An Internet of Accounting

Period Entry Methodology & Blockchain Infrastructure for Real-Time Financial Intelligence

Spencer Nash — December 2025

🧮 Try the Period Entry Demo

Executive Summary

This paper proposes a fundamental redesign of accounting infrastructure that eliminates reconciliation, enables real-time financial visibility, and transforms payments into email-like speed and ease of transacting. By combining blockchain technology with a novel "Period Entry" methodology, it creates a shared economic reality where all parties see the same transactions simultaneously—an "Internet of Accounting" that does for financial flows what the internet did for information flows.

The technology exists. The accounting logic is simple. The economic benefits are overwhelming. What remains is adoption.

The Problem with Current Accounting

Modern accounting suffers from three critical flaws:

First, the point-in-time problem: Balance sheets capture arbitrary snapshots (e.g., December 31st midnight), forcing companies to expend enormous effort "closing the books" rather than tracking continuous economic flows.
Second, the reconciliation problem: Each entity maintains separate ledgers for the same transactions, requiring constant cross-checking, variance chasing, and timing reconciliation across thousands of transactions and hundreds of counterparties.
Third, the forecasting problem: Double entry is backward looking and requires add-ons to make forecasts, and these systems rely on aggregated data.

The Solution: Period Entry + Blockchain

Period Entry Accounting tracks transactions over temporal intervals rather than at arbitrary points in time. Each transaction has an economic lifespan; the system calculates crossover periods to generate profit/loss, balance sheets, and cash reports continuously—daily, weekly, or monthly—without requiring a general ledger or periodic adjustments.

Blockchain infrastructure provides the shared, immutable transaction log where all authorized parties see the same data simultaneously. Combined with Period Entry tracking, this eliminates reconciliation entirely: there's only one ledger, one source of truth, visible to all participants in real-time.

Four parties, one record: Supplier, customer, supplier's bank, customer's bank — all seeing the same entry on a shared ledger. Creating a payment becomes as simple as sending an email, anywhere, any currency.

Chapter 1: The Point-in-Time Problem

Profit & Loss reports are captured at periodic moments—for example, the month of December. This creates artificial temporal boundaries. Companies spend enormous effort "closing the books"—reconciling accounts, chasing variances, ensuring everything balances at that precise moment.

Double Entry calculates invoice values and relative accounting adjustments to those values as points in time. So an invoice for the purchase of stationery on 12th December falls into the fixed accounting period of December 2025. But a one year insurance contract that is invoiced on 20th December must be adjusted via journals, one for each month. So the entries in the system are 13 journals: First to move the invoice value into prepayments and then one each month for the rest of the year.

The heart of accounting—the accruals process: depreciation, prepayments, deferrals and accrued revenue—is an add-on to the invoice processing system. This add-on is the general ledger. The adjustments require detailed reconciliation verified via audit. It is open to error and fraud and creates a mystique of complexity around accountancy, making financial processes an impenetrable wall in organizations.

The solution: Rather than an invoice with a point in time value there is a Period Entry which has the economic lifespan of the transaction. This enables adjustments to be made "on the fly" by juxtaposing the lifespan of the Period Entry and the reporting period.

Example: Insurance Contract

Period: 20th December 2025 to 19th December 2026

Value: £1,200

Reporting Period: 1st June to 30th June 2026

The crossovers are:

Prior period cost (20 Dec 2025 to 31 May 2026): 162 days / 365 days × £1,200 = £532.60 Current period cost (1 June to 30 June 2026): 30 days / 365 days × £1,200 = £98.63

Result: There is no need for a general ledger. There is no reconciliation and no audit of the bookkeeping needed.

Chapter 2: The Informational Silo Problem

Currently each company maintains separate books:

Same transaction, recorded twice, in separate ledgers. Now you must reconcile: Did they receive the invoice? Do their records match yours? Has payment been made? Do payment records match?

And again recorded separately on the customer's bank and again separately on the supplier's bank.

Multiply this by thousands of transactions, hundreds of counterparties, continuous activity. Companies spend enormous resources reconciling ledgers, chasing variances, managing timing differences.

The Internet revolutionized information flow by eliminating isolated silos (post boxes) requiring constant posting and emptying with data flows freely shared on a net. The Internet of Accounting does for financial information what the internet did for websites.
Blockchain (shared, immutable log) + Period Entry (no adjustments needed) = Unified Economic System

❌ Current System

  1. Customer places order (recorded in their system)
  2. Supplier receives order (recorded in their system)
  3. Supplier ships goods (recorded in their system)
  4. Customer receives goods (recorded in their system)
  5. Supplier sends invoice (recorded in their system)
  6. Customer receives invoice (recorded in their system)
  7. Customer pays (recorded in their system)
  8. Supplier receives payment (recorded in their system)
  9. Both parties reconcile (8 separate records)
  10. Payment order sent (and reconciled)
  11. Payment received (and reconciled)

✓ Internet of Accounting

Parties: Customer, Customer Bank, Supplier, Supplier's Bank

Single transaction on shared ledger:

  1. Order placed (visible to all four parties)
  2. Goods shipped (visible to all four parties)
  3. Payment made (visible to all four parties)

Result: No reconciliation needed. One truth. Real-time visibility.

