Build or buy?
29 March 2016
The in-house versus vendor debate polarises opinion. Consolo senior partner Richard Colvill speaks to four experts to see where the answer might lie
Image: Shutterstock
You鈥檝e concluded that your legacy platform is creaking at the seams, no longer fit for purpose, un-scalable and rapidly approaching the end of its life. Or, you鈥檙e a new business start-up and the world鈥檚 your oyster. In both cases, the million dollar question, or as is in many cases, much, much more, when deciding on a new system to support your business, is, do you build in-house or purchase from an established vendor. This is a question that polarises opinion.
This is likely to be a conundrum discussed at great length by every establishment in the financial industry, and one that quite often spans a period of many weeks, months and years. Unless you鈥檝e experienced it for yourself, you probably have no appreciation of how big a challenge this is. But those that have will testify that you would only want to do it once in your career. Unless, of course, you鈥檙e one of those change management or project management types, because they love it.
This is a task that is exacerbated in the larger firms where the number of stakeholders with differing opinions is far greater than in their smaller counterparts. The emails, the think-groups, the breakout area, manager鈥檚 office, clandestine chats over the coffee machine鈥攊t goes on and on.
While this decision ultimately rests with the board of directors and executive management, the key influencers reporting to these levels can be many in number and often with differing opinions and agendas. Trading, client, marketing, product, strategy, operations and IT managers will all have different objectives, such as increased volume, automation, profitability, scalability, job security, shareholder value, cost, support, development and flexibility, to name a few.
And so the process begins: whitepapers, proposals, business cases, feasibility and affordability studies, return-on-investment analysis, due-diligence and market research. The amount of paper seems endless, and it must all be juggled around your day-job (thank you very much).
Eventually, judgement day arrives and hurrah, a decision has been made. The lucky (or more influential) ones get their desired result while the others suffer an unbearable loss of professional pride. The victors soon lose the sweet smell of success from their nostrils, when they see how insufferable the 鈥榥o鈥� voters become for the duration of the rest of the project. Still it was nice while it lasted.
So, of in-house and vendor, which option is the best option? There are many factors that underpin the question, including: the speed at which it needs to be delivered; the cost; the risk; the knowledge, resources, capacity and infrastructure to support it; the up-tick; and people.
Sunil Daswani, senior vice president and head of international securities lending at Northern Trust Global Investors, provides his views on the decision to build in-house.
Global 色花堂Lending (GSL), Northern Trust鈥檚 securities lending proprietary trading system, is a single global 24/7 platform. It is a multi-function system for managing the securities lending business end to end.
GSL has electronic links in place to depositories, other securities lending vendors, custody systems, and EquiLend. The electronic links we have in place with borrowers, both direct and through EquiLend, automate the majority of loans. On average, 90 percent of our loan volume is automated without manual human intervention.
We also use similar automation via GSL for our post-trade reconcilement and mark-to-markets to be more efficient and maintain strong risk management. GSL was implemented in 2003 and we routinely enhance the system to meet new business requirements.
Having an in-house system means we can normally bring new variations of the product, set up lending in new markets and build customised solutions for clients in a manner that can transfer to be scalable in the future, should there be demand.
The decision we made to go in-house has certainly paid off given the large amount of new business we have brought in where we have capabilities that some of our peers may not have. Additionally, our peers may experience delays to-market for new or enhanced product offerings where off-the-shelf packages are used, which in our view gives us a major strategic advantage.
So a global custody agent lender leaves us in no doubt why the decision to build internally was the right one.
Bill Foley, director at securities finance specialist Foley O鈥橬eill, and, with a wealth of knowledge gained at a number of leading market participants, shares his thoughts on past experiences of both developing in-house and outsourcing to a vendor.
Having been through the process of implementing both off-the-shelf and in-house built securities finance systems, I have seen the merits and pitfalls of both. Plug-and-play solutions can sometimes offer exactly the speed and ease that the name suggests, but it can also sometimes seem more like 鈥榩lug-and-pray鈥� once the implementation process begins.
