Skip to content
Introducing complead. One AI-native platform for financial crime compliance. Read the story
New Ready Integrations available Check the new integrations

Swift screening, adverse media cases and rules from a prompt: complead April 2026 updates

In short

The April release brings screening into the payment flow and gives adverse media a workflow of its own. Swift messages are parsed and every party is screened. Adverse media can be scanned on its own, broken down by risk category and turned into a case. Customer risk formulas move to a visual canvas, an AI assistant builds them from a sentence, and customers can be handled in bulk.

A cross-border payment carries more names than most teams have time to read. There is the sender and the receiver, but also the banks in between, the party that started the payment and sometimes an ultimate beneficiary hidden a few lines further down the message. Checking them by hand means copying names out of a long, structured file and pasting them into a search box one at a time.

Adverse media had a similar gap. News about a customer appeared as a list of results, but there was no simple way to search the news on its own, see what kind of risk the articles pointed to, or keep the review in a case. And on the customer risk side, formulas that should have been simple often needed a long chain of conditions and a yearly update just to keep an age band correct.

We built the April release around these gaps. The highlights:

  • Swift messages are parsed, and the parties and banks inside them are screened in one step.
  • Adverse media can be scanned on its own, broken down by risk category and turned into a case.
  • Customer risk formulas are built on a visual canvas, and an AI assistant can draft them from a sentence.
  • Age is calculated from the date of birth, so age bands stay correct over time.
  • Customers can be screened, added to ongoing monitoring and rescored in bulk.

Payment screening

Swift message screening

The problem. A Swift message holds many parties, and screening them meant reading the message, finding each name and bank, and searching for them one by one. That is slow, and in a busy payment queue it is easy to miss the party that matters.

What changed. Swift messages are now parsed and screened in one step, in the interface and through the API.

How does Swift message screening work?

You paste a message, or send it through the API, and the platform reads its structure. It finds the parties, such as the debtor, the creditor, the ultimate debtor, the ultimate creditor and the initiating party, together with their bank identifiers. It also finds the banks in the chain, such as the instructing, instructed and intermediary agents. Each of them is screened against your lists in real time, people and companies by name and banks by BIC and bank name.

The screen supports the ISO 20022 MX formats used for customer and bank transfers, such as pacs.008, pacs.009 and pain.001, and you can still paste classic MT messages such as MT103. You choose whether a message is screened against your name screening lists, against adverse media, or both.

What happens when a party matches?

A message with a hit is recorded for review. The analyst sees which party produced the match, sets the match status, writes the reasoning and closes the case, with the same case tools used elsewhere in complead. Clean messages pass without extra work, so the team spends its time on the payments that need a second look.

Why it matters. Every party in the payment chain is screened, not only the names someone had time to copy, and the review stays on record.

Adverse media screening

Standalone scans, a category view and cases

The problem. Adverse media results came as a list attached to a name screening. There was no direct way to run a news search on its own, see at a glance what kind of risk the news described, or keep the review in a case of its own.

What changed. Adverse media became a workflow of its own, from the scan to the case.

How do you run a standalone adverse media scan?

From the left menu, open Daily Operations, then Manual Scan, then By Adverse Media. Type a name and the news about it is listed with its source, date and risk category. The search doesn't need a matching sanctions or PEP profile first, which is useful when you are checking a counterparty, a new supplier or a name that came up in an investigation.

What does the category breakdown show?

On the results and on the case detail, an infographic shows how the articles are spread across risk categories:

  • Financial Crime
  • Terrorism and Sanctions
  • Violent and Organized Crime
  • Legal and Regulatory
  • Cyber and Data Security
  • General Adverse Media

The breakdown tells you in a few seconds whether a name is in the news for a tax dispute or for something far more serious, before you read a single article. Each article also carries a short summary, so the reading that follows is faster too.

How does a scan become a case?

A new Adverse Media tab sits in Alert and Case Management. From a name screening with no match, you can run an adverse media search for the same name and keep the result as a case. From a list of profiles, you select the ones you want and turn them into an adverse media case in one click. The case can then be assigned, noted and closed like any other.

