.jpg)
Why the industry’s automation problem is really a data problem
10 Jun 2026
Michael Wood, general manager for Asset Servicing at Broadridge, speaks to Zarah Choudhary about how rising event volumes are exposing weaknesses in operating models and why many firms are placing greater emphasis on trusted data as the foundation for scalable automation and AI
Image: Broadridge
Asset servicing volumes are continuing to rise, yet many firms are seeing automation stall or even reverse. What is driving that trend, and what does it reveal about the current state of operations?
You are right, volumes are growing significantly. Our recent industry survey conducted with The ValueExchange found that global event volumes are up nearly 25 per cent year-on-year, while investor elections are up 31 per cent.
That is being driven by factors such as the democratisation of investing, generational wealth transfer, a growing variety of asset classes, and increasingly frequent and complex events. Looking ahead, developments such as tokenisation could further accelerate that trend.
Firms have invested heavily in automation, but the issue is not a lack of automation ambition. The challenge is operational confidence in that automation. Where data is inconsistent, incomplete, fragmented, or held in multiple formats, operations teams become reluctant to allow work?ows to run untouched.
The result is a pragmatic response: people are reinserted into the process to check, validate, and reconcile data. While that may reduce perceived risk, it also slows operations, increases costs, and limits scalability.
Increasingly, the challenge is less about automation itself than about the quality and consistency of the data feeding it. If firms do not trust the data and inputs, they cannot trust the automation. The level of scale they can achieve is therefore capped by data quality rather than the technology itself. AI can help, but its effectiveness depends heavily on the quality and standardisation of the underlying data.
Despite significant investment in automation over recent years, operations teams still rely heavily on manual checks, even in areas such as mandatory events. Why is manual intervention still viewed as necessary?
Manual intervention still feels like the safest option when there is uncertainty around the underlying data. Firms are not resisting automation, they are resisting the automation of poor-quality data and ?awed processes.
Even in mandatory events, where a high degree of standardisation exists and automation levels should be high, teams often chooseto validate information manually because the risk of acting on inaccurate or incomplete data remains too great.
Our survey found that activities such as income processing, which should be highly automatable, still account for around 58 per cent of total asset servicing spend. The issue is not a lack of belief in automation. Rather, firms do not want to automate bad data at scale.
Over time, however, manual controls cease to be true exception management and instead become the operating model itself, which is neither efficient nor scalable.
How serious is the risk of operations teams becoming the ‘control layer of last resort’, and why should senior leaders be paying closer attention to this issue?
It is a very real risk and, in many organisations, it is already happening.
You cannot solve structural fragmentation simply by adding more people. When infrastructure, processes, and data remain fragmented, operations teams inevitably become the control layer of last resort.
Increasingly, they are absorbing the consequences of upstream data issues, integration challenges, fragmented systems, inconsistent market practices, and unclear accountability across the asset servicing chain.
In effect, they become the final checkpoint before risk reaches the client or, in the worst-case scenario, the regulator.
Our survey suggests that while event volumes are growing by almost 25 per cent annually, headcount growth is running at only 2-6 per cent. Without greater automation, firms face significant scalability challenges, governance concerns, and growing keyperson risk.
What is the long-term cost of doing nothing? If firms continue relying on manual interventions and fragmented controls, what could that mean for operational risk, scalability, and costs over the next few years?
Ultimately, the cost of doing nothing is higher than many firms realise because it compounds over time.
Manual processes and controls can create a sense of stability, but they increase costs year after year while failing to address the underlying causes of inefficiency. They also cannot keep pace with growing volumes.
There is also a talent dimension. If experienced operational professionals are spending their time dealing with repetitive exceptions and routine repair work, it becomes increasingly difficult to retain and develop expertise.
While outsourcing can reduce costs and error rates by up to 25 per cent, long-term modernisation still depends on high-quality data and mature automation capabilities.
AI is powerful, but only when it sits on trusted and standardised data. Within our own outsourcing business, we have seen efficiency gains of around 30 per cent through the deployment of agentic AI, although the extent of these gains depends on the integrity of the underlying systems and data ?ows .
Data quality emerged as a major theme in the research. To what extent is today’s operational inefficiency a technology issue versus a problem of data trust, ownership, and accountability?