Similarly in-house builds should, in theory, see IT departments able to use their knowledge of the business and existing systems to build something bespoke in a cost-efficient and timely manner, and many are able to achieve this. However, without a full and honest appraisal of internal resources, skills and required connectivity to other internal platforms, in-house builds can become fraught affairs, often with much compromise, delay and additional costs incurred along the way, as well as running the risk of the project being re-prioritised, which rarely sees it moved up the list of priorities.
The decision to choose one route over the other will depend on a number of factors that will vary from business to business, so it is impossible to give a definitive view. However, the most important element to arriving at the right conclusion is to ensure that all stakeholders are involved from the outset. This includes both internal and external contributors.
Internally, the business case must be fully understood and supported by all and the entire process must be agreed by all those involved at the outset.
The entire process can often be managed better by utilising the right external resources, whose aim is solely to deliver the right solution and can focus resources completely on the delivery without being sidetracked by their day job. The most successful system implementation that I can recall from my own experience utilised exactly this approach.
Some thought provoking words of wisdom from an established industry veteran, who found that, as far as management of the job is concerned, external resources can be beneficial.
Tom Dibble, head of product management for securities finance at FIS Global, provides a view on why he believes his target audience prefers, and will benefit from, FIS鈥檚 solution and service offerings.
FIS has seen a significant upward trend as existing and prospective clients move away from in-house development of products and instead move to support services that enable these organisations to focus their resources on critical core business. As a result, there is growing demand for a comprehensive range of services and solutions that FIS offers across the securities finance landscape, all targeted at reducing centralised risk and lowering total cost of ownership.
FIS鈥檚 established securities finance solution set, such as Global One and Apex 色花堂Finance, is delivered via a framework of fully managed and domain specialist services. These services encompass the front-to-back business and technical requirements for our partner clients, where the responsibility for the solution environment remains with FIS across a strategic and long-standing service delivery model.
As we use our solution framework to create specific client delivery models, options range across our expertise, from our own data centres, through supporting mechanisms, change project delivery and product knowledge. This multi-strand approach that builds on our years of leading edge expertise in these areas enables us to talk across delivery boundaries and fill the gaps that other products, both in-house and solution providers, can sometimes have.
FIS is a trusted global financial technology partner to more than 20,000 clients across 130 countries. Our intention is always to be a willing and capable partner in helping our customers drive to success and we believe our wide range of services, from product to delivery, across the whole relationship with our customers, is vital and vibrant as we look to the future and empower the financial world.
A compelling argument from a vendor with a long-standing history in securities finance, and which many in the industry would view fondly as the first mainstream systems solution provider with Global One.
Paul Wilson, product director for global sales at 4sight Financial Software, gives his views on the debate.
It鈥檚 an interesting decision. In fact, the largest competitor to 4sight, and I鈥檓 sure to most of our peers in the vendor space, is a combination of doing nothing and building in-house.
Where 4sight does win a mandate to install a securities finance system, it鈥檚 always to replace an existing in-house system or an older vendor system that has become, on the whole, so ingrained and bespoke that it鈥檚 effectively in-house. The reasons why an institution might choose to do this is interesting. If I look at the most recent deals that 4sight has won, it鈥檚 a combination of factors. In order of priority:
Fit for purpose (both functionally and architecturally), and the business really wants it;
Functionality uplift;
Operational cost and risk reduction;
Cost effectiveness; and
Magic dust.
Ultimately, no-one in their right mind is going to buy the wrong, expensive, ill-fitting spanner for a job that, when applied to a stubborn nut, skins their knuckles painfully every time.
Interestingly, though, many people actually make the decision to hand-forge, finish and deploy their own bespoke spanners from scratch. The problem is that when the standards change, you end up re-engineering it all from scratch each time, and the spanner often doesn鈥檛 come out quite right, requiring considerable reengineering.
