Two wells produce the same 180 bbl/day. One has been online 14 months; the other, six years. On a rate sheet they look identical. But the first has 400,000 barrels of upside left and the second has maybe 60,000. If you're underwriting a package, that difference is the whole deal — and it never shows up in a snapshot of current production.
The question that actually matters is: how much of this well's total oil has already come out, and how much is left? That's a cumulative-versus-ultimate problem, and it's one of the most useful things you can pull straight from the production record.
Start With Cume, Not Rate
Every well has a running total — the sum of everything it has produced from first oil to the last reported month. Ask for the cumulative and you get the number that valuation hangs on: this well has made 312,000 bbl of oil and 480,000 mcf of gas across 71 months of production.
Cume is the honest measure of what a well has delivered. Rate tells you where it is on the curve today; cume tells you how much of the tank it has already emptied. The two together are what you need — a high cume with a low rate is a mature well winding down; a low cume with a strong rate is a young well with room to run.
Estimate the Ultimate From the Decline
To turn cume into "percent recovered," you need an estimate of the well's total lifetime production — its EUR. You don't guess at that; you build it from the well's own history.
Fit the decline to the producing months on record — the initial rate, the decline exponent, and how the curve is flattening as it moves from early hyperbolic decline toward a shallower terminal tail. Extrapolate that curve down to an economic cutoff, integrate the area under it, and add back the barrels already produced. That total is the estimated ultimate recovery.
The useful figure falls out immediately:
Percent recovered = cumulative to date ÷ estimated ultimate.
A well that has made 312,000 bbl against an estimated ultimate of 620,000 bbl is about 50% recovered — roughly half its life ahead of it. A well at 88% is nearly spent, and whatever rate it's holding today is the tail end of the story.
Why the Percentage Changes How You Read the Rate
The same 180 bbl/day means completely different things at 20% recovered versus 80% recovered. Early in a well's life, that rate sits high on a curve that still has years of meaningful volume beneath it. Late in life, it's a slow bleed off a nearly flat tail.
This is exactly where ranking wells by current rate misleads buyers. Sort a package by today's production and you'll reward tired wells that happen to be holding a flat tail while penalizing young wells still climbing or in early decline. Sort by percent of ultimate remaining and the picture inverts — you see where the unproduced barrels actually live.
Sanity-Check the EUR Against the Neighborhood
A decline fit on a single well can be noisy, especially if the history is short or interrupted by downtime. Pressure-test the estimated ultimate against context:
- The county or offset average. If comparable wells nearby recover 550,000–650,000 bbl and your fit spits out 1.1 million, the curve is too optimistic — probably reading early flush production as if it were the whole life.
- The operator's own book. An operator's completions in the same area tend to cluster around a characteristic ultimate. A well that lands far outside that band deserves a second look before you trust the number.
- The shape of the tail. A well that has already flattened into a slow, shallow decline is telling you most of the reserves are behind it, regardless of what today's rate looks like.
Benchmarking the ultimate against offsets and the operator's history keeps a single ragged decline fit from carrying more weight than it should.
Watch for the Traps
The percent-recovered calculation is only as good as the history feeding it. A few things to catch:
- Downtime dragging the fit. A well shut in for a workover or choked back will show months of depressed volume. Read those as reservoir decline and you'll understate both the ultimate and the barrels remaining. Separate genuine decline from downtime before you fit.
- Interventions that reset the curve. A refrac or a big workover can lift a well onto a new, higher decline. The pre-intervention decline no longer describes the well — fit the post-uplift segment.
- Short histories. A well with nine months online doesn't have enough curve to trust an ultimate yet. Treat those estimates as wide ranges, not point values.
The Number to Carry Into the Deal
Current rate is a photograph. Percent recovered is the whole reel. Pull the cume from the record, build the decline from the well's own months, extrapolate to an ultimate, and check it against the neighborhood — and you can say, for every well in a package, whether you're buying barrels still in the ground or paying for a tank that's mostly empty.
Ask it in plain language against the Wellsite data lake: what's this well's cume, what's its estimated ultimate, and what fraction is left? The answer is the difference between a well worth chasing and one that's already given up most of what it has.