Technology is definitely part of the equation, but the bigger issue is data trust and ownership.
Our research found that poor data quality accounts for a significant share of the errors firms experience in this area, while 51 per cent of costs are associated with sourcing and validating data.
When data quality is inconsistent, firms respond by layering on controls, adding people, introducing more checks, and relying on manual judgement. That creates delays, increases costs, and introduces additional operational risk.
In many cases, firms need a unified model, fewer systems, and a clearer view across the process. That allows organisations to manage risk centrally, facilitate trusted automation, and deploy AI more effectively.
Many firms still see adding people as the safest response to growing complexity. Why is that no longer a sustainable operating model for asset servicing teams?
You cannot solve structural fragmentation simply by adding more people, particularly in a world where volumes are growing much faster than headcount.
The goal should not be to increase exception handling and manual checks. It should be to reduce the number of exceptions and checks required in the first place.
If the operating model depends on people reviewing large volumes of routine activity, firms are introducing inconsistency, delays, and additional risk into the process.
A sustainable model is one where systems handle volume and standard processing, allowing people to focus on exceptions, oversight, and the areas where judgement genuinely adds value.
That delivers greater scalability, stronger controls, and a better use of scarce operational expertise.
One of the challenges facing the industry is that much of that expertise is ageing out of the workforce. Attracting younger talent into specialist operational roles is becoming increasingly difficult, while transaction volumes continue to grow.
The industry cannot rely indefinitely on replacing expertise through headcount growth alone. That’s why better data, greater automation, and more integrated end-to-end operating models are becoming increasingly important.
If senior operations leaders accept that the current model is under strain, what does a practical path forward look like? Where should firms begin if they want to reduce risk without launching a full-scale transformation programme overnight?
The important thing is that firms start now, but they should not feel they need to fix everything at once.
Investment budgets are constrained, the cost of change is high, and transformation always carries some degree of risk. The most effective approach is often to begin with targeted, highvolume pain points where work?ows can be redesigned and automation increased.
By focusing on a manageable area, firms can demonstrate that improved data discipline and work?ow design deliver measurable operational benefits. They can then expand that success over time.
Where appropriate, firms may find greater value in embedding AI within the core process rather than layer on top of disconnected systems, with human oversight remaining in place where necessary.
Firms that take this approach will be in a much stronger position than those that continue relying on manual workarounds that simply mask underlying issues.
You have spoken about the importance of trusted data, clearer accountability, and scalable work?ows. What does ‘good’ look like in practice for firms trying to modernise their operating models today?
Good starts with standardised, well-governed, and high-quality data. It also requires integrated systems with fewer hand-offs and less fragmentation. That naturally reduces the need for duplicated checks and manual intervention.
The result is greater automation of routine processing, allowing experienced staff to focus on oversight and genuine exceptions. Our research suggests firms that invest in these areas can achieve returns on investment of between 12 -13 per cent.
Increasingly, modern operating models also need to incorporate AI within a human-supervised framework, supported by embedded dashboards, system integration, and end-to-end visibility.
The emergence of tokenised assets is also increasing pressure for more integrated operating models. Firms need infrastructure capable of supporting both traditional and tokenised assets side by side. Creating entirely separate work?ows for tokenised assets would simply introduce new complexity and inefficiency.
As volumes, complexity, and regulatory expectations continue to grow, how do you see the role of operations teams evolving over the next five years?
Throughout my career, operations teams have consistently been asked to do more with less, and that will continue.
The difference is that they should be spending far less time on repetitive validation and routine repair work.
Their role will become increasingly strategic, focused on oversight, risk management, governance, client service, and exception management.
AI will help reduce some avoidable exceptions and free people to focus on higher-value activities. It will be one of the key enablers that allows firms to scale without continuously increasing headcount.
Our research found that 57 per cent of firms identified trusted technology partners as an important enabler of this transition, particularly where they can support a future-state golden-copy data model.
If you had to summarise the industry challenge in one sentence, what would it be?
I think it might be quite a long sentence, but I guess asset servicing has reached a point where maintaining the status quo is no longer a neutral option.
Rising event volumes, shorter settlement cycles, growing regulatory expectations, AI adoption, and the emergence of tokenised assets are all colliding with legacy systems, fragmented data, and ageing operating models.