Alternatively, you can buy an off-the-shelf toolkit with a lifetime guarantee, catering for all standards, and with a socket set, ratcheting spanners and free mug all included.
Why would people do anything but the latter?
In my experience, functionality is the key deciding factor, and in fact passing the 鈥榝it for purpose鈥� test is our clients鈥� primary decision. Which means that the tool (in this case 4sight) is fit for the job both functionally and architecturally. Put simply, they wouldn鈥檛 buy it if it wasn鈥檛. Those interested in the hand-forging route truly believe they can build something better than a vendor. This is true for some institutions, but our experience is that this is not a widespread truth and can be a very expensive function, particularly in our ever-changing regulatory landscape.
This decision is often made first by the business owners and their IT owners. Business also tend to see functionality uplift, for example, an improved ability to manage extendables, monitor balance sheet impact, process optional dividends, or achieve full straight-through processing for SWIFT. Interestingly, if the decision is front-office/business-led, a future vision of the 4sight roadmap and the ability to optimise trading decisions is often the key unique selling point, whereas if the decision is more operational or technical, the decision gets much more low level and delves into the architecture of the solution in terms of cost and risk reduction.
Obviously, cost is key. Nothing in life is free, and the old adage of pick two out of good, fast and cheap is paramount. Once procurement are involved, it鈥檚 a fairly standard negotiating process with the client. The hard part of this is having clients understand that it is difficult to predict the impact of large institutions鈥� reaction speed, bureaucracy, skillsets and access to actual users on project cost. However, on paper, it鈥檚 fairly cut and dried and a lot cheaper than many institutions might think.
There鈥檚 a finite set of fairly standard migration datasets, up and downstream integration points, extracts and reports to integrate, plus some inevitable gap development, so it is possible for 4sight to provide capped, fixed-price and surprisingly cost-effective high-quality projects. Factor in dozens of existing clients all feeding into the system and 4sight鈥檚 clear focus on securities finance, and the future costs are also attractive because, if your vendor is proactive, their system is always on-track with where the market is heading. And it鈥檚 heading into waters we have never imagined.
One last thing: the magic dust is whether the client considers that the vendor understands them, or that we 鈥榞et鈥� the crux of their problem and are convincing, straight up and compelling both now and as a long-term partner.
This is the 鈥榬elationship factor鈥� and it鈥檚 a truth that 4sight ultimately wins deals based on this final, hard-to-quantify consideration.
Having been involved in projects supporting both decisions on several occasions, Consolo knows that the process and challenges from a business or change perspective are practically identical.
The bottom line is that somebody, somewhere, at the top of the decision tree and holding the purse, will draw a line in the sand and demand delivery by a predetermined date and within budget. There鈥檚 always a little bit of wiggle-room but that good grace is limited and, in the end, the project will be made to deliver all or most of the requirements on or before that date.
I say 鈥榤ost鈥� because there is a constant with both vendor and self-build installations. Something is always decoupled from the main delivery and re-phased for a later installment, a victim of the budget and time constraint demands. What was once a 鈥榤ust-have鈥� becomes a 鈥榥ice-to-have鈥� and the term 鈥榳orkaround鈥� is attributed.
For the end-user, this workaround can be an eternal source of frustration, as popular belief is that it is something that all discerning systems should offer at entry level, yet the truth is it was dropped to meet a deadline, or assumed but never explicitly requested. These workarounds are now thrown into what us change types call 鈥榬etrofit鈥� and they are graded according to the business criticality versus development time-lapsed and cost equation versus the time it takes to perform the workaround. Only when this exercise is complete can the vendor or your internal developers prioritise these, subject to the allocation of further resources and budget.
