With pay transparency becoming an increasingly important issue, what are the biggest challenges employers face when trying to make fair and consistent pay decisions?
Employers primarily struggle with data siloed across disparate HRIS, benefits, and payroll systems, compounded by decentralised job architectures. This fragmentation obscures real-time visibility into internal peer positioning, making it nearly impossible to evaluate "work of equal value."
Another major hurdle is the absence of defensible documentation and audit trails. During merit cycles, pay adjustments are too often driven by unchecked manager discretion rather than objective performance metrics—or the rationale goes unrecorded, inadvertently widening pay gaps.
Practical Example: At beqom, we frequently work with global employers who discover that identical engineering roles across different business units carry unexplainable 15–20% salary discrepancies. Reconciling these requires establishing unified, objective pay bands alongside embedded workflow guardrails that prevent out-of-policy pay decisions before compensation statements are finalised.
Many organisations still rely heavily on spreadsheets to manage compensation. What risks does this create, and why are businesses finding it difficult to move away from these processes?
Spreadsheets introduce severe operational and compliance risks: the absence of a single source of truth, version control chaos, formula errors, leaks of sensitive pay, and non-existent audit trails. A single copy-paste mistake can trigger millions in payment errors while leaving the organization unable to document or defend pay decisions under regulatory scrutiny.
Organisations resist moving away because spreadsheets offer familiar, immediate flexibility without IT friction, whereas standard HRIS platforms are often too rigid to support complex compensation logic like deferrals, multi-tier bonuses, or global sales plans.
Practical Example: Frequently we see enterprises only transition to a dedicated comp engine like beqom after experiencing a high-profile spreadsheet error, such as bonus allocations exceeding approved budgets or unmasked salary sheets being emailed to all managers.
How can AI help HR teams identify potential pay inequalities and make better-informed compensation decisions? Where should human oversight remain essential?
At beqom we speak about Intentional AI, which means it’s explainable, collaborative and controllable, not a black-box and always with humans in the loop or even in the lead. Human oversight is essential for final decision-making, providing nuanced context, empathy, and final governance. Algorithms cannot evaluate exceptional individual contribution, non-traditional career paths, or nuanced organisational changes.
Our pay equity analyses, based on multivariate regression models, continuously surface group pay gaps and high risk employees. AI excels at helping ingest multi-source data, running complex scenario models and explaining the outputs in ways that HR can use to make decisions. It models remediation budgets in real time and provides managers with data-driven salary recommendations during allocation rounds.
Practical Example: AI can flag that a female director is paid 8% below the peer median, explain how pay factors contribute to some of those differences, and suggest a specific pay adjustment. It can also translate complex statistical model outputs into non-expert language. However, HR leaders and business managers must validate whether that difference is justified by objective, documented criteria or requires an immediate equity remediation.
With the EU Pay Transparency Directive approaching, what practical steps should employers be taking now to prepare, particularly those operating across multiple countries?
The first step employers should take immediately as a foundation is to audit their job architecture to define "work of equal value" across entities and harmonise cross-border title taxonomies. Next, preparing and sourcing their pay data for the various countries is essential to conducting initial pay gap audits. Identifying pay source systems, required fields, and ideally approaches to centralise and map the data into one source of truth.
Once those foundations are in place, key operational steps include:
- Running simulated gender pay gap audits to identify categories exceeding the 5% unjustified gap threshold, triggering mandatory joint pay assessments.
- Planning remediation strategies and where necessary allocating pay equity budget to upcoming compensation rounds.
- Updating recruitment processes to include transparent salary ranges in job postings and banning questions about candidate salary history.
Practical Example: Multi-entity clients are currently implementing beqom's pay equity analysis to run pre-directive pay gap simulations. It typically takes a few months of iteration to arrive at the models that will work across the organisation. And finally, they are ensuring pay remediation occurs proactively either in the current year-end cycle or mid-cycle next year, before statutory reporting becomes public.
Employees are increasingly expecting greater transparency around pay and progression. How can organisations build trust while managing sensitive compensation information?
Building trust does not mean making everyone's individual salary public. In fact the EU PTD specifically references safe guards and balance between pay transparency and data privacy. Employers are required to proactively share job levels, objective salary ranges, and the explicit criteria required to progress between bands with employees and job prospects.
In addition, at an employee's request, employers are obliged to share the employee's pay compared to employees performing work of equal value. Specific safeguards need to be built into the process to ensure the statements provided are accurate, complete and to ensure individual salaries are not disclosed.
Practical Example: Companies establish trust by responding to employee requests in a timely manner and by equipping frontline managers with compensation talking points. Managers can clearly explain: "You are at the 30th percentile of Band 4 or within your Category of Worker group, because of your tenure or performance rating; here is the criteria to reach the 50th."
Looking ahead, how do you see AI changing compensation management over the next few years, and what should HR leaders be prioritising today?
Looking ahead, compensation management will center on unified data lakes that consolidate disparate HRIS, payroll, and performance records into a single source of truth. Operating directly on this governed foundation, specialised AI agents deliver real-time pay intelligence, by continuously modeling market shifts, detecting wage compression, and surfacing equity gaps before they become systemic liabilities.
This equips HR and frontline managers to make informed, defensible pay decisions during hiring, promotions, and merit reviews, replacing subjective guesswork with explainable, data-backed guidance.
Today, HR leaders must prioritise unifying enterprise reward data and standardising job architectures; without this single source of truth, AI agents cannot generate reliable, compliant pay intelligence.

