Chapter 3: The Forecasting Problem

Double entry is backwards looking. The forecast is an add-on, normally in a spreadsheet. Huge effort is required to project this forward. This is both aggregate and point-in-time based. Forecasting systems duplicate the data already held in the transaction processing systems.

Period Entry turns this around in two ways:

  1. The economic lifespan is a perfect basis for forecasting forwards. The end date of the transaction is the start date of the next forecasted transaction.
  2. The Period Entry has a driver value and external and relative drivers produce the forecast.
There is no need for an add-on Financial Planning and Analysis tool. Forecasting is built into the transaction itself.

Chapter 4: Payments with the Speed and Ease of Email

Current payment systems are held back by settlement delays, intermediary verification, clearing houses.

Your company pays the supplier on their banking app. The supplier receives the payment on their banking app, banks clear funds. Both parties reconcile records. Days or weeks. Multiple intermediaries. Constant reconciliation. This means that FX gains and losses are incurred.

Internet of Accounting: Instant Settlement

The payment is like email: instant, direct, no intermediaries (for most cases), no reconciliation needed.

Technical Implementation

Benefits Compound

Zero reconciliation Single ledger eliminates need for cross-checking
Instant settlement Email-speed payments
Full transparency Economic flows visible to authorized parties
Real-time error correction Prediction errors detected and addressed immediately
Aggregate intelligence Calculate profit/loss across entire value chains, markets, countries

Chapter 5: Value Chain Visibility

Currently, each company tracks only its own Period Entries. They cannot see upstream or downstream—other customer's or supplier's transactions.

The manufacturer doesn't know the retailer's inventory levels. The retailer doesn't know the manufacturer's production schedules. Each optimizes locally, system suboptimal globally.

Shared ledgers enable value chain visibility: Track Period Entries across entire supply chains. See exactly where value is created or destroyed, bottlenecks occur, fraud happens, and optimization opportunities exist.

❌ Current System

  • Supplier A tracks: production, shipments to Manufacturer B
  • Manufacturer B tracks: inputs from A, production, shipments to Distributor C
  • Distributor C tracks: inputs from B, inventory, shipments to Retailer D
  • Retailer D tracks: inputs from C, sales to customers

Each has a partial view. Cannot optimize the chain as a whole.

✓ Shared Ledger System

  • All four parties see entire chain on shared ledger
  • Supplier A sees Retailer D's sales data → adjusts production
  • Retailer D sees Supplier A's production schedule → adjusts orders
  • Manufacturer B and Distributor C see both → optimize inventory

The system optimizes as a whole.

Result: Dramatically reduced waste (inventory, overproduction, stockouts), faster response to demand changes, better coordination without central planning.

Chapter 6: Business Modelling

The blockchain contains historic transactions up to the present moment. But this data is more than history—it contains the drivers and relationships that define how the business actually works. The historic data isn't just a record. It's a model of the business itself.

The Ledger IS the Model

Traditional approaches treat data and models as separate things: you extract data, then build a model to forecast from it. But in the Internet of Accounting:

The blockchain isn't data FOR a model. The blockchain IS the model.
  • Historic transactions = What happened
  • Relationships between entities = The causal drivers
  • Period Entry lifespans = The temporal mechanics
  • The whole structure = A living model of the business

This is exactly how brains work: memories aren't stored separately from the "model"—the memories ARE the model. The weights ARE the knowledge. The same principle applies here: the ledger IS the business model.

Graph Neural Networks for Privacy-Preserving Forecasting

The blockchain is naturally a graph:

Graph Neural Networks (GNNs) are designed exactly for this structure. They learn patterns from the relationships, not just the data points.

Privacy mechanism:
  1. Raw blockchain data stays private (permissioned access)
  2. GNN trains on the graph structure
  3. Only the model outputs (forecasts) are shared
  4. Individual transactions never exposed—only aggregate patterns

This is strengthened further with federated learning (each party trains locally, shares only model weights), differential privacy (noise added so individual transactions can't be reverse-engineered), and homomorphic encryption (compute on encrypted data without decrypting).

Continuous Learning

The system learns continuously as new transactions arrive:

New transaction arrives
    → Becomes history
    → Updates the graph
    → GNN reweights
    → Forecasts update
    → Expectations adjust
    → Next actual vs expectation = prediction error
    → Learning continues

This is ECF (Emotional Comparator Framework) applied to accounting:

Expected Value (forecast) − Actual Value (transaction) = Prediction Error

The whole system learns from prediction errors. No separate "planning model" maintained in a spreadsheet. The ledger and the forecast are ONE THING.

What the GNN Predicts

Cash flow forecasts For each entity, based on their transaction patterns
Supply chain demand Signals propagating through the graph
Sector-level indicators Economic patterns across industries
Anomaly detection Fraud patterns that deviate from learned norms
Counterparty risk Scores based on transaction graph health

The GNN learns: "When entity A's Period Entries look like X, entity B's typically follow with Y, at lag Z." Forecasting emerges from the graph structure itself.