Frustratingly, users will find that these retrofits are now directly contesting with new business and product development inventory for the ever-diminishing budget and resource pool, so some of these may not see the light of day for a very long time, if at all. However, if resourced correctly from the start, and with the business鈥檚 best interests at heart, these issues can be avoided, ensuring that not only will users get the system that they want, but the system that they need.
This is likely to be a conundrum discussed at great length by every establishment in the financial industry, and one that quite often spans a period of many weeks, months and years. Unless you鈥檝e experienced it for yourself, you probably have no appreciation of how big a challenge this is. But those that have will testify that you would only want to do it once in your career. Unless, of course, you鈥檙e one of those change management or project management types, because they love it.
This is a task that is exacerbated in the larger firms where the number of stakeholders with differing opinions is far greater than in their smaller counterparts. The emails, the think-groups, the breakout area, manager鈥檚 office, clandestine chats over the coffee machine鈥攊t goes on and on.
While this decision ultimately rests with the board of directors and executive management, the key influencers reporting to these levels can be many in number and often with differing opinions and agendas. Trading, client, marketing, product, strategy, operations and IT managers will all have different objectives, such as increased volume, automation, profitability, scalability, job security, shareholder value, cost, support, development and flexibility, to name a few.
And so the process begins: whitepapers, proposals, business cases, feasibility and affordability studies, return-on-investment analysis, due-diligence and market research. The amount of paper seems endless, and it must all be juggled around your day-job (thank you very much).
Eventually, judgement day arrives and hurrah, a decision has been made. The lucky (or more influential) ones get their desired result while the others suffer an unbearable loss of professional pride. The victors soon lose the sweet smell of success from their nostrils, when they see how insufferable the 鈥榥o鈥� voters become for the duration of the rest of the project. Still it was nice while it lasted.
So, of in-house and vendor, which option is the best option? There are many factors that underpin the question, including: the speed at which it needs to be delivered; the cost; the risk; the knowledge, resources, capacity and infrastructure to support it; the up-tick; and people.
Sunil Daswani, senior vice president and head of international securities lending at Northern Trust Global Investors, provides his views on the decision to build in-house.
Global 色花堂Lending (GSL), Northern Trust鈥檚 securities lending proprietary trading system, is a single global 24/7 platform. It is a multi-function system for managing the securities lending business end to end.
GSL has electronic links in place to depositories, other securities lending vendors, custody systems, and EquiLend. The electronic links we have in place with borrowers, both direct and through EquiLend, automate the majority of loans. On average, 90 percent of our loan volume is automated without manual human intervention.
We also use similar automation via GSL for our post-trade reconcilement and mark-to-markets to be more efficient and maintain strong risk management. GSL was implemented in 2003 and we routinely enhance the system to meet new business requirements.
Having an in-house system means we can normally bring new variations of the product, set up lending in new markets and build customised solutions for clients in a manner that can transfer to be scalable in the future, should there be demand.
The decision we made to go in-house has certainly paid off given the large amount of new business we have brought in where we have capabilities that some of our peers may not have. Additionally, our peers may experience delays to-market for new or enhanced product offerings where off-the-shelf packages are used, which in our view gives us a major strategic advantage.
So a global custody agent lender leaves us in no doubt why the decision to build internally was the right one.
Bill Foley, director at securities finance specialist Foley O鈥橬eill, and, with a wealth of knowledge gained at a number of leading market participants, shares his thoughts on past experiences of both developing in-house and outsourcing to a vendor.
Having been through the process of implementing both off-the-shelf and in-house built securities finance systems, I have seen the merits and pitfalls of both. Plug-and-play solutions can sometimes offer exactly the speed and ease that the name suggests, but it can also sometimes seem more like 鈥榩lug-and-pray鈥� once the implementation process begins.
Similarly in-house builds should, in theory, see IT departments able to use their knowledge of the business and existing systems to build something bespoke in a cost-efficient and timely manner, and many are able to achieve this. However, without a full and honest appraisal of internal resources, skills and required connectivity to other internal platforms, in-house builds can become fraught affairs, often with much compromise, delay and additional costs incurred along the way, as well as running the risk of the project being re-prioritised, which rarely sees it moved up the list of priorities.
