What is on-time delivery rate?
On-time delivery rate (OTD) is the percentage of orders or deliveries a supplier completes by the agreed delivery date. It is the single cleanest read on supply reliability, and for most procurement teams it is the headline supplier-performance KPI — the early-warning signal that continuity is at risk before a stock-out actually hits the line.
OTD belongs to the supplier-performance family of procurement metrics and maps directly to the continuity goal among the wider goals of procurement. This reference covers the formula, the calculation choices that quietly change the number, typical benchmarks, what drives OTD up or down, and the levers that actually improve it.
Key Takeaways
- OTD = on-time deliveries ÷ total deliveries × 100. Simple to compute, easy to game.
- Definition choices matter: requested vs. confirmed date, and whether early counts as on time, can swing the number by points.
- Typical benchmark: >95% is world-class, 90–95% acceptable, <90% a red flag.
- OTIF is stricter than OTD — it also requires the order to be complete.
- Improvement is a system: realistic lead times, shared forecasts, scorecards, buffers, and dual sourcing.
The on-time delivery rate formula
The core calculation is straightforward:
For example, if a supplier made 1,000 deliveries in a quarter and 920 arrived on or before the promised date, OTD is 92%. The arithmetic is trivial; the judgement is in defining "on time" and "delivery." Two organisations measuring the same supplier can report materially different numbers depending on three choices.
Choice 1: Which date counts as the target?
Measure against the original requested date and you capture the customer experience but penalise suppliers for delays you agreed to. Measure against the supplier-confirmed date and you reward suppliers for hitting their own (sometimes padded) commitments. Mature teams track both: requested-date OTD for planning reality, confirmed-date OTD for supplier accountability.
Choice 2: Does early count as on time?
In a just-in-time environment, an early delivery creates carrying cost and storage problems, so many manufacturers define an acceptance window (for example, the promised date plus or minus one day). Outside JIT, early is usually treated as on time.
Choice 3: Line, order, or unit?
You can count at the delivery level, the order-line level, or the unit level. Line-level is the most common and the most honest for multi-item orders, because a single late line shouldn't make a 50-line order look fully on time.
On-time delivery rate benchmarks
Benchmarks vary widely by industry, lead time, and how strictly "on time" is defined, so treat the table below as typical directional ranges from public and buyer-reported data rather than fixed targets. The trend in your own data matters more than the absolute number.
| OTD Band | Typical Range | Interpretation |
|---|---|---|
| World-class | 95–100% | Reliable supplier; safe to reduce buffers selectively. |
| Acceptable | 90–95% | Workable, but monitor trend and critical SKUs. |
| Watch list | 85–90% | Reliability slipping; open a performance dialogue. |
| At risk | Below 85% | Material continuity risk; develop or replace the supplier. |
ProcurementAIAgents.com analysis. Set category-specific targets — a long-lead engineered part is not comparable to a catalogue commodity.
OTD vs OTIF: why the stricter metric matters
OTD only asks whether the delivery arrived on time. OTIF — on time in full — asks whether it arrived on time and complete, with the right items in the right quantity. A shipment that is punctual but short still stops the line, so OTIF is the more meaningful service measure. Because OTIF combines two conditions, it is always equal to or lower than OTD for the same supplier; a big gap between the two points to fulfilment-completeness problems rather than scheduling problems.
Many teams report OTD as the public scorecard number and use OTIF internally to diagnose root cause. If your OTD is healthy but operations still report shortages, OTIF is where the truth hides.
Catch slippage before it becomes a stock-out
Continuous supplier monitoring surfaces OTD decline earlier than quarterly reviews. See which platforms automate it.
What drives on-time delivery rate
OTD is an outcome metric, so improving it means addressing the upstream causes. The most common drivers of late delivery:
- Unrealistic lead times — promising dates the supplier's process can't hit. Often the biggest single cause and the easiest to fix.
- Forecast inaccuracy — suppliers can't position capacity and inventory for demand they didn't see coming.
