3 way reconciliation

A step-by-step guide to combining two or more two-sided reconciliations into a single consolidated view.

In this article you will find below sections:

• Overview
• Building your process 
• Verifying the result
• Working with results
• Troubleshooting / FAQ

Overview

3 way reconciliation, also known as a Multi recon view (MRV) is an overlay layer that sits on top of reconciliations you already run. It doesn’t touch the underlying processes, it consolidates their results so you can see how records line up across all of them at once. This article walks through building an MRV process that consolidates two two-sided reconciliations sharing a common middle office data source: one against front office data, one against broker data.

By the end of this article you’ll be able to:

• Create a multi recon view process and add the underlying processes it should consolidate
• Map fields and choose grouping keys so records from each source join up correctly
• Read the new grouping statuses (Match, Partial match, Missing, Incomplete, Duplicate, Standalone) and investigate a break
• Use workflow actions and comments from the consolidated view and understand how they sync back to the underlying processes

Before you start

Each two-sided process you want to bring into an 3 way recon process must have:

• Record tracking enabled
• Exceptions workflow turned on
• Not already be part of a connected process chain

Building your process

Step 1 - Create the multi recon view process

1. From the homepage, go to Processes and click Create process.
2. Under process types, select Multi-Recon View.
3. Choose a Use case (for example, Trade lifecycle and control).
4. Enter a Process name, Code, and Data label (plus a plural form if you’d like one). These work the same way as for any other Duco process.
5. Click Create process.

You can also add a description on the General settings page afterwards, and delete the process from there if you no longer need it.

Step 2 - Add the underlying processes

1. Open the Data inputs settings page.
2. Click Add process and search for each two-sided process you want to consolidate.
3. Select a process and click Add. Repeat for the second process.

Not every process is eligible. If a process can’t be selected, click its warning triangle to see why.

Once both processes are added, choose the base source: the data input that’s common across both underlying processes (in this example, the middle office trades data). This is what the other inputs are compared against.

Step 3 - Map the fields

Field mappings tell Duco which fields from each underlying process should be compared to each other in the consolidated results.

1. Open the Field mappings settings page.
2. Click Add field, give it a name, and choose its data type (Text, Decimal, Date, and so on).
3. For each of the two underlying processes, choose the matching field from the dropdown. Field names don’t need to match across processes, only the data type does.
4. Repeat for every field you want visible in the results, for example Trade ID, Instrument, Account, Amount, Price, Trade Date, Settlement Date, Currency, and Counterparty.

Step 4 - Set grouping keys

Grouping keys determine how records from the underlying processes are joined together into a single row in the consolidated results. Every field you choose here must be present in all of the underlying processes, and there’s only one set of grouping keys per multi recon view process.

Back on the Data inputs page, select the fields you want to group on, for example Trade ID, Account, Currency, and Trade Date. A preview of sample data from each process is shown underneath so you can check the values line up as expected.

Step 5 - Set permissions

Permissions are inherited automatically from the underlying processes, so you don’t need to configure them twice. Anything a user or group can do in an underlying process, they can also do from the multi recon view.

The only things you can add directly at the multi recon view level are config admin and user admin access. The Permissions page also shows you the users and groups already permissioned on each underlying process, so you don’t have to check each one separately, but you can’t edit those permissions from here. Click through to the underlying process if you need to change them.

Running it

Before you exit settings, rerun each of the underlying processes so the latest data is pulled through under your new configuration. There’s a short wait after the runs complete before the multi recon view results are ready, while the consolidated data is built.

Verifying the result

Once the runs finish, the Overview page gives you the same kind of summary you’d see for any other process: submission data, run status per underlying process, and a breakdown by grouping status and exception workflow status. Every chart is interactive, click a bar to jump straight to that bucket in the results.

The six grouping statuses

Multi recon view introduces six new statuses that describe how a group of records compares across your sources:

• Match - all sources are present and agree.
• Partial match - some sources agree, but others break (for example, a mismatched account number).
• Missing - one or more expected sources are absent from the group.
• Incomplete - a record exists in the base source and only one of the underlying processes.
• Duplicate - more than one record exists for a source within the group.
• Standalone - a record exists in one of the other sources but has no corresponding record in the base source, so it isn’t grouped with anything.

Investigating a partial match

Use Expand all to open every group and see the parent record alongside its children from each underlying process. Fields that are part of your grouping keys are marked with a star. Where a partial match breaks on a specific field, for example an account number, that field is highlighted so you can see exactly what’s causing the group not to fully agree.

Click a record’s ID to jump to a pre-filtered view of that record inside its underlying process, useful when you want to see the full detail, history, and comments for that specific trade.

Missing and other statuses

Records in the Missing bucket show which source or sources didn’t produce a matching record for that group, for example a trade the middle office has but the broker hasn’t reported.

Working with results

Everything you’d normally do in a two-sided process is available from the multi recon view too: adjusting, hiding, and reordering columns, filtering, sorting, adding comments, and changing workflow status.

You can also assign a group of records to a user or group, or run a workflow action such as Start review directly from the multi recon view. The workflow status columns show you the state of each source’s record, alongside the overall grouping status.

Troubleshooting / FAQ

Why can’t I select a process when adding it as a data input?

The process is either missing record tracking, doesn’t have exceptions workflow turned on, or is already part of a connected process chain. Click the warning triangle next to the process to see which restriction applies.

Why are some fields missing from the Field mappings dropdown?

At the moment, only match fields are supported in field mappings. Reported fields and calculated results aren’t available to select yet.

Does a multi recon view process count towards our process cap?

No. Multi recon view processes don’t count towards your organisation’s process cap, only the underlying two-sided reconciliations they consolidate do.

Can I change permissions for an underlying process from the multi recon view?

You can see who has access, but not change it from there. Click through to the underlying process’s own Permissions page to make changes.

How many processes can I combine in one multi recon view?

At the moment, a multi recon view process consolidates exactly two underlying two-sided processes, sharing one common base source.