Things your mother wants you to know about Payment Orchestration — Part 1
I’ve wanted to write this for a while as there’s a growing amount of noise about payment orchestration and as new terms & words come to market there’s always a lot of questions being asked about what it is? how does it work? is it right for me? why would my customer use it? Presently there isn’t too much content out there that caters to this so I’m having a crack. In here we’ll discuss what is payment orchestration, why 2020 was the year the term & concept became a thing and what are the reasons businesses look at payment orchestration.
From my perspective I’ve spent c.13 years in payments and have seen more and more businesses needing multiple payment & fraud providers and in 2020 I jumped into the payment orchestration world. Now, I’d love to say I am responsible for 2020 being the biggest year yet within payment orchestration but I think that would be misleading, in truth it was very much a coincidence ha.
So what is Payment Orchestration?
Payment Orchestration at its core is about businesses having one integration to access multiple payment providers & methods. Associated with this comes a single payment fields/page, a cross provider token vault, a cross provider 3DS engine, a single (& standardized) reconciliation api/file, a smart routing engine, a smart retry logic, and so on and so on. Mature/Large businesses had developed this infrastructure in house years ago, but it’s been rare for people to talk about it, it’s been rare for people to explain how their smart routing works, it’s been rare for people to talk about why they started orchestrating their payments and as the payment world has got more complex so have these systems and the code.
In the last 12 months businesses have begun to have a real choice; build/continue building in house or buy a Payment Orchestration platform from one of a number of companies.
I hope some people find this content useful, from my thousands of discussions with businesses, payment providers, payment methods, fraud providers, platforms and the schemes I welcome more debate! Disclaimer, I’m talking about companies who typically have online sales of >$150M
Why Orchestration couldn’t really happen until 2020
In my head (which is dangerous) I think it’s good to understand why buying Payment Orchestration couldn’t really happen 5–10 years ago. Here’s a few fundamental reasons:
- Infrastructure in the cloud — Without the cloud (AWS, GCP, Azure) there is no real concept of payment orchestration and I’d even argue without technology that allows you to be across multiple clouds at the same time there is still no real concept — wondering why, picture this, I’m at home setting up an orchestration business with my friends, putting my software on a CD for people to run on their premise. I get my first meeting and off I go walking into John Lewis (this is an example name, this meeting did not happen) and saying ‘on this CD exists the greatest payment infrastructure that could ever be built’. It’s pretty unlikely they will buy from me and not just build what they need themselves. Option 2; perhaps I rented some space in a data center in Cambridge, UK (my hometown) and off I go again to my meeting, this time with ASOS (again, this didn’t happen) say ‘hey, here is the best payment infrastructure ever built’ — but as I’m a start up the performance of my cloud would be very poor compared to ASOS and if it goes down I can’t afford a second DC so Asos.com you now go down too, and if miraculously I don’t go down I don’t have much budget for security so perhaps my systems gets compromised and ASOS and it’s customers are now a victim in a data breach, or the last scenario (equally unlikely) is that I’m an incredibly well funded start up who has managed to build multiple data centers around the world, with disaster recovery and great latency and security (weird that I did this but wasn’t part of the founding team of AWS, Azure or GCP) and on top of doing that I built the worlds best payment infrastructure…
- Trust — Touched on a little bit above, trust plays a huge role in deciding whether to buy payment orchestration software, rather than build in house. Trust in security, people, infrastructure, product roadmap, vision. Would you put your $400M in online sales through a company making less than $2M in revenue? That $1M in revenue doesn’t leave much room for spend on security, PCI DSS, infrastructure for uptime/processing time etc. For many companies this risk has been too high. The old saying of nobody gets fired for buying IBM might not be so true if the sentence is changed to ‘nobody get’s fired for putting 100% of their revenues through a company whose revenues are less than the spend of your work Christmas party’ (or perhaps a more applicable saying might be ‘who might not exist in 18 months’…)
- Payment providers were not truly global — Imagine being an online retailer in the US. It was pretty common to have one provider in North America and then another provider in Europe and another in Asia. (This is a very simple example, I’ve seen some with over 15 providers). This forced the business into some form of orchestration and as there was no one really they could buy from they were forced to build in house. They took on the task of managing multiple providers, they were deciding which traffic goes where, handling multiple reconciliation files/formats, multiple fraud tools etc. I don’t think many merchants wanted to handle this complexity, but the nature of the payment industry forced them to. It’s like buying a car but then finding out you have to be your own mechanic…all you wanted to do was drive the kids to school and do the shopping.
- ‘Smart’ payments didn’t exist until recently — A version of smart payments 10 years ago was “if a transaction fails with one provider send it to my back up provider to try again”. I’d argue not quite meeting the definition/expectation we have of ‘Smart’ nowadays. However, look at the top payment providers now. Smart exists everywhere and quite rightly businesses want to access the latest functionality. Businesses payment maturity/knowledge has greatly deepened over the last 10 years and they need to know who has got the smartest ‘smart’ tools.