Without trusted data, integrated platforms, and more scalable operating frameworks, firms will struggle to keep pace with the demands being placed on them. That is the industry’s central challenge.
You are right, volumes are growing significantly. Our recent industry survey conducted with The ValueExchange found that global event volumes are up nearly 25 per cent year-on-year, while investor elections are up 31 per cent.
That is being driven by factors such as the democratisation of investing, generational wealth transfer, a growing variety of asset classes, and increasingly frequent and complex events. Looking ahead, developments such as tokenisation could further accelerate that trend.
Firms have invested heavily in automation, but the issue is not a lack of automation ambition. The challenge is operational confidence in that automation. Where data is inconsistent, incomplete, fragmented, or held in multiple formats, operations teams become reluctant to allow work?ows to run untouched.
The result is a pragmatic response: people are reinserted into the process to check, validate, and reconcile data. While that may reduce perceived risk, it also slows operations, increases costs, and limits scalability.
Increasingly, the challenge is less about automation itself than about the quality and consistency of the data feeding it. If firms do not trust the data and inputs, they cannot trust the automation. The level of scale they can achieve is therefore capped by data quality rather than the technology itself. AI can help, but its effectiveness depends heavily on the quality and standardisation of the underlying data.
Despite significant investment in automation over recent years, operations teams still rely heavily on manual checks, even in areas such as mandatory events. Why is manual intervention still viewed as necessary?
Manual intervention still feels like the safest option when there is uncertainty around the underlying data. Firms are not resisting automation, they are resisting the automation of poor-quality data and ?awed processes.
Even in mandatory events, where a high degree of standardisation exists and automation levels should be high, teams often chooseto validate information manually because the risk of acting on inaccurate or incomplete data remains too great.
Our survey found that activities such as income processing, which should be highly automatable, still account for around 58 per cent of total asset servicing spend. The issue is not a lack of belief in automation. Rather, firms do not want to automate bad data at scale.
Over time, however, manual controls cease to be true exception management and instead become the operating model itself, which is neither efficient nor scalable.
How serious is the risk of operations teams becoming the ‘control layer of last resort’, and why should senior leaders be paying closer attention to this issue?
It is a very real risk and, in many organisations, it is already happening.
You cannot solve structural fragmentation simply by adding more people. When infrastructure, processes, and data remain fragmented, operations teams inevitably become the control layer of last resort.
Increasingly, they are absorbing the consequences of upstream data issues, integration challenges, fragmented systems, inconsistent market practices, and unclear accountability across the asset servicing chain.
In effect, they become the final checkpoint before risk reaches the client or, in the worst-case scenario, the regulator.
Our survey suggests that while event volumes are growing by almost 25 per cent annually, headcount growth is running at only 2-6 per cent. Without greater automation, firms face significant scalability challenges, governance concerns, and growing keyperson risk.
What is the long-term cost of doing nothing? If firms continue relying on manual interventions and fragmented controls, what could that mean for operational risk, scalability, and costs over the next few years?
Ultimately, the cost of doing nothing is higher than many firms realise because it compounds over time.
Manual processes and controls can create a sense of stability, but they increase costs year after year while failing to address the underlying causes of inefficiency. They also cannot keep pace with growing volumes.
There is also a talent dimension. If experienced operational professionals are spending their time dealing with repetitive exceptions and routine repair work, it becomes increasingly difficult to retain and develop expertise.
While outsourcing can reduce costs and error rates by up to 25 per cent, long-term modernisation still depends on high-quality data and mature automation capabilities.
AI is powerful, but only when it sits on trusted and standardised data. Within our own outsourcing business, we have seen efficiency gains of around 30 per cent through the deployment of agentic AI, although the extent of these gains depends on the integrity of the underlying systems and data ?ows .
Data quality emerged as a major theme in the research. To what extent is today’s operational inefficiency a technology issue versus a problem of data trust, ownership, and accountability?
Technology is definitely part of the equation, but the bigger issue is data trust and ownership.
Our research found that poor data quality accounts for a significant share of the errors firms experience in this area, while 51 per cent of costs are associated with sourcing and validating data.
When data quality is inconsistent, firms respond by layering on controls, adding people, introducing more checks, and relying on manual judgement. That creates delays, increases costs, and introduces additional operational risk.
