Sign in

Blog · Alternatives and comparisons · Trading

How to choose analytics software for trading: questions, data and traps

How desks should choose analytics software: start from the questions, check the data you hold, ask vendors ten questions, avoid the traps.

By CoviragePublished 24 September 20264 min read
The short answerStart from the questions desks ask every month, not from features. List the exports you already hold, ask every vendor what it needs before the first answer and whether its AI calculates figures, and check that every total reconciles. Then compare the first-year cost, all in.

Most buying decisions for analytics start from a feature list. For desks the better start is the questions that come back every month, the files already on hand, and the traps that make a tool look right in a demonstration and wrong in the first board meeting.

Start from the questions

The question The measure behind it
Which clients trade with us by count but not by size? Size-weighted hit ratio
Who has gone quiet? Quiet client list
What share of each client's inquiries do we see? Inquiry share
Are clients trading the products we are axed in? Axe hit rate
What did we trade that no salesperson had sent? Traded unsent
How fast do we answer, and does it matter? Response time on inquiries

Any tool you consider should answer these from your data, not from a sample. Ask to see it.

The data you already hold

  • Inquiry log
  • Blotter
  • Venue or client-stated data
  • Revenue attribution
  • Axe log
  • Distribution records
  • CRM or chat logs

If a vendor needs a warehouse built before it can read these, count that in the cost and the time.

Ten questions to ask any vendor

  1. What does it need in place before the first answer? A warehouse, a data model, a modelling language, a partner? Ask for the list and the typical weeks.
  2. Who does the setup, and who maintains it? Your team, the vendor, or a partner, and what that costs after year one.
  3. Does the AI calculate figures, or choose from computed ones? A language model that writes queries or code can produce a plausible wrong number. Ask what it is allowed to do.
  4. Does every total reconcile to a control figure? Ask to see a bridge that does not sum and what the product does about it.
  5. Can every figure be opened to its rows? An answer nobody can check becomes a debate in the meeting.
  6. What does it cost in the first year, all in? Licences, consumption, implementation, modelling and training, not only the seat price.
  7. How does data arrive, and who holds credentials? A file your systems already export, a scheduled drop, or a live connection with the vendor holding keys.
  8. What happens to the data, and where is it stored? Residency, retention, deletion, and whether names can be replaced with identifiers.
  9. Can we see it on our own data before we sign? A demonstration on a sample dataset tells you little about your own.
  10. What does the tool do when it cannot answer? It should say so. A confident guess does more harm than no answer.

Checks specific to trading

Ask whether the tool enforces these, and what it does when they fail:

  • Hit ratio: Inquiries = traded + traded away + passed + no quote
  • Axes: Matched trades are a subset of the blotter; each trade matches at most one axe
  • Tiers: Every client is in one tier; client revenue sums to the desk total
  • Coverage: Clients = covered + uncovered

The traps

Size-weighted hit ratio. Count-weighted is the default on every venue report.

The quiet client list. Revenue reports show who traded, not who stopped.

Traded unsent. Nobody joins the axe log to the blotter for clients who were not on the list.

Measures to leave out

Inquiry count as a measure of relationship. Price-checkers inquire most.

Axes sent. Volume of distribution is not effectiveness.

Voice and electronic hit ratios blended. Two different competitions averaged into one number.

A scorecard

Criterion Weight Tool A Tool B Covirage
Answers our six questions on our own data High
Time to the first answer High
Needs a warehouse or data team Medium
AI calculates figures, or only explains computed ones High
Every total reconciles; figures open to rows High
First-year cost, all in Medium

Where Covirage fits

Covirage reads the exports above, answers the questions with figures our tools compute and check, and is set up for you within a week. See analytics software for trading compared, AI analytics for trading and Covirage for Trading.

For the measures in full, with formulas and exports, read Client KPIs for sales and trading desks.

Questions people ask

What should desks look for in analytics software?

The answer to their own questions, from the data they already hold, with every figure reconciled. Features matter less than what the tool needs before the first answer and who maintains it.

Is a BI suite enough for desks?

It can be, with a warehouse and someone to build and maintain the model. Without them, the dashboard shows what changed and the explanation is still an analyst's job.

What data do desks already hold?

Usually: inquiry log, blotter, venue or client-stated data, revenue attribution, axe log, distribution records. Most analytics questions in this industry can be answered from those exports.

Related reading

Coverage and territory · Trading

Axe distribution: which clients receive the desk's axes, and which ones trade them

How a trading desk measures whether its axes reach the clients who trade them, from the axe log, the distribution records and the blotter: axes sent per client per product, axes traded within a stated window, the clients who receive many and trade none, the clients who trade the desk's axes but receive few, the salespeople whose distribution lists have not changed in a year, and the identity that trades against axes sit inside the blotter.

16 Sept 20262 min readTerritory, capacity and quota planning · Trading

Client tiering by flow and hit ratio: where the desk should spend its axe

How a trading desk tiers its clients from the RFQ log and the blotter rather than from the coverage list: two axes, notional traded and hit ratio, four tiers that follow, the clients with high inquiry volume and low hit ratio who consume pricing effort for little, the clients with high hit ratio and modest flow who would trade more if shown more, and the quarterly re-tiering that moves clients on their own numbers.

16 Sept 20262 min readWallet share and penetration · Trading

Hit ratio by client and product: a trading desk's guide to flow share

How a sell-side desk computes hit ratio and flow share per client and product from its own RFQ logs and blotter, why the two numbers must be read together, and the three drifts that predict a client is leaving before the revenue shows it.

16 Sept 20264 min readWallet share and penetration · Trading

Size-weighted hit ratio: why the count-weighted number hides the losses

Why a trading desk's hit ratio should be weighted by ticket size as well as counted, how the two diverge when a competitor commits balance sheet, the computation from the RFQ log, and a worked example where a rising count-weighted ratio masks a falling share of volume.

16 Sept 20263 min read