Chapter 7: Market Intelligence

Accurate and transparent financial data transform the world.

With shared ledgers, calculate aggregate economic prediction errors across markets and countries.

Central banks currently rely on lagging indicators: GDP (reported quarterly, 1-2 months delayed), unemployment (monthly), inflation (monthly). Policy made with stale data.

Real-time Economic Prediction Error Tracking

Monitor aggregate variance between expected and actual:

Economic prediction errors are real-time signals of economic state. Rising negative variances (actuals below expectations) signal contraction. Rising positive variances signal expansion.

Policy Response Becomes Adaptive

Instead of waiting for quarterly GDP reports showing the economy already in recession, detect rising negative prediction errors in real-time and respond immediately.

Privacy Preservation

Blockchain enables selective disclosure. Participants see only authorized transactions. Regulators see aggregates without individual details. Not everything public, but no private inconsistencies—your ledger matches my ledger by construction.

Chapter 8: Complying with Accounting Standards

Why Period Entry Is Naturally Standards-Compliant

The first question from every accountant, auditor, and CFO: "Does this comply with IFRS and GAAP?"

The short answer: Yes. More naturally and far more easily than traditional accounting.

Period Entry doesn't just achieve compliance—it embodies the core principles that accounting standards are trying to enforce. While traditional accounting achieves those principles through elaborate corrections (adjusting entries, accruals, deferrals), Period Entry achieves them by design.

IFRS cares about outcomes, not methods:
  • Is revenue recognized when earned?
  • Are expenses matched to the periods they benefit?
  • Are assets valued appropriately?
  • Are liabilities complete and accurate?
  • Do financial statements fairly represent economic reality?

IFRS doesn't mandate double-entry bookkeeping. It doesn't require general ledgers. It doesn't specify that you must use debits and credits. It specifies economic principles. Any system that achieves those principles complies with IFRS. Period Entry achieves them more directly.

Complex Example Made Simple: Mobile Phone Contract

Over the last twenty years revenue recognition in IFRS and US GAAP has been transformed. Perhaps the most complex example is revenue recognition of mobile service payments.

Scenario

You sign a £300 phone contract. You are contracted to pay £25 per month for 12 months. You receive a handset worth £180 at the start of the contract. But the handset is delivered on day one so the revenue of the handset is incurred on day 1. There is a horrible mismatch.

The Period Entry Solution

A Period Entry which has the 12 monthly payments made at the end of each month totalling £300 and two revenue periods:

  • The first revenue is the handset revenue of £180 on day one.
  • The second revenue is the service fee of £120 (£300-£180) with a 12 month period.

In the first month:

Handset revenue: £180 Service fee: 1/12 × £120 = £10 Total revenue: £180 + £10 = £190 Payments received: £25 Accrued revenue = £190 - £25 = £165 owing

In the last month:

Current income: 1/12 × £120 = £10 Prior period income: 11/12 × £120 + £180 = £290 Total income: £300 Total payments: 12 × £25 = £300 Accrued revenue = £300 - £300 = £0 owing

Once the Period Entry is set up, all the monthly adjustments are calculated automatically. There is no closing period. Reporting is real time, any period, one day, one week, one month. The contract is also perfectly forecast into the future. There is no need for a separate cash forecast.

Conclusion

The Internet of Accounting represents a structural shift in how economic activity is recorded, shared, analysed, and acted upon. By replacing point-in-time double-entry adjustments with Period Entry Accounting, financial statements become real-time outputs of economic lifespans rather than the by-product of a reconciliation process. By placing these Period Entries onto shared blockchain infrastructure, reconciliation itself disappears—replaced by a single, common economic reality visible to all authorised parties.

The implications are profound: payments settle instantly; financial reporting becomes continuous; forecasting emerges naturally from the end dates of transactions; and entire value chains can coordinate through transparent, real-time prediction error signals. What once required general ledgers, adjusting journals, clearing houses, and siloed systems becomes a unified flow of economic information operating with the speed and simplicity of email.

The technology exists. The accounting logic is simple. The economic benefits are overwhelming.

The Internet of Accounting is not merely an improvement—it is the natural next stage in the evolution of economic systems.

Appendix: Period Entry Logic

Calculating current income or expense

Earliest End = the earlier of transaction end date and report end date
Latest Start = the later of transaction start and report start
Crossover days current = earliest end - latest start
Crossover portion current only = crossover days current / length of period in days
Current Income/Expense = crossover portion × transaction value

Calculating the prior period income or expense

Earliest End = the earlier of transaction end date and report end date
Earliest Start = the earlier of transaction start date and report end date
Crossover days All = earliest end - earliest start
Crossover portion current + prior period = crossover days All / length of period in days
Total Income/Expense = (Crossover portion current + prior period) × transaction value
Prior period income/expense = Total Income/Expense - Current Income/Expense

Calculating work done less payments made give accruals

Where payment date <= Report end date, payment was made
Income/Expense - receipt/payment made to date = accrual