- Upstream supply disruption — the supplier's own suppliers are late, propagating delay down the chain.
- Capacity constraints — the supplier is over-committed across its customer base.
- Order errors and changes — late specification changes or PO errors reset the clock.
- Logistics and customs — transit, freight, and border delays outside the supplier's plant.
Notice how many of these are partly within the buyer's control. Lead time and forecast quality in particular are shared responsibilities, which is why OTD improvement is rarely just a matter of pressuring the supplier. The interplay with inventory buffers is direct: when OTD is volatile, teams compensate with safety stock, and the right buffer depends on both OTD variability and supplier lead time.
How to improve on-time delivery rate
Sustained OTD improvement is a system, not a single intervention. The proven levers, roughly in order of impact:
1. Set realistic, capacity-aware lead times
Agree lead times the supplier can actually meet rather than the ones you wish for. A truthful longer lead time beats an aspirational short one that's missed half the time.
2. Share forecasts and demand signals
Give suppliers forward visibility so they can position capacity and materials. Collaborative forecasting is one of the highest-return OTD levers and costs little.
3. Run supplier scorecards with consequences
Measure OTD per supplier, review it on a cadence, and tie it to award decisions. What gets measured and acted on improves; scorecards with no consequence do not.
4. Build buffers for critical, volatile items
Where OTD is structurally unreliable and the item is critical, hold safety stock or qualify a second source. This is risk management, not waste.
5. Diversify single-source exposure
A sole supplier with mediocre OTD is a continuity risk with no fallback. Dual sourcing critical categories caps the downside, a theme we develop in our reference on supplier risk assessment.
6. Develop or replace chronic underperformers
For suppliers stuck below target, run a structured improvement plan with milestones. If it doesn't move, replacement is the disciplined choice.
The sequence matters as much as the individual levers. Start with the cheap, high-leverage fixes — realistic lead times and forecast sharing cost almost nothing and often recover several points of OTD on their own. Only once those are exhausted should you reach for the expensive levers of buffer stock and dual sourcing, which carry real working-capital and management cost. Teams that invert this order — buying inventory to paper over a lead-time promise nobody could hit — spend money to hide a problem they could have fixed for free. Treat OTD improvement as a staircase: collaboration first, redundancy second, replacement last.
"Most chronic OTD problems trace back to a lead time nobody could ever hit. Fix the promise before you blame the supplier, and the number often moves on its own."
Putting a number on OTD improvement
OTD has a direct financial signature: poor delivery reliability forces higher safety stock, expedited freight, production downtime, and lost sales. Quantifying those costs turns an operational metric into a business case for supplier development or a tooling investment. Our ROI calculator helps frame the savings from reducing buffer stock and expediting, and the procurement AI ROI business-case model shows how reliability gains feed a wider automation case. For the assumptions behind a credible estimate, the procurement AI ROI calculator guide is the companion walkthrough.
A worked OTD calculation
Walking through a real calculation exposes the judgement calls that move the number. Suppose a supplier made the following deliveries against purchase orders in a quarter:
| Measure | Count | Notes |
|---|---|---|
| Total deliveries | 1,000 | All POs delivered in the period |
| On or before requested date | 880 | Customer-experience view |
| On or before confirmed date | 940 | Supplier-accountability view |
| Early (more than 2 days) | 40 | Counts against OTD in a JIT window |
Measured against the requested date, OTD is 880 ÷ 1,000 = 88%. Measured against the supplier's confirmed date, it is 940 ÷ 1,000 = 94%. If you operate a just-in-time acceptance window and treat the 40 early deliveries as not on time, the figure drops further. Same supplier, same quarter, three defensible numbers ranging from the low 80s to the mid 90s — which is exactly why the definition choices in the previous section matter more than the arithmetic.