The decision to choose one route over the other will depend on a number of factors that will vary from business to business, so it is impossible to give a definitive view. However, the most important element to arriving at the right conclusion is to ensure that all stakeholders are involved from the outset. This includes both internal and external contributors.
Internally, the business case must be fully understood and supported by all and the entire process must be agreed by all those involved at the outset.
The entire process can often be managed better by utilising the right external resources, whose aim is solely to deliver the right solution and can focus resources completely on the delivery without being sidetracked by their day job. The most successful system implementation that I can recall from my own experience utilised exactly this approach.
Some thought provoking words of wisdom from an established industry veteran, who found that, as far as management of the job is concerned, external resources can be beneficial.
Tom Dibble, head of product management for securities finance at FIS Global, provides a view on why he believes his target audience prefers, and will benefit from, FIS鈥檚 solution and service offerings.
FIS has seen a significant upward trend as existing and prospective clients move away from in-house development of products and instead move to support services that enable these organisations to focus their resources on critical core business. As a result, there is growing demand for a comprehensive range of services and solutions that FIS offers across the securities finance landscape, all targeted at reducing centralised risk and lowering total cost of ownership.
FIS鈥檚 established securities finance solution set, such as Global One and Apex 色花堂Finance, is delivered via a framework of fully managed and domain specialist services. These services encompass the front-to-back business and technical requirements for our partner clients, where the responsibility for the solution environment remains with FIS across a strategic and long-standing service delivery model.
As we use our solution framework to create specific client delivery models, options range across our expertise, from our own data centres, through supporting mechanisms, change project delivery and product knowledge. This multi-strand approach that builds on our years of leading edge expertise in these areas enables us to talk across delivery boundaries and fill the gaps that other products, both in-house and solution providers, can sometimes have.
FIS is a trusted global financial technology partner to more than 20,000 clients across 130 countries. Our intention is always to be a willing and capable partner in helping our customers drive to success and we believe our wide range of services, from product to delivery, across the whole relationship with our customers, is vital and vibrant as we look to the future and empower the financial world.
A compelling argument from a vendor with a long-standing history in securities finance, and which many in the industry would view fondly as the first mainstream systems solution provider with Global One.
Paul Wilson, product director for global sales at 4sight Financial Software, gives his views on the debate.
It鈥檚 an interesting decision. In fact, the largest competitor to 4sight, and I鈥檓 sure to most of our peers in the vendor space, is a combination of doing nothing and building in-house.
Where 4sight does win a mandate to install a securities finance system, it鈥檚 always to replace an existing in-house system or an older vendor system that has become, on the whole, so ingrained and bespoke that it鈥檚 effectively in-house. The reasons why an institution might choose to do this is interesting. If I look at the most recent deals that 4sight has won, it鈥檚 a combination of factors. In order of priority:
Fit for purpose (both functionally and architecturally), and the business really wants it;
Functionality uplift;
Operational cost and risk reduction;
Cost effectiveness; and
Magic dust.
Ultimately, no-one in their right mind is going to buy the wrong, expensive, ill-fitting spanner for a job that, when applied to a stubborn nut, skins their knuckles painfully every time.
Interestingly, though, many people actually make the decision to hand-forge, finish and deploy their own bespoke spanners from scratch. The problem is that when the standards change, you end up re-engineering it all from scratch each time, and the spanner often doesn鈥檛 come out quite right, requiring considerable reengineering.
Alternatively, you can buy an off-the-shelf toolkit with a lifetime guarantee, catering for all standards, and with a socket set, ratcheting spanners and free mug all included.
Why would people do anything but the latter?
