Why not integrate Payment Methods directly?
Recently someone asked generally why a merchant should work with a payment service provider (PSP), when they could integrate into a payment method directly e.g., integrate something like a Gcash (Philippines digital wallet) or Grabpay (SEA digital wallet) directly rather than through a Payment Service Provider (PSP).
My instinctive reaction was “Nooooo!! Why would you want to do so much work?)”. But I realized I should examine this on both sides before going “Noooo”! In this post I will:
- Share the three options merchants have
- Discuss the case for and against direct integration to a payment method
- Propose some considerations for merchants
1/ Merchants have three main options
To accept payments, especially local payment methods (putting cards aside here) Merchants can choose between these three partnership and integration models:
- Payment Aggregators
- Payment Service Providers
- Direct to payment methods (the focus of this post)
See diagram below for how the partnerships model relate to one another.
2a/ The case for direct integration to a payment method
Summary: Potentially lower costs and own a direct relationship.
Lower your costs
Every third party layer between you and the payment method is additional margin you give to each layer for their services.
Direct integration and relationship
- Direct integration: Make direct technical changes since you are building closest to the metal.
- Direct relationship: Own the relationship directly to negotiate the best terms possible.
2b/ The case against a direct integration to a payment method
Summary: Creates significant upfront work and scaling challenges.
High investment costs (engineering)
Are you trying to be a payment provider? A single payment method integration is fine, but if you plan to accept ten payment methods, then the work repeats 10x times. And if you want to scale the work, you have to create a platform that covers:
- Payment in flows (and out flows)
- Reporting and reconciliation
- Operations: refunds, settlement, transaction errors, ledger
- Monitoring reliability, security
- There’s more…the Pragmatic Engineer does a deeper dive into this. The point is — you have to build a platform if you want to integrate multiple partners. You’ll have to think carefully through the platform design. Ultimately, do you want to be a payments provider (Stripe, Braintree), or do you want to focus on your core business?
High investment costs (partnerships)
To enable payment methods — or any banking/payment partners, you will need someone to handle Partnerships. Then they will have to find the partner, negotiate a deal, onboard, and then manage that partner long term. Imagine doing it once, and then doing it 10x. Ideally you want each partnership to unlock more than one payment method and scale the efforts.
3/ Merchant considerations
- Payment method companies might not want to directly engage merchants. There are payment methods that want merchants to work with them via a third party payments partner (for larger distribution), vs the payment method company doing direct merchant acquisition. For example, to accept Wechatpay, merchants must go through a third party acquirer (e.g., Stripe, Adyen)
- Payment method companies might only want to engage directly with merchants of a large size. If you’re a Google or Amazon— you’ll probably one of the few ones who can do this. Even if there’s a commercial engagement, payment method providers may prefer to avoid the burdens of maintaining custom technical integrations.
Looking back: The power of global payment networks like Visa, Mastercard, Amex, UnionPay, is that they’ve given a select pool of PSPs access to their global rails, and then those PSPs developed payment platforms for merchants integrate into. This enables networks to distribute their payment rails more widely.
Looking forward: Newer payment methods— mobile wallets (the Gcash, Gopay, Grabpay — amongst many others) have at times allowed merchants to directly integrate them. For wider distribution and adoption of their payment method, it makes more sense for them to focus on mass distribution through PSP route.
Looking inwards: For a merchant evaluating the option of directly integrating a payment method, the core question is — do I have the people (engineering, partnerships, operations) to integrate, execute and maintain individual payment method integrations? What is the return on this investment and is it significant enough? Have I reviewed the PSP landscape to see who might be able to do the work better — scale and speed wise? Ask these questions before leaping into direct payment method integrations.