Processes

How to Speed Up and Simplify Your Security Process

A combination padlock, a toy rocket, and a small black alarm clock

Published On: 3rd Nov 2026 08:00

Security checks are essential to prevent fraud, but that doesn’t mean they need to slow every customer conversation to a crawl. So, how can contact centres keep the right safeguards in place without adding unnecessary friction?

It’s a fine balancing act, but it isn’t impossible – as our consultants panel shared when we asked them for their top tips and advice on speeding up and simplifying security processes.

Swap Memory Tests for Things Customers Have

Katie Stabler, Founder and Director of Customer Experience at CULTIVATE Customer Experience by Design — Katie Stabler

Swap memory tests for things customers have! Questions like “what was your last transaction?” trip up genuine customers and are easy for fraudsters to research.

A one-time passcode to a registered device, or passive voice biometrics, is faster, more secure and far less stressful.

Explain “Why” to Help Customers Relax and Cooperate

Always explain “why” you are doing security checks. When agents say, “I’m asking this to keep your account safe” customers relax and cooperate.

And design for the edges. Customers who are vulnerable, bereaved, or without a smartphone need an alternative route that doesn’t leave them stuck.

Contributed by: Katie Stabler, Founder and Director of Customer Experience at CULTIVATE Customer Experience by Design

Make Sure Their Verified Identity Travels with Every Transfer

Chris Marron, CMA Intelligence — Chris Marron

Too often, security checks create friction because the contact centre treats each stage of the customer journey as a fresh interaction. The handover from bot to agent is a well-known problem, but one many brands still get wrong.

Customers should not have to authenticate again simply because they are being transferred, especially after explaining their issue to the bot. Their verified identity and the conversation context should travel with them.

Getting this right requires compliance and IT teams to agree how authentication is recorded, shared and trusted, as well as for the right supporting technology. It can be challenging, but it is a challenge worth working through. If your contact centre isn’t getting this right, this is the first place to start.

This same principle also applies to mobile apps and web portals. If a customer is already logged in, use that authenticated context in the support experience.

Contributed by: Chris Marron, Industry Analyst at CMA Intelligence

Stop Putting Customers Through Questions for No Clear Reason

Dave D’Arcy, Global award-winning customer experience leader — Dave D’Arcy

Are we checking security, or buying the agent time to read the notes?It’s an uncomfortable question, but I’ve seen security routines become so embedded in the opening of a call that nobody stops to ask what each question is achieving.

If the agent needs a minute to understand the account, let’s be honest with the customer: “Give me a moment to read through what’s happened so I can help you properly.” Most people will appreciate that far more than being put through five questions for no clear reason.

Contributed by: Dave D’Arcy, Founder and MD of Laughing Leadership

Take Sensitive Data Out of the Conversation Entirely

Iqbal Javaid, CTO and Director at Evolved — Iqbal Javaid

We recently introduced a voice bot to a customer that handles a simple ID&V process before the call reaches an agent. It asks the customer for their details, checks them against the backend database in real time, and passes the call through already verified.

The agent sees a clear verified flag and can open with “I can see you’re already verified, how can I help?”. On average, that trimmed 90 seconds off every call the agent handled.

Also, take sensitive data out of the conversation entirely. For payments, keypad entry with tone masking means card details never reach the agent or the call recording.

Contributed by: Iqbal Javaid, CTO and Director of Evolved

Switch to Risk-Based Authentication for a Better Starting Point

Most security processes are built to catch the one-in-whatever caller that is an impostor, so all the others face the same barrage of questions, often just to ask something simple.

Risk-based authentication offers a better starting point: context. Why is this customer calling?

Picture two customers calling their bank. The first wants to check whether yesterday’s payment went through. The second wants to make a transfer. Today, both get the same questions. Only one of them really needs them.

The first customer only wants confirmation of something she already knows. The risk is low, so the check can be light: a recognised number may be enough, while an unknown number adds a question or two.
For the second, full verification makes complete sense because money is about to move, and the customer understands that too. (If the first customer then asks for a transaction, verification steps up at that moment.)

