Reducing Involuntary Churn

How Can I Reduce Involuntary Churn? This is one of the most important questions every business owner wants answered. When an attempt to charge your client results in a decline, that’s involuntary churn. If you’re a subscription or recurring business, they’re the most frustrating customers to lose – the ones that still want your service, but whose money you can’t receive. Your client wants the service, and you want to serve them, but there is some sort of technical block that is getting in the way.

In this article, we’ll share advice and tips to help reduce involuntary churn.

Payment Processor

The first place to look is at your payment processor; you want to make sure that there isn’t something systemically broken. When a transaction is declined, your payment gateway returns a response code (sometimes called a decline code) that explains what happened with the transaction. You may be surprised to discover several issues can be resolved by reconfiguring your CRM, sending more data, or making an adjustment to your payment gateway.

Common Response Codes:

Here are some common examples of response codes that you might be able to address, and what you can do about them:

  • You need to refer to the Card Issuer
    • Either call the issuing bank for manual authorization or retry at a better day/time.
  • Invalid card: Luhn algorithm failed (MOD – 10)
    • This isn’t a valid card number. You should have a check on your side to know this (Luhn algorithm) so that you aren’t sending fake credit card numbers to your gateway.
  • Transaction cannot be completed
    • A common, and unhelpful response code. There are many reasons you may get this, but it is most commonly resolved by resending the transaction at a better day/time.
  • Transaction required data are missing or invalid
    • Varying by gateway and merchant account provider, you may need to include some fields, like Address2, EmailAddress and/or others.
  • Gateway declined – CVV is missing or incorrect
    • If this is a recurring transaction, you need to set the recurring flag on the initial transaction (where you must include the CVV) so that subsequent charges don’t require it.
  • Transaction limit for merchant exceeded
    • Your merchant account provider has likely placed limits on how many transactions (both in terms of count and dollars) that you can conduct each hour, day, week, and month. This is common, so you should space out your rebills to not run into this issue.
  • You submitted an unsupported card type with your request
    • Some card types (i.e. government-issued prepaid cards) are not supported by all gateways. It’s best to move these to other merchant accounts or get an alternate payment method from the customer.
  • Restricted card
    • This is almost always due to a card breach and should never be reattempted.

You can analyze your response codes and come up with a variety of fixes that will pare down these structural errors. This should be part of your regular process as it’s a freebee. The improvements you make will help over the long term giving you a great ROI on any money spent on consulting.

Bad Data

After looking through your response codes, you may have seen some warnings related to the client’s name and address. It’s normal for bad data to slip into your CRM. Make sure that the client’s name and billing address use a standard format. And, where possible, remove special characters and accents. You may only want to make those changes as the card is processed, leaving the original visible to your client, so this might require some software to do this on the fly.

MCC

The next part will require some research and is more of an art, but this technique will work if you can figure out how to best position your business. Merchant banks assign each merchant a Merchant Category Code (MCC) as prescribed by the Card Networks (see our many other articles on MCCs!). This code identifies the type of merchant and card-issuing banks weigh their fraud scores by MCC. It makes sense – which MCC would you expect to have a bigger problem with fraud, 5967 (Inbound Telemarketing) or 5949 (Needlework, Sewing or Fabric and Piece Goods Stores)? As expected, the banking data we’ve analyzed for 2019-20 saw nearly four times the decline rate for MCC 5967.

The delta in approval rates can be massive between MCCs. Where some will have average approval rates in the 90’s, others will be in the 70’s, or lower! Work with your campaign manager to find out what MCCs are available to your business, then work with your merchant bank to ensure that you get a favourable MCC that’s still properly descriptive.

Timing

Next is timing, which can be a real issue. There are times, days and dates which yield better approval rates for your clients. You can imagine that the likelihood of an NSF would change over the month, based on when your customer is likely to get paid (in the case of a debit card), or when they’re likely to pay their bill (in the case of a credit card). There are also certain cards that get refilled on specific days of the month. If you have a good number of those in your client base, then you might change your billing cycle for them.

We’ve seen that different times of day have significant variations in average approval rates. Run a test charging cards at different times of the day and see if there is a variation. Remember to use large data sets to ensure statistical significance. With any luck, you’ll find bands of time with higher approval rates and switch your traffic into those bands. But remember to run this test from time to time to capture any fluctuations.

