Blog · Coverage and territory · Trading
How a sales-trading desk finds the clients whose inquiry or trade count has fallen against their own baseline from the RFQ and trade logs, per product, why a client can be silent on one desk and active on another, the quiet score that ranks them by the flow at stake, and the weekly list that reaches the salesperson before the quarter-end review does.
A salesperson learns a client has moved its rates flow to another dealer at the quarterly review, from the client. The RFQ log knew in week three. This guide sets out the quiet client list built from the client's own baseline per product, the desk-wide adjustment, and the score that ranks it.
Per client, per product:
Baseline = median weekly inquiry count and trade count, trailing quarter Quiet ratio = this month's weekly average ÷ baseline Quiet if ratio < threshold, and the product's desk-wide ratio is above it Quiet score = prior monthly notional × desk hit ratio with this client × (1 − quiet ratio)
Client identifiers only.
| Signal | Reading | Owner |
|---|---|---|
| Inquiries down, trades down in proportion | The client has moved or gone quiet | Salesperson |
| Inquiries steady, trades down | Hit ratio fell: pricing, speed or axe | Trader, with the salesperson |
Threshold 0.5. One product, this month.
| Client | Salesperson | Baseline inquiries/wk | This month | Ratio | Desk ratio | Prior notional/month | Quiet score |
|---|---|---|---|---|---|---|---|
| 2207 | S-04 | 18 | 6 | 0.33 | 0.92 | $410m | $61m |
| 4471 | S-04 | 9 | 4 | 0.44 | 0.92 | $120m | $15m |
| 9034 | S-11 | 22 | 12 | 0.55 | 0.92 | $380m | not quiet |
| 1187 | S-11 | 14 | 5 | 0.36 | 0.41 | $200m | market, not client |
Client 2207 has cut its inquiries by two thirds in a product where the desk is flat, and the flow at stake is the largest on the list. Client 1187 is down by a similar proportion in a product where every client is, and it is not on the list.
| Client 2207 | Rates | Credit | FX |
|---|---|---|---|
| Quiet ratio | 0.33 | 1.05 | 0.98 |
Active in two products, quiet in one. The conversation is about rates pricing or coverage, and the report says so before the salesperson picks up the phone.
Global threshold on counts. The largest clients are always quiet and the smallest never.
Desk-wide decline ignored. The whole client list is quiet in a slow month.
Inquiries and trades blended. A hit-ratio problem is worked as a relationship problem.
Products blended. The client is fine on average and gone in rates.
Mapped once, the RFQ log, the blotter and the coverage file produce the baselines, the quiet lists and the scores every week. Covirage builds this from the exports as they are. The trading page describes the setup, and the hit ratio guide covers the second list's measure in depth.
Both, separately. A client still sending inquiries but no longer trading has a pricing or hit-ratio problem; one no longer sending inquiries has gone elsewhere or gone quiet in the market. The two lists have different owners.
By comparing the client's decline to the product's decline across all clients in the same period. A client down 60 percent in a product where the desk is down 55 percent is not quiet; the market is. The report shows both figures.
The RFQ log and the trade blotter with client identifier, product, date and notional. No names. Where the desk has a coverage assignment file, the salesperson per client; otherwise the list is per product.