In many cases, firms need a unified model, fewer systems, and a clearer view across the process. That allows organisations to manage risk centrally, facilitate trusted automation, and deploy AI more effectively.
Many firms still see adding people as the safest response to growing complexity. Why is that no longer a sustainable operating model for asset servicing teams?
You cannot solve structural fragmentation simply by adding more people, particularly in a world where volumes are growing much faster than headcount.
The goal should not be to increase exception handling and manual checks. It should be to reduce the number of exceptions and checks required in the first place.
If the operating model depends on people reviewing large volumes of routine activity, firms are introducing inconsistency, delays, and additional risk into the process.
A sustainable model is one where systems handle volume and standard processing, allowing people to focus on exceptions, oversight, and the areas where judgement genuinely adds value.
That delivers greater scalability, stronger controls, and a better use of scarce operational expertise.
One of the challenges facing the industry is that much of that expertise is ageing out of the workforce. Attracting younger talent into specialist operational roles is becoming increasingly difficult, while transaction volumes continue to grow.
The industry cannot rely indefinitely on replacing expertise through headcount growth alone. That’s why better data, greater automation, and more integrated end-to-end operating models are becoming increasingly important.
If senior operations leaders accept that the current model is under strain, what does a practical path forward look like? Where should firms begin if they want to reduce risk without launching a full-scale transformation programme overnight?
The important thing is that firms start now, but they should not feel they need to fix everything at once.
Investment budgets are constrained, the cost of change is high, and transformation always carries some degree of risk. The most effective approach is often to begin with targeted, highvolume pain points where work?ows can be redesigned and automation increased.
By focusing on a manageable area, firms can demonstrate that improved data discipline and work?ow design deliver measurable operational benefits. They can then expand that success over time.
Where appropriate, firms may find greater value in embedding AI within the core process rather than layer on top of disconnected systems, with human oversight remaining in place where necessary.
Firms that take this approach will be in a much stronger position than those that continue relying on manual workarounds that simply mask underlying issues.
You have spoken about the importance of trusted data, clearer accountability, and scalable work?ows. What does ‘good’ look like in practice for firms trying to modernise their operating models today?
Good starts with standardised, well-governed, and high-quality data. It also requires integrated systems with fewer hand-offs and less fragmentation. That naturally reduces the need for duplicated checks and manual intervention.
The result is greater automation of routine processing, allowing experienced staff to focus on oversight and genuine exceptions. Our research suggests firms that invest in these areas can achieve returns on investment of between 12 -13 per cent.
Increasingly, modern operating models also need to incorporate AI within a human-supervised framework, supported by embedded dashboards, system integration, and end-to-end visibility.
The emergence of tokenised assets is also increasing pressure for more integrated operating models. Firms need infrastructure capable of supporting both traditional and tokenised assets side by side. Creating entirely separate work?ows for tokenised assets would simply introduce new complexity and inefficiency.
As volumes, complexity, and regulatory expectations continue to grow, how do you see the role of operations teams evolving over the next five years?
Throughout my career, operations teams have consistently been asked to do more with less, and that will continue.
The difference is that they should be spending far less time on repetitive validation and routine repair work.
Their role will become increasingly strategic, focused on oversight, risk management, governance, client service, and exception management.
AI will help reduce some avoidable exceptions and free people to focus on higher-value activities. It will be one of the key enablers that allows firms to scale without continuously increasing headcount.
Our research found that 57 per cent of firms identified trusted technology partners as an important enabler of this transition, particularly where they can support a future-state golden-copy data model.
If you had to summarise the industry challenge in one sentence, what would it be?
I think it might be quite a long sentence, but I guess asset servicing has reached a point where maintaining the status quo is no longer a neutral option.
Rising event volumes, shorter settlement cycles, growing regulatory expectations, AI adoption, and the emergence of tokenised assets are all colliding with legacy systems, fragmented data, and ageing operating models.
Without trusted data, integrated platforms, and more scalable operating frameworks, firms will struggle to keep pace with the demands being placed on them. That is the industry’s central challenge.
NO FEE, NO RISK
100% ON RETURNS If you invest in only one asset servicing news source this year, make sure it is your free subscription to Asset Servicing Times
100% ON RETURNS If you invest in only one asset servicing news source this year, make sure it is your free subscription to Asset Servicing Times
