Your data. Your licences. Your numbers.
We managed portfolios before we built this, and we built it as a portfolio construction aid — not a reporting system that PMs are shown once a month. It runs daily, from your transactions, inside your own Azure tenancy, on your prices, your classifications and your analytics.
We are onboarding a small number of early partner firms ahead of general availability. Give us a month of your own data and we will return a full attribution — in numbers you can check against your own systems. The first month is free when we open, and we would welcome your view on what is missing.
Your licences, reused — not replaced
Benchmark and analytics data stay inside your tenancy, under the agreements you already hold. Nothing is redistributed, so there is nothing new to licence or pay for.
No residual to explain away
Every effect is derived from the same transaction-level return calculation, so the effects sum exactly to the active return. There is no smoothing term, because there is nothing left over.
Per portfolio, not per seat — on purpose
Unlimited users, deliberately: the more people looking at the results, the more the results are worth. Mapping, re-runs and new portfolios are self-service too.
Why the calculation is the product
Daily · transaction-level · decision supportMonthly, holdings-based attribution quietly hides the things a portfolio manager most wants explained: the intra-month trade, the corporate action, the cash that sat idle for nine days. Attribution Express calculates every portfolio daily, from the transactions, so returns compound the way the book actually behaved and every effect traces back to a fill.
Because every effect falls out of that one calculation rather than a separate analytics model, the parts add up. There is no residual — no plug, no smoothing term, and no conversation about why the pieces do not reconcile to the total. Attribution that arrives monthly explains history; attribution that arrives daily can still change a decision.
Instead of debating whether the numbers are right, people start looking at what the numbers mean
It runs on the prices, classifications, durations and yields you already use — data that is already feeding your other systems, with people looking at it every day. That is precisely why it is cleaner than anything re-derived downstream, and why the exposures your PMs see are the ones they expect.
Brinson-Fachler
Equity · Multi-assetAllocation measured against the benchmark’s own total return
Brinson-Hood-Beebower
EquityWhere the house convention is absolute sector allocation
Karnosky-Singer
Global · HedgedCurrency separated from the market decision, with the forward premium shown explicitly
Modified van Breukelen
Fixed incomeCarry, shift, twist, spread and issue selection from the same daily calculation
GIPS composite management
IncludedComposite membership, dispersion and compliant presentations
Classifications are yours — sector scheme, desk, rating band, strategy, desk, rating band, duration bucket or any custom hierarchy you already maintain — and drill from the total effect straight down to the individual security.
Where the data comes from, and where it stays
Straight-through, no human in the middleIndependently certified — the company, not just the claim
CloudAttribution Ltd, the company behind Attribution Express, is assessed and registered by NQA under UKAS accreditation. Certification covers our management systems for the provision of SaaS software and an outsourced performance and attribution team to the fund management industry.
- ISO/IEC 27001:2022 Information security management View certificate 175611 Valid to 3 February 2028
- ISO 9001:2015 Quality management View certificate 177960 Valid to 19 May 2028
What your analysts actually see
Illustrative · figures are worked examples| Sector | Port wt % | Bmk wt % | Port ret % | Bmk ret % | Allocation | Selection | Interaction | Total bp |
|---|---|---|---|---|---|---|---|---|
| Information Technology | 24.8 | 20.1 | 9.42 | 8.05 | +23.2 | +27.5 | +6.4 | +57.2 |
| Financials | 18.2 | 16.4 | 4.16 | 3.88 | +1.4 | +4.6 | +0.5 | +6.5 |
| Health Care | 11.9 | 14.7 | -1.35 | -0.42 | +9.9 | -13.7 | +2.6 | -1.2 |
| Industrials | 13.6 | 12.8 | 5.71 | 4.93 | +1.5 | +10.0 | +0.6 | +12.1 |
| Consumer Discretionary | 9.4 | 11.2 | 2.08 | 3.36 | -0.5 | -14.3 | +2.3 | -12.5 |
| Energy | 5.1 | 7.6 | -2.94 | -3.71 | +17.0 | +5.9 | -1.9 | +21.0 |
| Consumer Staples | 7.8 | 8.9 | 1.22 | 1.05 | +2.3 | +1.5 | -0.2 | +3.6 |
| Utilities & Other | 6.5 | 8.3 | 0.84 | 1.17 | +3.5 | -2.7 | +0.6 | +1.3 |
| Cash | 2.7 | 0.0 | 0.31 | 0.00 | -8.4 | +0.0 | +0.8 | -7.6 |
| Total portfolio | 100.0 | 100.0 | 3.91 | 3.11 | +49.9 | +18.7 | +11.8 | +80.4 |
Read it as a portfolio manager would: 80.4 bp of active return — the technology overweight and the energy underweight gave 23.2 bp and 17.0 bp of allocation, health care cost 13.7 bp in selection, and a 2.7% cash balance cost 8.4 bp. The effects sum to the total exactly, and every cell drills through.
