list instead. A gap in promotion rates with no obvious mechanism is a reason for listening sessions or exit interviews this year, not for a funded program built on a guess.

Step three is designing actions that apply to everyone

The strongest actions change a process rather than a person's opinion. In Why Diversity Programs Fail, Frank Dobbin and Alexandra Kalev found that mandatory training was often followed by no gains in management representation, while approaches that engaged managers in solving the problem and made decisions more accountable did better.

In practice, that points to actions such as posting every opening internally, using structured interviews with written scoring criteria, auditing pay within each job, and opening mentoring programs to all employees. Each is neutral on its face and in its use, and each can be measured.

Step four is assigning measures, owners and a budget

Every action needs one measure, one owner and one budget line. The table below shows the shape for an illustrative plan with three priorities, which is the format the DEI strategic plan template uses.

Table / 3 prioritiesEach priority gets one action and one measureAn illustrative plan for a mid-sized employer
PriorityActionMeasure
Night shift turnoverSupervisor training and monthly one-to-onesVoluntary exits per quarter
Few internal applicantsPost every opening for ten working daysInternal applicants per opening
Unexplained pay gapsAnnual pay audit within each jobGaps unexplained by stated factors

Each measure is a count the organization can produce every quarter without a new system. The DEI metrics dashboard template lays these out over time, and the DEI program budget calculator helps price the actions before the plan goes to leadership.

Step five is reviewing the plan on a calendar

A strategy without review dates becomes a document nobody opens. The usual rhythm is a short quarterly report to the executive sponsor on each measure, and a fuller annual review that rechecks the data and decides whether each priority stays, changes or closes.

Closing a priority is a good outcome, not a failure. If internal applications have doubled and held for a year, the action becomes normal practice and the slot goes to the next problem on the watch list.

Keeping the strategy within the law

Every action in a strategy has to comply with anti-discrimination law, which protects all employees and applicants equally. In July 2025 the Department of Justice issued guidance for recipients of federal funding stating that using race, sex or other protected characteristics as criteria for employment or program participation is generally unlawful, whatever the program is called.

The same guidance warns about facially neutral criteria used as proxies for protected characteristics. A practitioner therefore checks both what an action says and how it will be applied, and has employment counsel review the plan before launch, especially in a federal contractor or grantee.

Mistakes that stall a strategy

Most stalled strategies fail in predictable ways, and each one is avoidable at the drafting stage.

  • Too many priorities. Eight goals spread the budget and the owners thin, so none of them moves.
  • Measures nobody can produce. A measure that needs a new survey tool will not be reported on time.
  • No named owner. An action owned by "HR" or "leadership" belongs to no one in particular.
  • Training as the whole plan. Training can support a process change, but on its own the research suggests it rarely shifts outcomes.

Where the certification goes further

The Professional DEI Certification Course teaches this process in full. Unit 3, DEI Strategic Planning and Business Integration, covers three-year and five-year plans, budgets built line by line with their owner, and the KPIs and scorecards that track progress.

Unit 1, Foundations of DEI Leadership, covers the business case and the stakeholders a plan depends on. Unit 4, Managing Diversity Initiatives, covers running the programs themselves and analyzing whether they changed anything, which is what the annual review asks.