The Acquirer Divergence: Why Your Authorisation Rate is a Product, Not Plumbing
- Apr 8
- 6 min read
Updated: Apr 10
For years, the payment industry has been obsessed with one thing: cost. When a merchant looks for a new payment provider, the first question is usually, "What’s your take rate?" It’s a race to the bottom where everyone is trying to shave a few basis points off the transaction fee.
In this world, payments are treated like plumbing. You want the pipes to be invisible, the water to flow, and the bill to be as low as possible. But here’s the problem: if your plumbing is cheap but leaks 10% of your water before it reaches the tap, it’s not actually cheap, it’s expensive.
At RivaTech Consulting, we’re seeing a massive shift in how the world’s most successful fintechs and enterprise merchants view their payment stack. They’ve stopped looking at authorisation rates as a technical metric and started viewing them as a core product feature.
We call this The Acquirer Divergence. It’s the growing gap between the "commodity" acquirers and the "product-led" acquirers. And if you’re on the wrong side of that gap, you’re likely leaving millions of dollars on the table.
The 5-12% Performance Gap
The data is startling. When you look at comparable transaction profiles, same industry, same geography, same ticket size, there is often a 5% to 12% difference in authorisation rates between top-tier acquirers and the rest of the pack.
Think about that for a second. If you’re a merchant processing $100 million a year, a 5% gap is $5 million in lost revenue. That isn't just a "cost of doing business." It’s a catastrophic leak in your conversion funnel.
For a long time, merchants assumed that if a transaction failed, it was the customer’s fault (insufficient funds) or the bank's fault (random decline). While that’s sometimes true, we now know that a huge chunk of those "declines" are actually preventable. They are the result of poor routing, bad data, or lacklustre retry strategies.
The Shift: From Plumbing to Product
When you treat payments as plumbing, you focus on the "pipe", the connection between the merchant and the card scheme. You want it to be stable and cheap.
When you treat payments as a product, you focus on the "output", the successful authorisation. A product-led approach means you are constantly iterating, experimenting, and optimising to ensure that every legitimate customer who wants to pay actually can.
Top-tier players like Adyen and Stripe haven't built their empires just by being "easy to integrate." They’ve won because they treat the authorisation rate as a competitive advantage. They’ve realised that an EBITDA margin is built by processing transactions better, not just cheaper.
So, how do the leaders do it? It comes down to four key levers.
Lever 1: Smart Routing (Local vs. Cross-Border)
The quickest way to tank your authorisation rate is to send a transaction across a border unnecessarily. If an Australian customer tries to buy from a merchant whose acquirer is based in the US, the issuing bank in Australia sees a "cross-border" flag. To the bank, this looks like high-risk activity, and they are much more likely to hit the "decline" button.
A product-led acquirer uses local acquiring whenever possible. By having local entities and direct connections to domestic schemes in dozens of countries, they make a global transaction look like a local one to the issuing bank.
We’ve seen cases where a marketplace improved its authorisation rate by over 13% simply by switching from cross-border processing to local acquiring in markets like Brazil or Mexico. This isn't just about avoiding fees; it's about building trust with the issuer.
Lever 2: Smart Retry Logic
A decline isn't always the end of the road. There is a whole "taxonomy of declines" that most merchants never see. Some are "hard" declines (e.g., "stolen card"), which you should never retry. But many are "soft" declines (e.g., "technical error" or "temporary insufficient funds").
Generic acquirers treat all declines the same. Product-led acquirers use Smart Retry Logic. This involves:
Timing: If a subscription payment fails at 2:00 AM on a Tuesday, retrying it at 9:00 AM on a Friday (payday) might yield a success.
Parameter Adjustments: Sometimes retrying the transaction without the CVV or via a different network path can bypass a temporary glitch in the issuer’s system.
Machine Learning: The best systems learn which issuers respond best to specific retry patterns.