The lesson for any scorecard is to fix the definition once, document it, and apply it consistently across suppliers. A supplier that looks better only because you measure it against its own padded confirmed dates is not actually more reliable. Pair OTD with the broader supplier-performance measures in our reference on procurement metrics so a single flattering definition cannot hide a continuity problem.
OTD expectations vary by industry
There is no universal "good" OTD because the cost of a miss differs enormously by sector. In high-volume manufacturing running lean, a one-day slip can halt a production line, so OTD targets are aggressive and acceptance windows are tight. In project-based construction or capital equipment, lead times are long and a few days' variance on a multi-month delivery is tolerable, so targets are looser but the consequences of a major miss are larger.
Distribution and retail sit in between, with OTD tightly linked to shelf availability and promotional commitments. Services categories often track milestone delivery rather than physical OTD altogether. The practical implication is that OTD targets should be set per category against the operational cost of lateness, not copied from a generic benchmark. A blanket 95% target applied across a diverse spend base will be too soft for the critical lines and pointlessly punishing for the ones where a day either way is irrelevant.
This category-by-category sensitivity is also why OTD feeds directly into inventory policy. Where lateness is expensive and OTD is volatile, the rational response is to carry more safety stock or qualify a second source — a trade-off that connects this single metric to the wider continuity goal among the goals of procurement.
OTD and its sibling delivery metrics
OTD rarely travels alone. It belongs to a small family of delivery-performance measures, and understanding how they relate prevents the common error of treating OTD as the whole picture when it captures only one dimension.
Fill rate measures completeness — the percentage of ordered units actually delivered, regardless of timing. A supplier can hit a high OTD while running a poor fill rate if it ships punctually but short. Reading the two together tells you whether a problem is about scheduling or about supply availability.
OTIF (on time in full), covered earlier, combines OTD and fill rate into a single stricter measure. Perfect order rate goes further still, requiring a delivery to be on time, complete, damage-free, and correctly documented — the most demanding service measure of all, and the one that best reflects the end-customer experience. Because each added condition can only lower the score, perfect order rate sits at or below OTIF, which sits at or below OTD.
Lead-time reliability rounds out the family by measuring the consistency of delivery timing rather than adherence to a single date. A supplier with predictable, if longer, lead times is often easier to plan around than one whose OTD is high on average but wildly variable. The practical guidance is to lead your scorecard with OTD for its simplicity, but pair it with fill rate or OTIF so a punctual-but-incomplete supplier cannot hide. These measures sit alongside the broader supplier KPIs in our reference on procurement metrics, and their volatility is what ultimately sizes the safety stock you need to hold.
Frequently asked questions
What is on-time delivery rate?
On-time delivery rate (OTD) is the percentage of orders or deliveries a supplier completes by the agreed delivery date. It is calculated as on-time deliveries divided by total deliveries, multiplied by 100, and is the primary procurement metric for supply reliability and continuity.
How do you calculate on-time delivery rate?
Divide the deliveries that arrived on or before the promised date by total deliveries in the period, then multiply by 100. For example, 920 on-time out of 1,000 gives 92%. Decide upfront whether you measure against the requested or confirmed date, and whether early deliveries count as on time.
What is a good on-time delivery rate?
As a typical benchmark, world-class suppliers run above 95%, an acceptable range is roughly 90 to 95%, and below 90% usually signals a reliability problem worth investigating. The right target varies by industry, lead time, and how strictly you define on time.
What is the difference between OTD and OTIF?
OTD measures only whether a delivery arrived on time. OTIF — on time in full — also requires the delivery to be complete, with the right quantity and items. OTIF is the tougher, more meaningful measure because a punctual but incomplete shipment still disrupts operations.
How can you improve on-time delivery rate?
Set realistic lead times, share forecasts, monitor with scorecards, build buffers for critical items, diversify single-source categories, and address chronic underperformers through development or replacement. Continuous monitoring tools surface slippage earlier than periodic reviews.
Quantify the cost of unreliable delivery
Model the buffer stock and expediting you could remove by lifting OTD, with our independent ROI calculator.