A total alone does not tell you whether the increase is expected, unsupported or worth investigating.
See what your occupancy costs are trying to tell you.
A reconciliation should do more than produce a number. Premetry connects source evidence, confirmed inputs, calculations, exceptions and portfolio context so you can understand what changed, what is supported, what needs attention and what to do next.
The problem is rarely that you have no data. It is that the data is disconnected.
Lease terms live in one place. Reconciliation statements live somewhere else. Supporting invoices arrive later. Spreadsheet calculations are difficult to reproduce. The person who remembers why something was done may not be available next year. Financial intelligence begins by reconnecting those fragments.
Financial effect, timing, evidence quality and unresolved exceptions should influence priority—not whoever asks loudest.
If the source, rule and arithmetic are disconnected, the answer becomes dependent on one person’s memory.
Recurring patterns disappear when every reconciliation is treated as an isolated one-time task.
From fragmented evidence to an explainable operating signal.
Premetry separates the work into explicit stages so evidence, authority, arithmetic and judgment do not disappear inside a black box.
Connect the relevant lease, statement, support and source context.
Record the operative numeric input and who confirmed it.
Apply deterministic logic that can be independently reproduced.
Separate factual QA, missing evidence and Legal Review Required states.
Prioritize the finding and connect it to portfolio-level context and next action.
Research-backed financial intelligence for better managerial decisions.
Premetry maintains a governed research inventory and publishes only a selective active cohort. Explore the eight evidence-reviewed resources below, then open the topic most relevant to the decision in front of you. Each resource separates what research supports, what it does not establish and Premetry's practitioner interpretation.
Start with the decision challenge you are trying to solve.
Select a resource below. The reading experience keeps one topic in focus at a time and gives you a clear path forward or back.
FI-001 From Financial Data to Decision IntelligenceWhat separates having financial data from having information that is genuinely useful for a decision?
Why it matters
Organizations can collect more financial and operational data than ever before, yet more data does not automatically create better decisions. Managers still have to determine whether information is reliable, relevant, understandable, timely and connected to the decision in front of them.
What the research shows
Business intelligence and analytics are established fields concerned with using organizational data to support action and decision processes. Information-quality research shows that accuracy is only one dimension of useful information; context, representation, accessibility, completeness and fitness for the task also matter. Empirical analytics research further suggests that value depends on combinations of data, technology, people, organizational capabilities, governance and context rather than on a tool alone.
A useful practical distinction is: data are recorded facts; reporting organizes them; analysis examines relationships or patterns; and decision-useful intelligence connects credible evidence and analysis to a specific question, its limitations and an appropriate next action. That final progression is a Premetry synthesis, not a universal academic hierarchy.
What the evidence does not establish
The literature does not establish “Financial Intelligence” as one universally accepted academic construct. It does not show that more data, dashboards or analytics automatically improve financial performance, and it does not establish that a particular Premetry workflow causes superior outcomes.
Practical implication
Before asking whether an organization needs more data, ask whether the available evidence is good enough for the decision, current enough for the period, traceable to its source and explicit about what remains unknown.
Premetry perspective
Premetry uses “Financial Intelligence” as a practitioner term for evidence that has been organized, checked, interpreted in an explicit decision context and translated into a next step without hiding material uncertainty.
Selected research
FI-002 Evidence Quality and Financial Decision ReliabilityHow does the quality of evidence affect the reliability of a financial decision?
Why it matters
A precise calculation can still support a poor decision when the underlying evidence is incomplete, stale, poorly represented, difficult to trace or irrelevant to the question. Mathematical correctness and evidence readiness are different things.
What the research shows
Information-quality research treats quality as multidimensional rather than synonymous with accuracy. Established work distinguishes intrinsic, contextual, representational and accessibility dimensions; information-systems and business-intelligence research adds the importance of organizational use, maturity and governance.
For managers, a practical evidence review should ask whether the source is credible, the evidence is complete enough for this decision, the period is current, the conclusion is traceable, the information is clearly represented and the evidence is relevant to the task.
What the evidence does not establish
Research does not support a universal “evidence quality score” that predicts financial outcomes across contexts. No single quality dimension guarantees a better decision, and complete information can still require judgment.
Practical implication
Separate calculation confidence from evidence readiness. Preserve source provenance so a later reviewer can understand where the conclusion came from and which assumptions were necessary.