Why it matters. News checks get the same trail as sanctions checks: a scan, a case, a decision and a reason.

One platform for AML, KYC and fraud.

Screening, transaction monitoring and client risk on one engine, with one risk score for every client.

800+companies use complead
80+countries served

Customer risk

Customer formulas on a canvas, with an AI assistant

The problem. Customer risk formulas were written in a form, and a formula with several conditions grew into a long chain of nested statements that was hard to read and harder to change. An age band also needed a stored age that someone had to refresh every year.

What changed. Customer formulas moved to the Canvas Rule Editor, and AI can build them for you.

How does the canvas work for customer formulas?

The Customer Formulas screen now uses the same drag-and-drop canvas as transaction rules. You place nodes for customer fields, comparisons and score actions, and connect them to describe the logic. Numeric, text and date conditions can all be combined on the same canvas. A new component replaces long chains of nested conditions, so a formula with several branches stays readable.

How does the AI rule assistant help?

Next to the canvas sits an AI assistant. Describe the formula in a sentence, for example "add risk points when a corporate customer is registered in a high-risk country and was founded less than two years ago", and the assistant draws the nodes on the canvas. You review the result, adjust what you need and save. The formula is yours to read and change, the assistant only saves you the first draft.

How is age calculated?

A customer field called Age is calculated from the date of birth at the moment the formula runs. A rule such as "customers under 25" stays correct as time passes, without anyone updating stored ages each year.

Why it matters. Risk logic is easier to build, easier to read in an audit and stays accurate without yearly maintenance.

Customer management

Bulk actions

The problem. Screening a group of customers, adding them to ongoing monitoring or changing their risk scores meant opening each record in turn.

What changed. The customer list gained bulk actions.

Which actions can be run in bulk?

Select customers on the Customer Management screen and run, in one step:

  • A name screening for all of them.
  • Adding them to ongoing monitoring.
  • Raising or lowering their risk score.

A typical use is a periodic review. Filter the list down to a segment, for example customers from a country whose risk rating has just changed, select them, screen them again and adjust their scores, without opening the records one by one.

Why it matters. Work that used to take an afternoon of clicks takes a few selections.

Everyday work

Smaller changes you'll notice during the day

  • Auto resolution by list type: case-based auto resolution can use a separate fuzzy rate threshold for sanctions, PEP and watchlist results.
  • One transaction, several rulesets: a transaction can run through more than one ruleset at the same time, for example one for the card and one for the merchant.
  • Rule performance dashboard: a dashboard for analysing how transaction monitoring rules perform.
  • Canvas rules as JSON: rule scenarios built on the canvas can be exported and imported as JSON files.
  • Company or individual: Alert and Case Management can be filtered by merchant (company) and individual customers.
  • Editable customer numbers: admins can add or edit a customer number after the customer is created.
  • Lists and exports: Customer and Account Management gained blacklist and whitelist filters and an Excel export.
  • Exports that follow your columns: PDF and Excel exports keep the columns you chose on the screen.

If you already use complead, the details of each update are in the Help Center inside the app. For other months, see the Product Updates category.

Next month: May 2026 updates.

See it with your own payments.

We'll show you Swift message screening, standalone adverse media scans and AI-built customer formulas with your own examples.

7-day free trialNo credit card required
800+companies use complead
80+countries served

Frequently asked questions

Which Swift formats can be screened?

ISO 20022 MX messages such as pacs.008, pacs.009 and pain.001, and classic MT messages such as MT103.

Are banks in the payment chain screened too?

Yes. Agents such as the instructing, instructed and intermediary banks are screened by BIC and bank name.

Do I need a name screening match before an adverse media scan?

No. A standalone scan searches the news for any name, and the result can be kept as a case.

Does the AI assistant save formulas on its own?

No. It draws the formula on the canvas, and you review and save it.

Enes Keltepe
Written by Enes Keltepe Technical Product Manager Enes Keltepe is a Technical Product Manager at complead with a background in software development. He focuses on product development, software engineering and financial technology.

Originally published

Back to blog