Reverse-engineer competitor and market API cost from declared trade. The p25 is the cheapest observable buyer; the gap to your internal cost is the procurement mandate.
Tofacitinib API clears at $9,862/kg median with a $5,785 to $14,447 interquartile band across 19 active suppliers and 53 trimmed shipments in 2024. The cheapest buyer is paying $1,707/kg trimmed minimum; the p25 represents a defensible procurement floor sitting at 59% of median. A buyer holding internal contracts at $11,000/kg sits 12% above the median; the procurement mandate is to renegotiate toward the p25, with $5,785 as the structural target.
Type the API INN. The AI agent pulls the global API percentile band, the supplier count, and the trimmed minimum clearance price.
p25 is the procurement target. The trimmed minimum is the absolute floor (one cheap buyer somewhere; not negotiable but informative). The median is the volume-weighted anchor.
Citation-anchored brief: your internal price (if supplied), the p25 floor, the median anchor, and the gap as a dollar figure and percentage. Plus a recommended negotiation posture.
The p25 is what 25% of declared clearance volume cleared at. It is a real floor in the market, not an industry rule of thumb or a vendor brochure target.
Internal price minus p25 = your COGS gap. The brief expresses the gap in both USD per kg and percentage terms; multiplied by your annual volume, it is the negotiation prize.
The supplier count tells the user whether the floor is defended by competition or held by a single dominant supplier. Floors held by 19 suppliers are negotiable. Floors held by 2 are not.
API aggregates refresh weekly. The floor moves with the market; the brief carries the exact computation timestamp.
A real chat thread in PharmaDB. Type a question, the AI agent runs the tools, the answer lands as a saveable note.
pharmagraph_query Twelve API molecules · global 2024 · p25 is the floor 25% of declared volume cleared at.
COGS reverse-engineering is the procurement floor view. PharmaDB surfaces the global API p25 with the supplier count alongside; the user’s internal price minus the p25 is the negotiation prize. The table is molecule_api_prices. Refresh cadence is weekly.
25% of declared clearance volume on the molecule cleared at or below this price. It is the empirical floor that real buyers are negotiating to. Higher than the absolute minimum (held by one cheap buyer); lower than the volume-weighted median (where most volume sits).
Generally no. The trimmed minimum represents one buyer at one moment with terms a typical buyer cannot replicate (forward purchase, multi-year offtake, captive supply). It is informative as a structural floor but not a negotiable target.
Molecules with 15+ active suppliers (tofacitinib, sitagliptin, dolutegravir) have floors held by competition; p25 is negotiable. Molecules with under 6 suppliers (apalutamide, saxagliptin, baricitinib) have floors held by oligopoly; p25 is the price the cheapest two suppliers grant, not a structural negotiation target.
You are negotiating better than 75% of declared buyers. The brief flags this as procurement-leading. The defensive question becomes: is the contract sustainable? If your single-source supplier has thin margin at your price, the supply-chain risk is the bigger concern.
Trimmed: rows above 5x median or below 0.01x median are excluded from the percentile compute. The original rows are retained in the audit log. The trimmed counts surface in the headline so the user sees how much of the raw data made the cut.
API aggregates refresh weekly. The cost floor reruns on every refresh; the catalog row carries the exact timestamp.
Bring the molecule, the lane, or the supplier you're sourcing this week. The AI agent runs it on PharmaDB in 30 minutes. You keep the brief.