- Many payment providers now have awesome functionality — Similar to the above, with investment in payment companies being so large it’s creating so much proprietary functionality and business are wanting to access the best of each provider. This is very similar to when I used to go to Woolworths (I miss this store) as a kid and wanted the best sweets in my bag. I was the one who decided what was best for me, and what would go in my bag and in what order. A pick n mix approach is deemed critical by most payment mature companies now.
- There was no one to buy payment orchestration software from — It’s still fairly recent that providers existed in this space and for the large enterprises there wasn’t a provider that was suitable for them. Many businesses will still believe this last comment is true and thus they take the decision to build in house (we’ll discuss this more later).
Why buying Payment Orchestration now makes sense:
- Developer costs in the last ten years have rocketed and payment engineers are in short supply. This is forcing many businesses to buy an orchestration platform because they can’t get the 10/12/20/24 engineers they need to do this themselves (want to see the efforts some companies go to when building their layer in house — I recommend reading up on Uber — it’s awesome)
- Businesses need agility — Whether the driver is Covid, competition or customer experience demands; it’s widely being stated that businesses with the most agility are going to be the winners. If you have to wait 14 months to get an integration to a local provider in Brazil is that agile? (if you say yes, …)
- Businesses are going through cloud transformations — The old programing languages and legacy monoliths these original in house payment orchestration code/logic was written on is being ripped out and thus it’s the perfect time to re-evaluate.
- Businesses are under more scrutiny and regulation — With regards to payment and data than ever before. This is forcing greater accountability and transparency with stronger processes, a change in payment page offerings, who handles data and scrutiny over this data. Nobody wants to be the employee in the company that insisted they still handled raw card data in 2021…
- Ecommerce volumes have sky-rocketed and the worlds got smaller. With that comes increased complexity. Selling to domestic US is very different to selling to 150 countries for instance. With complexity comes cost and if you can’t stay on top of the complexity you can be creating a large opportunity cost for your business. No c suite member wants to be the unpopular kid at school and be responsible for slowing the business down.
What are some of the specific drivers that ignite the discussion of buying a payment orchestration platform:
- Business Continuity guidelines mean it’s too risky to have one provider. This is one of the primary reasons I’ve seen. If your business makes c.100% of its revenues online and you only use one provider someone in your business will point out this is too risky and will likely be trying to persuade you to have additional providers (sometimes this is as well received by colleagues as the real estate agent you once gave your email address to and they wont give up). Some famous businesses (I found this a fun google search task which explains a little about me as a person) have revealed in their IPO prospectus they are at risk by relying on a single payment provider. Adding a second provider brings you the benefit of reduced risk but you are immediately thrown into the complexity of managing technically & commercially multiple providers. It doesn’t stop there, it also includes; failover strategy, fraud tools, token vaults, payment pages and thus before you know it you’re having to decide whether you write this code or buy this code.
- Geo expansion is another very common reason, particularly for ones who have built in house. Most commonly I see companies who already have multiple providers getting fed up with the complexity of adding many more direct integrations (I’ve seen this trend accelerate with brands now targeting Asia as their revenue driver given Covid. To target Asia well you need local payment methods and the thought of adding 7 (e.g.) new payment integrations is a scary thought for anyone’s roadmap). The wait on the product roadmap, the cost/effort of maintaining the connections, the pressure from colleagues about not moving fast enough, …., the list of downsides continue. Leveraging a payment orchestration platform and doing a single integration to access your new/additional markets is very common.
- Optimization of revenue is the top criteria for the most payment mature businesses. Testing providers for authorization or cost performance, real time analytics, modelling of performance, contracts tied to KPI’s. Often this is a multi-year dream for most companies but for an increasing few this is the requirement today. Quite often they are looking externally because they are concerned that either 1) their internal roadmap isn’t allowing them to keep up with their payment demands or 2) they are scanning the market to see the latest innovations and opportunities. (In a future post i’ll talk about what I think a ‘payment mature business’ is)
- Internal consolidation for some businesses is the critical driver for buying a payment orchestration product. For example, imagine being a company with multiple business units, all with their own payment set ups, infrastructure, teams, you’re buying more companies which further adds complexities and your finance function is seeing costs go up and up. At some point it’s understandable that someone questions if there is another way to do this and consolidate efforts.
Why some businesses make the right decision and don’t buy a payment orchestration platform now?
There’s definitely a number of legit reasons why some companies don’t buy a payment orchestration platform today. I’d love to write up about this topic and many more. I hope this content helps, I hope it creates debate and I hope it’s helped people learn a little bit.
Thank you for reading! and welcome any comments, any feedback on the writing and of course I’d love to hear your viewpoint on things!
