How to Verify a Well-Test Allocation Factor
An allocation factor is the number that decides how much of a battery's measured sales volume gets credited to each well, and it is only as good as the test behind it and the arithmetic that combines the tests. This page is for the engineer or allocation analyst checking that a well's factor is defensible before it drives a month of production accounting. It walks the verification in order - trace the factor to its test, confirm the battery factors reconcile, and check theoretical against measured - so a bad factor is caught before it misallocates volume.
Verify a well-test allocation factor in one line: To verify a well-test allocation factor, trace it back to the specific well test that produced it and confirm that test was stable, clean, and correctly closed out, then confirm the factors across the battery combine correctly so the allocated volumes sum to the measured sales volume. Finally, reconcile the theoretical allocated total against the actual metered sales and confirm the difference is small and randomly distributed, not a persistent bias pointing at one well's factor.
Trace the Factor Back to Its Test
An allocation factor is derived from a well test, so the first verification is to open the test behind it. Confirm the test was recent enough to still represent the well, that the separator was stable and level control was clean during it, and that the closeout math was correct. A factor built on a test with carryover, an unstable start, or a closeout error is wrong at the root, and no downstream reconciliation will reveal which well is off unless you check the tests. The concept the factor implements is covered in the guide on what a well-test allocation factor is.
Confirm the factor uses the right test for the right well and phase. Allocation is done per phase - oil, water, gas - so each phase has its own factor from the test's corresponding rate, and a factor that pulled the wrong phase's rate or an old superseded test is a mapping error. Verifying that the factor traces to the correct current test for the correct phase is the step most likely to catch a silent misconfiguration, and it ties to the closeout discipline in the guide on closing out a well test to allocation.
Confirm the Battery Factors Reconcile
Allocation splits a shared measured volume among wells, so the factors across the battery have to combine correctly - the allocated volumes must sum to the measured sales volume, no more and no less. If the sum of allocated oil exceeds or falls short of the metered sales oil, a factor is wrong, a well is missing from the model, or a phantom well is included. This sum check is the fastest way to catch a gross error before chasing individual wells.
A worked check makes it concrete. If three wells tested at 300, 200, and 100 barrels of oil per day, their theoretical total is 600 and their factors are 0.50, 0.333, and 0.167, which sum to 1.0. If the battery actually sold 570 barrels that day, each well is scaled by 570 over 600, giving 285, 190, and 95, which sum back to 570. If your allocated volumes do not sum to the measured sales, the factors do not reconcile and the model is broken before any individual factor is even in question. This is the theoretical-versus-actual mechanic described in the guide on theoretical vs actual allocation.
Reconcile Theoretical Against Measured
The deeper verification is watching how the allocation performs over time. Each period, the theoretical total from the tests is scaled to the measured sales, and the scaling factor - how far the tests' sum sits from reality - should be small and should wander randomly around one. A scaling factor that is persistently far from one, or that trends, says the tests as a group are stale or biased and the whole battery's factors need refreshing.
More useful is spotting a single well whose factor is off. If one well's allocated volume consistently disagrees with an independent check - a well's own dedicated meter, a downhole indication, or the next test - that well's factor is the suspect, not the whole model. This is the difference between a model that is uniformly a little stale and one distorted by a single bad test, and only the reconciliation over several periods distinguishes them, connecting to the back-allocation logic in the guide on back allocation of well production.
Use Continuous Data to Keep Factors Honest
Allocation factors decay as wells change, and the only defense is knowing when a well has moved off its last test. A platform such as Merobix trending each well's available indications - injection rates on lifted wells, casing pressures, run times - flags a well whose behavior has diverged from the conditions of its last test, which is exactly the well whose factor is now suspect and whose test is overdue.
That continuous view turns allocation verification from a monthly paper reconciliation into an ongoing watch. Rather than discovering at month-end that a well drifted three weeks ago, you see the divergence when it happens, prioritize that well's next test, and refresh its factor before it distorts a full period's accounting. The test proves the factor; the continuous data tells you when the factor has gone stale, which is the harder half of keeping allocation honest.
Frequently Asked Questions
How do I check that allocation factors are consistent?
Confirm the allocated volumes across the battery sum back to the measured sales volume for each phase. If the sum of allocated oil does not equal the metered sales oil, a factor is wrong, a well is missing, or a phantom well is in the model. This sum check catches gross errors immediately, before you start examining any individual well's factor, and it is the fastest first verification of an allocation model.
What does it mean if the theoretical total keeps missing the measured sales?
The scaling factor between the tests' theoretical total and the measured sales should be small and wander randomly around one. If it is persistently far from one or trends over time, the tests as a group are stale or biased and the whole battery's factors need refreshing. If instead a single well consistently disagrees with an independent check, that one well's factor is the suspect and its test is overdue, not the entire model.
How often should an allocation factor be refreshed?
Whenever the well behind it has changed enough that the last test no longer represents it, which is a condition rather than a fixed calendar. Watching a well's available indications - lift injection rate, casing pressure, run time - tells you when its behavior has diverged from its last test, and that divergence is the signal to retest and refresh the factor before it distorts a full period of allocation.
Automation services
Need help turning this into a working system?
Merobix integrates SCADA, programs Allen-Bradley and Siemens PLCs, and designs and fabricates industrial control panels.
Meeting requests are reviewed before confirmation.