Location

You might be surprised to know that there are strong correlations on approval rates when comparing your client’s address with that of your company. Generally, if your business and your customer are in the same state, approval rates will be higher. If you dig into the data, you’ll see all sorts of insights on geography’s effect on approval rates. We’ve written an article that gets into it a little bit more – check that out here. You can set up merchant accounts for your different locations to mitigate the worst deltas in approval rates caused by location.

Price

Unsurprisingly, your price point also influences your approval rates. In general, the higher the price, the greater the likelihood that the transaction will be declined – and not just for Insufficient Funds! Fraud detection is more stringent on higher prices. We’ve also seen variations right down to the number of cents in the price. Run another experiment where instead of charging $29.99, you change the cents to $29.75 or maybe even $30 (equally applicable in all other currencies we’ve analyzed).

If you do all the above and other things like them, you should yield a nice increase in your approval rates. Keep in mind that while there is some effort here, if you implement systems to make these adjustments regularly, you will reap the benefits automatically, for years to come.

Recovery Efforts

Despite these efforts, you are still going to get some declines. When looking at response codes, you will see the dreaded ‘Bank Decline’ providing little clue as to what happened and other common codes, like Insufficient Funds. These are often unpreventable. However, recovering these declines can be more impactful to your bottom line than anything else, including adding new customers.

You may have access to an automated recovery engine through your CRM or gateway. If you use it, you will get some customers recovered. Most likely, you will see the best results with “Insufficient Funds” declines. However, their core business is holding your client information and executing your subscription schedule – decline recovery is a secondary or tertiary service to them and not their core competence. And if decline recovery is not what you’re doing all day long, then you’re leaving a lot on the table. It’s a tough discipline.

There are a few companies, including ours, that make decline salvage a high priority. To our knowledge, we’re the only company that is wholly focused on automated, AI-driven decline salvage. Look at this article that shows how using machine learning is the best way to ensure that the recovery service gets the highest recovery rates.

Picking a Recovery Service

When you talk to a recovery service provider, you should carefully consider their pricing model’s implications. Are they based on lift? What if their fees are greater than your gross margin? How many additional payments will you get after the recovery? What will be the impact on your bottom line in the first month? What will be the effect six months down the road? Ask them to model it out for you.

While we have evidence that shows that decline salvage has no appreciable affect on chargebacks, adding a whole bunch of new approvals to your book of business will, necessarily, increase the count. Be sure that you understand what that means for your business.

Ask about their strategies. While a careful and considered approach to recovery won’t increase chargeback and refund rates, there are strategies that will. Ask about forced capture. This is an aggressive recovery strategy that is certain to cause your customers difficulties ranging from over-draft fees to displeasure with your brand.

What’s their target average retry count? It’s important that recovery balances the upside of more approved payments with the downside of lower average approval rates on your merchant account. If the provider is too aggressive, they can drive your approval rate so low that the merchant account provider might close your account. In addition, we have some evidence showing that lowering your average approval rate can negatively impact your approval rate the first time a transaction is attempted. This results in even more declines!

Make sure you’re well informed, and you’ll find a partner that has your long-term interests at heart. (They should as it’s also in their interest that you do well)

Dunning

Our last piece of advice is to consider a dunning service. When automated recovery fails (it happens), this is a perfect time to use a dunning service to try to get new payment information. We’ve seen success with email, SMS, and even direct outbound calling. The advantage of this approach is that new payment information is likely the only way to get. This kind of service can be expensive, so our recommendation is to let automated recovery run first. A good recovery service will know which transactions are most likely to need dunning for recovery, so get a rejection list from them, and use that to seed your dunning service. That should speed up the process and get those clients back in action.

Conclusion

This is a lot to digest. In essence, the approach is to avoid a decline in the first place with good business practices and a buttoned-down backend. Then, use automated recovery and dunning to recover the rest. These are customers that want your service. They’ve already been sold. You’ve already done the hardest part when they signed up in the first place. It’s hard to imagine much that will have a bigger effect on your bottom line than recovering them and getting that long stream of future payments. Following this approach will help you reduce involuntary churn.


FlexPay can help you reduce involuntary churn! Contact us to find out more.

Categories: Knowledge Base