Premetry perspective
Premetry treats evidence readiness as explicit checks rather than a hidden confidence score. Any future scoring model would require separate validation before being represented as predictive or scientifically established.
Selected research
FI-003 Decision-Making Under Uncertainty: Known, Unknown, and PauseWhat should a decision system do when important evidence is uncertain, incomplete or unconfirmed?
Why it matters
Financial systems often create pressure for a definitive answer. That pressure can convert uncertainty into false precision: a clean number or recommendation that looks more certain than the evidence allows.
What the research shows
Research distinguishes uncertainty about a fact or number from uncertainty about the quality of the evidence underlying it. Experimental and professional studies show that uncertainty information can affect decisions and does not necessarily destroy trust, but the effects depend on the task, format, users and environment. Importantly, a production-planning experiment found that prediction intervals did not improve decision quality in that setting, reinforcing the need to avoid simple “more uncertainty information is better” claims.
What the evidence does not establish
Research does not show that every uncertainty label improves financial decisions or trust. “Pause” is not a universal research-established decision rule; it is a Premetry control principle informed by broader evidence.
Practical implication
Separate known facts, unresolved facts, assumptions and confirmation-dependent items. When a missing or unconfirmed input could materially change a calculation or action, obtaining the missing evidence or pausing can be more responsible than manufacturing precision.
Premetry perspective
Premetry's authored operating model makes several states explicit: Known, Unknown, Confirmation Needed, Calculable and Pause. These states are design choices intended to keep incomplete evidence from silently becoming a definitive conclusion.
Selected research
FI-004 Variance Analysis: Separating Signal From NoiseWhen does a financial difference deserve attention instead of becoming another alert?
Why it matters
Variance analysis can reveal changes, errors, emerging risks or opportunities. But a system that flags everything can be as ineffective as a system that flags nothing because managerial attention is limited.
What the research shows
Management-by-exception research provides a basis for directing attention toward selected deviations rather than reviewing every item equally. More recent theoretical work shows that threshold-based communication can reduce information burden under modeled conditions. Dashboard and information-overload research adds that usefulness depends on context, information load, metric selection, risk and the task the manager must perform.
The defensible public principle is narrow: effective exception management requires more than sorting by the largest percentage or dollar change.
What the evidence does not establish
There is no universal research-supported variance threshold, cross-industry formula or Premetry-specific prioritization rule. Older exception-reporting research is not direct validation of a modern production engine.
Practical implication
Use variance as the beginning of an investigation. Consider scale, recurrence, evidence quality, timing, context, consequence and the competing demand for attention.
Premetry perspective
Premetry's public principle is selective, evidence-aware attention. Production thresholds, weights and exception-routing logic remain private and require operating validation.
Selected research
FI-005 Management Attention: What Deserves Investigation?How should managers decide which financial differences or exceptions deserve investigation first?
Why it matters
A technically correct system can still fail if it puts low-value information in front of decision makers while more consequential issues remain buried. Prioritization is an attention-allocation problem as much as a reporting problem.
What the research shows
The attention-based view of the firm argues that organizational structures and processes shape what decision makers notice and act on. Information-overload research similarly shows that excess information can exceed processing capacity. Audit materiality research provides a separate professional example that even formal materiality judgments are not reducible to one percentage because qualitative context can matter.
What the evidence does not establish
The literature does not justify a universal management-attention formula or a Premetry “materiality percentage.” Audit materiality has standards and purposes distinct from managerial prioritization, and Premetry does not adopt audit materiality standards for operating decisions.
Practical implication
Prioritization can consider magnitude, context, recurrence, timing, evidence quality, decision consequence and actionability. The goal is to make the basis for attention explicit instead of ranking by whichever number looks largest.
Premetry perspective
Premetry asks “What deserves investigation?” rather than “What is material?” Any production thresholds or weights are separate from the research concept, remain private and require empirical validation.
Selected research
FI-007 Portfolio Intelligence: Learning Across Locations and Business UnitsWhat can an organization learn by comparing locations or business units instead of reviewing each in isolation?
Why it matters
A single location can show whether something changed. A portfolio view can show whether the same pattern is recurring, isolated, spreading or associated with a shared process.
What the research shows
Organizational-learning research provides a strong foundation for the value—and difficulty—of transferring knowledge across units. Studies of franchises and best-practice transfer show that experience can move across locations under some conditions, but recipient capability, causal ambiguity, source-recipient relationships and what can be codified all affect whether knowledge travels successfully.