Currency, decomposed not netted
| Currency | Active exp % | Ccy return % | Fwd prem % | Effect bp |
|---|---|---|---|---|
| USD | +3.20 | +2.45 | -0.18 | +7.3 |
| EUR | -2.40 | -1.62 | +0.09 | +3.7 |
| JPY | +1.10 | -3.10 | -0.44 | -3.9 |
| CHF | -1.90 | +1.08 | +0.21 | -2.5 |
| Currency management | +4.6 |
The currency move and the forward premium are shown separately, so a hedging decision is judged on its own merits. Exposures come from your analytics, so they match what your PMs already quote.
Fixed income, same calculation
| Effect | bp |
|---|---|
| Carry & roll-down | +14.2 |
| Curve — shift | +9.6 |
| Curve — twist | -3.1 |
| Spread / sector | +11.4 |
| Issue selection | +6.8 |
| Currency (hedged) | -2.3 |
| Active return | +36.6 bp |
Sterling credit sleeve, same quarter. Note what is missing: no residual line — the effects come out of the return calculation, not a model fitted alongside it.
From signature to first signed-off month
Weeks, not months or years-
Week 1
Tenancy up
Your Azure environment is provisioned and joined to your identity provider — SSO from day one.
-
Weeks 1–2
You map your data
Drag your extract in. The mapper shows real values beside each field, so you pick the column by looking at it.
-
Weeks 2–3
Check, don’t reconcile
Same prices and analytics as your book of record, so differences are real exceptions — with the trade that caused them.
-
Week 4
Sign off and report
First month closed and reported. Output lands in Power BI, Snowflake or your own warehouse.
Who supplies what
You provide
- Holdings, transactions & your prices — SFTP, API or file drop
- Sector, strategy & rating classifications
- Durations, yields & analytics your PMs already quote
- Benchmark data, on your existing licence
We add
- Market reference rates — FX, cash, forwards, curves
- The calculation engine, models and interface
- The Azure environment, in the UK or Europe
Why this is different
By portfolio managers
We managed portfolios in the front office before we wrote a line of this, and designed it as a portfolio construction aid — so the interface is built for a PM’s desk, not a monthly pack.
Per portfolio, never per seat
Deliberate. Charging per user puts a price on every extra pair of eyes, and the value here is more people looking at the same numbers. So we don’t.
Already checked, by your own people
Your prices and analytics feed your other systems, where people look at them daily. Data under that many eyes is cleaner than anything re-derived elsewhere — and clients tell us that is what they notice first.
Nothing licensed twice
Benchmark and analytics data never leave your ownership, so the agreements you hold still cover it. No redistribution, nothing new to negotiate or buy.
See it run on your portfolios.
We are onboarding a small number of early partner firms ahead of general availability. Give us a month of your own data and we will return a full attribution — in numbers you can check against your own systems.
Register your interest — it takes a minute, and tells us what to build next.
Per portfolio · Unlimited users
First month free when we open. No implementation fee, no per-seat licence, and no data you have to licence twice.