In my experience, functionality is the key deciding factor, and in fact passing the 鈥榝it for purpose鈥� test is our clients鈥� primary decision. Which means that the tool (in this case 4sight) is fit for the job both functionally and architecturally. Put simply, they wouldn鈥檛 buy it if it wasn鈥檛. Those interested in the hand-forging route truly believe they can build something better than a vendor. This is true for some institutions, but our experience is that this is not a widespread truth and can be a very expensive function, particularly in our ever-changing regulatory landscape.
This decision is often made first by the business owners and their IT owners. Business also tend to see functionality uplift, for example, an improved ability to manage extendables, monitor balance sheet impact, process optional dividends, or achieve full straight-through processing for SWIFT. Interestingly, if the decision is front-office/business-led, a future vision of the 4sight roadmap and the ability to optimise trading decisions is often the key unique selling point, whereas if the decision is more operational or technical, the decision gets much more low level and delves into the architecture of the solution in terms of cost and risk reduction.
Obviously, cost is key. Nothing in life is free, and the old adage of pick two out of good, fast and cheap is paramount. Once procurement are involved, it鈥檚 a fairly standard negotiating process with the client. The hard part of this is having clients understand that it is difficult to predict the impact of large institutions鈥� reaction speed, bureaucracy, skillsets and access to actual users on project cost. However, on paper, it鈥檚 fairly cut and dried and a lot cheaper than many institutions might think.
There鈥檚 a finite set of fairly standard migration datasets, up and downstream integration points, extracts and reports to integrate, plus some inevitable gap development, so it is possible for 4sight to provide capped, fixed-price and surprisingly cost-effective high-quality projects. Factor in dozens of existing clients all feeding into the system and 4sight鈥檚 clear focus on securities finance, and the future costs are also attractive because, if your vendor is proactive, their system is always on-track with where the market is heading. And it鈥檚 heading into waters we have never imagined.
One last thing: the magic dust is whether the client considers that the vendor understands them, or that we 鈥榞et鈥� the crux of their problem and are convincing, straight up and compelling both now and as a long-term partner.
This is the 鈥榬elationship factor鈥� and it鈥檚 a truth that 4sight ultimately wins deals based on this final, hard-to-quantify consideration.
Having been involved in projects supporting both decisions on several occasions, Consolo knows that the process and challenges from a business or change perspective are practically identical.
The bottom line is that somebody, somewhere, at the top of the decision tree and holding the purse, will draw a line in the sand and demand delivery by a predetermined date and within budget. There鈥檚 always a little bit of wiggle-room but that good grace is limited and, in the end, the project will be made to deliver all or most of the requirements on or before that date.
I say 鈥榤ost鈥� because there is a constant with both vendor and self-build installations. Something is always decoupled from the main delivery and re-phased for a later installment, a victim of the budget and time constraint demands. What was once a 鈥榤ust-have鈥� becomes a 鈥榥ice-to-have鈥� and the term 鈥榳orkaround鈥� is attributed.
For the end-user, this workaround can be an eternal source of frustration, as popular belief is that it is something that all discerning systems should offer at entry level, yet the truth is it was dropped to meet a deadline, or assumed but never explicitly requested. These workarounds are now thrown into what us change types call 鈥榬etrofit鈥� and they are graded according to the business criticality versus development time-lapsed and cost equation versus the time it takes to perform the workaround. Only when this exercise is complete can the vendor or your internal developers prioritise these, subject to the allocation of further resources and budget.
Frustratingly, users will find that these retrofits are now directly contesting with new business and product development inventory for the ever-diminishing budget and resource pool, so some of these may not see the light of day for a very long time, if at all. However, if resourced correctly from the start, and with the business鈥檚 best interests at heart, these issues can be avoided, ensuring that not only will users get the system that they want, but the system that they need.
NO FEE, NO RISK
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to 色花堂Finance Times
100% ON RETURNS If you invest in only one securities finance news source this year, make sure it is your free subscription to 色花堂Finance Times