What the evidence does not establish
The research does not prove that a financial portfolio dashboard automatically improves profit, detects every error or successfully transfers best practice. It also does not establish that locations are directly comparable without meaningful context.
Practical implication
Use portfolio views to ask better questions: Which exceptions recur? Which locations differ after relevant context is considered? Where are evidence gaps concentrated? Did a corrected practice reduce recurrence elsewhere?
Premetry perspective
Premetry treats portfolio intelligence as a learning layer above individual reconciliations. Claims about financial improvement should follow operating evidence, not be assumed from visibility alone.
Selected research
FI-008 Human Judgment and Automation in Financial DecisionsWhich parts of a financial decision process should be automated, and where should human judgment remain?
Why it matters
“Automate everything” is not a research-supported design principle. The opposite assumption—put a human in every decision—is not supported either. The useful question is how tasks, authority, context, consequence and user capabilities should be divided.
What the research shows
Human-factors research distinguishes functions and levels of automation rather than treating automation as binary. Behavioral research documents both algorithm aversion and algorithm appreciation. Management research frames automation and augmentation as interdependent. Recent field evidence further reinforces context dependence: human-centered interaction design can matter, and a financial-services field experiment found that giving a human banker final say over AI-produced advice did not compromise advice quality in the studied setting.
What the evidence does not establish
Research does not validate Premetry's specific routing of automated versus human tasks or any universal “human-in-the-loop” percentage. Human judgment can also introduce bias and inconsistency, so adding a human is not automatically safer or more accurate.
Practical implication
Break the process into functions: information acquisition, organization, rule application, calculation, classification, recommendation, authorization, execution and exception resolution. Then determine which functions are structured enough to automate and which require explicit judgment or qualified professional review.
Premetry perspective
Premetry's design objective is calibrated task allocation: deterministic work should not consume human time unnecessarily, while ambiguity and legally or professionally sensitive judgments should not be hidden inside automation.
Selected research
- Parasuraman, Sheridan & Wickens (2000), Types and levels of automation
- Dietvorst, Simmons & Massey (2015), Algorithm aversion
- Logg, Minson & Moore (2019), Algorithm appreciation
- Raisch & Krakowski (2021), Automation–augmentation paradox
- Yang, Bauer, Li & Hinz (published online 2025), Human–AI collaboration and investment decisions
FI-009 Explainable Financial Intelligence, Trust, and AdoptionHow should a decision system explain its output so users can rely on it appropriately rather than blindly trust or reject it?
Why it matters
A recommendation that cannot be understood is difficult to challenge, govern or use responsibly. But adding more explanation does not automatically solve the problem; explanations can create cognitive load, encourage over-reliance or raise perceived confidence without improving accuracy.
What the research shows
Trust-in-automation research treats appropriate reliance—not maximum trust—as the goal. Explainability research adds that explanations should reflect how people actually understand explanations rather than merely exposing technical internals. Empirical XAI findings are mixed: AI predictions can improve human decisions while explanations may add no incremental accuracy in some tasks; explanation type, user control, task complexity and domain expertise can change results.
What the evidence does not establish
Explanation does not guarantee accuracy, appropriate trust, compliance, adoption or better decisions. The evidence does not prove that Premetry's specific Evidence Graph or explanation design causes improved outcomes.
Practical implication
A useful explanation should help answer four questions: What evidence supports this output? What logic connects the evidence to the conclusion? What important limitations or unknowns remain? What action follows, and why?
Premetry perspective
Premetry uses an authored candidate explanation pattern: Evidence → Logic → Limitation → Action. It is informed by research on trust, explainability, cognitive load, decision control and expertise, but it is not presented as a validated academic standard. The production Evidence Graph and control logic remain separate proprietary implementation.
Selected research
The same evidence should answer different questions for different people.
Which locations, dollars and patterns deserve attention?
Learn how Premetry translates individual reconciliations into portfolio-level financial visibility.
Can the evidence, calculation and QA decision be reproduced?
Learn how Premetry keeps source evidence, confirmed inputs, calculations and exception states connected.
How do you scale insight across clients without blending their context?
Learn how Premetry supports authorized multi-client prioritization while preserving client-specific evidence.
You do not need to send a lease to learn whether Premetry may fit.
The short assessment helps identify your portfolio scale, current review process and likely next step. No documents, purchase or obligation are required.