Note, this only works as part of the overall security strategy – not as a shortcut on the phone.

Contributed by: Dave Sneyders, Technology Strategist & Business Innovator

Use the Number the Customer Calls From as a Helpful Risk Signal

The number a customer calls from is already in your data. A number that has called many times, consistent with their app and web activity, is a lower-risk signal than a first-time one.

Tier the checks: familiar patterns get a lighter touch, while anything new, or any high-risk request like changing bank details, steps up.

Customers feel the experience is built around them, which builds trust. Just be aware that caller numbers can be spoofed, so treat them as one signal – never proof!

Contributed by: Natasha Doren, Customer Experience Consultant

Continuously Adjust ID&V Software to Recognise Different Dialects

Simon Korsholm, Head of Contact Centre Transformation, PMO & Special Projects at HGS - Hinduja Global Solutions — Simon Korsholm

In a previous leadership role with a high-volume retail organisation, I introduced speech-enabled IVR to provide self-service ID&V.

With the CS team receiving 68,000 calls a week, this was a costly operation, so every small gain in efficiency had a significant commercial impact, as well as providing an opportunity to improve the customer experience.

The solution achieved an 85%+ handling rate across all ID&V calls, with each successful process saving an average of 30–40 seconds per call.

The agent simply greeted the customer, thanked them for completing the security check and immediately moved on to address their reason for calling.

But technology is only part of the answer. An additional piece of work often neglected is the time spent listening to live interactions and continuously adjusting the ability of the software to recognise different dialects and speech characteristics.

Contributed by: Simon Korsholm, Head of Contact Centre Transformation, PMO & Special Projects at HGS – Hinduja Global Solutions

Take the Clock off Security for Callers Who Stammer

Steve Morrell, Managing Director, ContactBabel — Steve Morrell

For many people who stammer – this applies to well over 500,000 adults in the UK alone – security is the hardest part of the call: saying specific words, numbers and answers aloud, under time pressure, to a stranger.

Yet ContactBabel research for STAMMA (the charity representing people who stammer) finds that where a speech disability is on record, only 12% of UK contact centres show it to the agent before identity verification. 21% only show it afterwards, and 37% leave agents to find it in the account notes.

A voicebot handling authentication can change that. Recognising the caller’s number, it can find the flag before anyone speaks and run the same ID&V checks an agent would, but without the clock: it doesn’t interrupt, doesn’t finish sentences and can offer keypad entry or a text-message passcode instead of speaking. None of this needs the expense of voice biometrics, as the bot simply takes over the existing process from the agent or IVR.

The caller then reaches an agent already verified, with the flag on screen, so they don’t have to disclose again or go through security twice. Customers dislike breaking channel, but this is the same call, and with 44% of contact centres never training agents on speech disability, the most pressured moment no longer depends on an agent knowing how to help.

The wider saving holds even when the bot can’t resolve the enquiry: agent-handled verification averages 48 seconds and costs the UK industry over £3bn a year.

The catch is configuration: a voicebot tuned for the average caller will cut in on a block or a long pause, which is worse than a patient agent, so generous silence timeouts, keypad alternatives and an easy route to a person at any point are essential. Get those right and security becomes quicker for every caller, and far less stressful for those who stammer.

Contributed by: Steve Morrell, Principal Analyst, ContactBabel

**Figures from ContactBabel’s Serving Callers Who Stammer and The Inner Circle Guide to Voice AI (UK edition)

If you want to find out more about handling contacts from customer who stammer, read our article: Simple Changes to Help Customers Who Stammer

★★★★★

What Have You Tried to Speed Up and Simplify Your Security Process?

Join our LinkedIn community and let us know.

If you are looking for more information to help improve your contact centre processes, read these articles next:

Author
Megan Jones

Megan Jones, Editor, Call Centre Helper

Megan interviews inspiring people across the contact centre industry to capture these amazing voices and turn them into best practice articles – covering all elements of what it takes to deliver truly outstanding employee and customer experiences.

Connect with Megan on LinkedIn