Card account updater services compared: Stripe ACU vs alternatives

Roughly one in four failed Stripe charges in month 12 of a subscription comes back with an expired_card decline code. That single bucket is what card account updater services were built to fix — and where most stores either over-spend or under-build.

What a card account updater service actually does

A card account updater service queries Visa, MasterCard, and American Express for refreshed cardholder details when a stored card is replaced, reissued, or its expiry date moves. When it works, the new card details land in your vault automatically and the next renewal charge succeeds — no customer touch required.

The category has been quietly maturing for ten years. Visa Account Updater (VAU) and MasterCard Automatic Billing Updater (ABU) are the underlying network services. Every commercial card account updater service — including Stripe's free ACU — is a wrapper around those network feeds plus some logic about how often to query, when to retry, and which results to trust.

Three things determine how much value you actually get from one:

1. How often it polls the networks. Once a month vs once a week is a meaningful difference for monthly billers. 2. How it handles "no match" responses. A good service flags these for manual update flows instead of silently failing. 3. Whether it queries pre-emptively, on-decline, or both. Pre-emptive (before the card expires) is the highest-leverage mode — and the one Stripe's free tier skips.

For background on the decline codes that ACU is designed to fix, anything returning expired_card, incorrect_number, or some incorrect_cvc cases is ACU territory. do_not_honor and insufficient_funds are not — those need retry-timing fixes, not card refreshes.

Stripe ACU: the free baseline

Stripe's Card Account Updater is included with every Stripe account at no additional cost. It runs on Visa, MasterCard, and American Express feeds with sensible defaults:

• Polling cadence: weekly for active cards, monthly for dormant ones.
• Trigger model: mostly reactive — it queries on decline events for issuers that support real-time lookup, plus the periodic batch.
• Coverage: Visa, MasterCard, Amex. Discover ACU exists but Stripe's coverage there is patchier.

Across the 800+ Stripe accounts in our customer base, Stripe ACU resolves roughly 60% of expired-card declines within 30 days. That drifts up to 65-70% on B2C consumer subscriptions where issuers push updates aggressively. It drops to 45-55% on B2B corporate cards, which are slower to publish updates to the network feeds.

The 30-40% that ACU misses split into three buckets:

• The card was cancelled, not reissued (no replacement exists in the network feed).
• The cardholder switched issuers entirely (Bank of America → Chase). ABU/VAU only sees changes within the same issuer.
• The issuer is small enough that they don't push updates to VAU/ABU at all — common for credit unions and regional banks.

The honest read: for a monthly D2C subscription under $100/month, free Stripe ACU is enough. The marginal revenue from layering a paid service on top usually doesn't clear the subscription cost. For anything else, keep reading.

The third-party tier: Pulsate, SwitchKit, and Recurly's built-in

Three services dominate the beyond-Stripe-ACU market: Pulsate (recently rebranded from Pulse Commerce), SwitchKit, and Recurly's bundled updater. They all sit on top of the same underlying VAU/ABU feeds, but they differ in three structural ways: polling frequency, pre-emptive vs reactive triggers, and direct-bank integrations.

Pulsate

Pulsate's pitch is pre-emptive ACU: instead of waiting for a decline, it queries the networks 30-60 days before a stored card's known expiry date and proactively updates. For annual SaaS subscriptions or high-AOV memberships, this matters a lot — you'd rather catch the expired card in February than discover it on January 1st when 4,000 annual renewals run at once.

Their reported recovery rate on expired_card declines sits around 78-82%, against Stripe ACU's 60%. The marginal lift is meaningful, but only if your AOV justifies it. Pricing is typically 1.5-2.5% of recovered revenue — honest performance-based pricing similar to what Churnbuster charges on the dunning side.

Best fit: annual SaaS, high-AOV memberships ($500+/year), B2B subscriptions.

SwitchKit

SwitchKit takes a different angle — it integrates directly with several large issuing banks (Chase, Bank of America, Capital One, a handful of others) for real-time card-update feeds rather than relying entirely on the once-weekly VAU/ABU batch. For issuers it covers, cards get updated within hours, not days.

The catch: the issuer-direct coverage is U.S.-centric and skews toward the top 15 banks. For a customer base that's 70% Chase + Bank of America + Citi, SwitchKit shows meaningful lift. For European customers or anyone with a long tail of small issuers, the lift is closer to flat vs Stripe ACU.

Their published recovery rate is 72-75% on the specific issuers they cover directly. Pricing is per-card-updated, typically in the $0.40-$0.80 range.

Best fit: U.S.-only D2C with a top-bank-heavy customer base, monthly billing.

Recurly's built-in updater

If you're already on Recurly as a billing platform, their built-in account updater is bundled into the platform fee. It's roughly on par with Stripe ACU on coverage but adds a pre-emptive 60-day-before-expiry query that Stripe ACU doesn't do by default.