This can easily add 2-3 percentage points back to your bottom line without the customer ever knowing there was a hiccup.
Lever 3: Signal Enrichment (Giving Issuers What They Want)
Issuing banks are terrified of fraud. When they receive an authorisation request, they have milliseconds to decide if it’s legitimate. If the data they receive is sparse, they default to "no" to protect themselves.
Most legacy payment messages use the ISO 8583 standard, which was designed in the 1980s. It’s the digital equivalent of a postcard: there’s not much room for detail. Product-led acquirers use Signal Enrichment to pack that message with as much high-quality data as possible.
This includes:
Enhanced 3DS 2.0 Data: Sharing over 150 data elements (device ID, IP address, shipping speed) with the issuer.
Fraud Scores: Telling the issuer, "We’ve vetted this customer, and our fraud engine says they are 99% safe."
Optimised Formatting: Issuers are picky. Some prefer address data in a specific format; others want the merchant descriptor to look a certain way. Small tweaks in how the data is presented can lead to higher approval probabilities.
Lever 4: Tokenization & Message Experimentation
Network tokens (like those provided by Visa and Mastercard) are a game-changer. Unlike a standard "vault" token, a network token is kept up to date by the card schemes. If a customer’s physical card expires and they get a new one, the network token stays valid.
Using network tokens doesn't just reduce "card expired" declines; it also signals to the issuer that the transaction is secure, often leading to lower interchange fees and higher auth rates.
Beyond tokens, the leaders are constantly experimenting with message formatting. They treat the payment message like a marketing A/B test. Does this issuer prefer the merchant’s phone number in the descriptor? Does that issuer prefer a specific transaction category code? These tiny iterations compound over millions of transactions.
The Compounding Effect
The real magic happens when these features talk to each other. In a "plumbing" setup, your fraud tool, your gateway, and your acquirer are often three different companies that don't share data.
In a "product" setup, the data flows freely. The fraud engine informs the routing engine, which informs the retry logic, which is powered by enriched signals. This integrated loop creates a compounding effect where the system gets smarter with every transaction.
As we discussed in our post on Stripe vs Adyen, the "full-stack" approach is winning because it eliminates the data silos that kill authorisation rates.
The Stick: Industry Pressures (VAMP and TPE)
If the "carrot" of increased revenue isn't enough to make you care about auth rates, the "stick" definitely will.
Visa and Mastercard have introduced programmes like Visa VAMP (Visa Acquirer Monitoring Program) and Mastercard’s Transaction Processing Excellence (TPE). These programmes penalise acquirers: and by extension, merchants: who have excessive decline rates or poor retry practices.
If you keep retrying a "hard" decline, you won't just fail to get the money; you’ll actually be fined for "excessive retries." Compliance is no longer just about staying out of trouble; it’s now a performance metric that directly impacts your costs.
What Should You Do Now?
If you’re a fintech or an enterprise merchant, it’s time to stop looking at your payment provider as a utility. Here is your roadmap:
Benchmark Your Rates: You can’t fix what you don’t measure. Do you know your authorisation rate by country? By issuer? By card type?
Audit Your Declines: Ask your provider for a breakdown of decline codes. If more than 10% of your declines are "Generic" or "Do Not Honour," you have a data enrichment problem.
Evaluate Your "Product": Is your acquirer providing you with tools to improve these rates (like network tokens and smart retries), or are they just giving you a pipe?
Consider Orchestration: If you’re large enough, you might need a payment orchestration layer to dynamically route transactions to the best-performing acquirer for a specific region.
The "Acquirer Divergence" is only going to get wider. As AI and machine learning become more integrated into the payment flow, the gap between the leaders and the laggards will become an unbridgeable chasm.
At RivaTech Consulting, we help businesses navigate these complexities. Whether you’re looking to recover lost revenue or rethink your entire payment stack, remember: your authorisation rate isn't just plumbing. It’s the most important product you’ll ever manage.