The trade-off: Recurly's billing platform itself is a multi-thousand-dollar-a-month commitment for most accounts. Don't migrate to Recurly *for* the ACU — but if you're already there, you're already paying for one. No reason to layer Pulsate on top unless your AOV is unusually high.

When ACU alone isn't enough

Three signals tell you it's time to look beyond free Stripe ACU:

1. Annual billing. A missed expired-card update on an annual renewal is a $500-$5,000 mistake. The math on pre-emptive ACU is overwhelming when the alternative is a 12-month gap to the next attempt. 2. High customer LTV. If recovering one extra customer is worth $1,000+, a 1-2% performance fee on a paid service is trivial. Below $300 LTV, the math gets harder. 3. Material chunk of cards from major U.S. issuers. If 60%+ of your stored cards are Chase, BofA, Capital One, or Citi, SwitchKit's direct-feed lift compounds.

If none of the three apply, free Stripe ACU plus a tight magic-link card-update flow (covered in our Stripe failed payment recovery playbook) recovers nearly everything ACU misses, at no extra cost.

The math: what a paid ACU is actually worth

A worked example. Say you run a $5M ARR subscription with a 7% monthly involuntary churn rate. Of that 7%, roughly 25-30% is expired_card declines — call it 2% of MRR per month, or $100,000 ARR at risk annually.

Free Stripe ACU recovers about 60% of that pool, or $60,000. Pulsate or SwitchKit recovers about 78%, or $78,000 — a marginal lift of $18,000.

At 2% of recovered revenue, Pulsate's fee on the full $78,000 is $1,560/year. Net lift: ~$16,500. That's a clear yes.

Now run it on a $500K ARR subscription. Same math: $10,000 ARR at risk, $6,000 recovered by Stripe ACU, $7,800 by Pulsate. Marginal lift: $1,800. Fee: $156. Net: $1,644 — still positive, but small enough that the engineering time to integrate and monitor a second service rarely clears the bar.

This is the same shape of analysis we use for Stripe Smart Retries vs a timing-optimized recovery layer: the free baseline gets you most of the way, the paid service captures the long tail, and whether that tail is worth chasing depends on your AOV.

The three failure modes nobody documents

Even with a paid ACU layered on top, three patterns cause silent losses:

1. The "no match" silent fail. When ACU returns "no replacement card available," most stores treat it as a permanent decline and never retry. In practice, around 25% of "no match" cards come back into the network feed within 60 days as the issuer eventually pushes the update. A good system re-queries quarterly on these, not just once.

2. Issuer-side BIN range changes. Once or twice a year, an issuer migrates a chunk of customers from one BIN range to another (typically during portfolio acquisitions or rebrand events). ACU eventually catches these, but the lag can be 60-90 days. During the gap, retries fail with generic_decline or do_not_honor and look indistinguishable from the customer's account being closed. Most stores write these off — they shouldn't.

3. Network token drift. If you're using Stripe's network tokens (and you should — they survive card replacements better than raw PAN storage), ACU updates can fall out of sync with the token vault. Stripe's reconciliation is good but not perfect. We see roughly 1 in 200 token-stored cards develop a sync issue after an ACU update. The fix is a periodic audit query, but most stores don't run one.

A fourth, milder pattern: ACU services sometimes mark a card as updated when only the expiry date changed but the underlying PAN is the same. This is technically correct — the card *was* updated — but it doesn't help you recover a decline that came back with do_not_honor rather than expired_card. Watch the decline code, not just the ACU success rate.

Build vs buy, redux

Three honest answers, ARR-bucketed:

• Under $1M ARR: Stripe ACU + a magic-link card-update flow on the front end. Don't pay for a service yet. The marginal recovery from Pulsate or SwitchKit at this volume is usually under $200/month — not worth the integration time.
• $1M-$20M ARR with annual billing or high LTV: Pulsate or SwitchKit pays for itself within 90 days. Pick based on customer geography: SwitchKit for U.S.-heavy, Pulsate for international or mixed. If you're already on Recurly, you've got one — don't double up.
• $20M+ ARR with internal engineering: Building direct VAU/ABU integrations is technically possible but rarely justified. The underlying networks charge per query, the operational overhead of running an ACU pipeline is real, and the vendor stack at this scale tends toward "buy the specialist." Most stores at this scale still buy.

The point isn't which service is best in the abstract — it's which one closes the specific decline-code distribution you actually have. A SwitchKit account that doesn't cover your customers' issuers is worse than free Stripe ACU. A Pulsate account on a monthly-billing D2C business pays for itself slowly, if at all.

Run a leak scan on your own stack

Picking the right card account updater service is one decision in a stack with twelve more like it — retry routing, dunning cadence, magic-link flows, BIN-level analysis. Omesta scans for 147 leak patterns across the whole recovery surface in 2 minutes, no code changes, read-only OAuth. We'll show you which ACU layer (if any) actually moves the number on your specific data.

Start the leak scan — Only $7 activates 30 days, $1,000+ recovered or your $7 back.