Attentive SMS: Consent Capture, Segments, Send Limits

What actually syncs when you connect Attentive SMS to Shopify

Attentive SMS pulls a defined set of Shopify data on a schedule and via webhook: customer profiles, order events, cart activity and product catalogue fields you map during setup. That sync is real and it’s the part every integration page covers well — customer joins your list, places an order, abandons a cart, and Attentive has the data to trigger a message within a reasonable window.

What doesn’t get the same attention is what happens once that sync is running against $3M+ of order volume, multiple lists, and a compliance regime that assumes every message has a clean, verifiable consent trail behind it. That’s the part this article covers: the three things that break once volume goes up, and the fix for each.

This article is written for operators running Shopify Plus or a paid subscription platform at $3M–$30M in revenue, who already have Attentive connected and are past the initial setup. If you’re evaluating whether to buy Attentive SMS at all, or you’re under the $3M floor, this isn’t the article: the three failure modes covered next only show up once you’re sending real volume to a real list.

What doesn’t sync automatically

Three things live outside the standard Shopify-Attentive data flow, and each one is a place teams assume synchronisation that doesn’t exist:

• Consent scope. Shopify’s own checkout consent checkbox and Attentive’s opt-in event are recorded in two different systems with two different legal bases, and nothing forces them to agree.
• Segment freshness. Attentive segments are built from Shopify data on Attentive’s refresh cycle, not the instant a Shopify event fires.
• Send-volume signals. Attentive’s send queue doesn’t know what a mobile carrier is about to do with that volume once it leaves Attentive’s infrastructure.

Each of those gaps becomes a specific, repeatable failure mode. Here they are, with what actually fixes each one.

Break 1: consent scope drifts between Shopify’s checkbox and Attentive’s opt-in record

A shopper ticks the SMS box at Shopify checkout. Weeks later, marketing sends a campaign through Attentive assuming that checkbox created a valid Attentive subscriber. Sometimes it did — if the checkout integration is configured to push that consent event into Attentive as a proper opt-in. Sometimes it didn’t, because the checkbox copy, the sync timing or a theme customisation meant the event never reached Attentive as a recognised subscription.

The result is a list that looks compliant in Shopify’s customer record and isn’t verifiable in Attentive’s own consent log, or the reverse — a shopper who unsubscribed through one channel and never dropped out of the other. Neither state is visible from the campaign-builder screen, which is exactly why it survives past launch.

The fix is to treat Attentive’s own consent record as the source of truth for what Attentive is legally allowed to message, and audit it against the Shopify checkout event on a schedule, not once at setup. Pull a sample of recent Shopify SMS opt-ins and confirm each one has a matching Attentive subscription event with a timestamp and a source. TCPA-shaped compliance rules, and how strict a consent trail needs to be for your message type and state mix, are the kind of thing to confirm directly with counsel rather than infer from a vendor’s setup guide — the requirements are specific enough, and change often enough, that a general description isn’t a safe substitute.

Break 2: segments lag behind the Shopify customer state they’re built on

An Attentive segment built on “customers who ordered in the last 30 days” or “subscribers who haven’t purchased” is a snapshot, refreshed on Attentive’s own cycle. At low send volume the lag between that snapshot and the live Shopify customer state is invisible — nobody notices a segment that’s a few hours stale when you send once a week.

At $3M+ order volume, with daily or multi-daily sends, that lag becomes visible fast: a shopper who cancelled a subscription still gets the win-back text; someone who just placed a first order still lands in the “never purchased” campaign; a returned item doesn’t pull the customer out of a post-purchase upsell segment in time. None of this is a bug in Attentive — it’s the ordinary cost of building live-feeling messaging on a periodically refreshed segment.

The fix is to know your segment refresh cadence for each segment type and build a buffer into anything time-sensitive — exclude orders from the last day or two in a win-back segment rather than trusting the segment to have already dropped them, and re-check high-stakes segments (cancellation-triggered, refund-triggered) against a live Shopify query before a send goes out, not just before the campaign is built. If you’re also running Klaviyo flows against the same customer base, this is the same class of problem Klaviyo’s own segment building has — see how it plays out on the email side in Klaviyo’s benchmark reporting across 183,000+ brands, where 41% of total email revenue now comes from automated flows rather than one-off campaigns (Klaviyo, vendor-reported): the flows carrying that revenue share are exactly the ones built on segments, which makes segment freshness a revenue question, not just a compliance one.

Break 3: send frequency and carrier throttling collide during your busiest sends

Attentive queues and sends your message; what happens next is between the carrier network and the recipient’s device, and carriers apply their own filtering logic that Attentive doesn’t fully control. A shared short code splits its reputation across every brand using it, so a spike in complaint volume from another sender on that code can degrade your deliverability without you changing anything. A dedicated 10DLC or toll-free number isolates your reputation, but has its own volume thresholds a carrier watches for sudden spikes.

The failure shows up as messages that send successfully from Attentive’s side but arrive late, arrive out of order, or don’t arrive at all — and because the failure happens after Attentive’s own send confirmation, it’s easy to miss unless you’re actively watching delivery and complaint metrics by campaign, not just send confirmations.

The fix has two parts. First, match your number type to your volume and message mix before you scale sends, not after deliverability degrades — a dedicated number is worth the registration overhead once you’re sending consistent marketing volume rather than occasional alerts. Second, watch unsubscribe and complaint rate per campaign as your early warning signal; a rising complaint rate on a specific segment or send time is usually the leading indicator of carrier throttling, arriving before open or click data would show it. Send frequency caps set per segment in Attentive reduce the complaint volume that triggers this in the first place — there’s no universally safe cadence, so set the cap from your own unsubscribe data rather than a number carried over from email.

Growing the list without diluting what you can actually message

Every Attentive SMS programme needs new subscribers coming in, and the mechanism most teams use is a two-tap sign-up unit: a shopper taps once to reveal a phone number field, taps again to confirm, and the opt-in event fires without a full-page redirect. Where that unit sits matters more than its copy. A unit embedded in the cart drawer or post-add-to-cart confirmation catches shoppers already committed to buying, which tends to produce a smaller but more purchase-intent list. A unit surfaced as an on-load popup across the whole site catches far more traffic, including browsers who were never close to buying, which grows the list faster but adds subscribers who are more likely to sit unengaged or unsubscribe at the first promotional send.

That’s the real trade-off behind list growth, and it doesn’t resolve itself — a bigger list is not automatically a better one for SMS the way it can be for email, because carrier reputation (covered in Break 3) responds to complaint rate, not list size. A site-wide popup that doubles your subscriber count while tripling your complaint rate is a net loss even though the top-line number looks like progress. Before adding or widening a sign-up unit, decide what you’re optimising for: a smaller list that opens cleanly for post-purchase and shipping messages, or a broader list you intend to message less often and filter harder before every marketing send. Running both, a high-intent unit at cart and a broader one at exit-intent, each feeding a different segment and cadence, is a reasonable middle path, but only if the two feeds are tagged separately in Attentive so you can treat them differently at send time rather than merging them into one undifferentiated list.

Which triggered flows belong on SMS, and which don’t

Not every flow that works in email is worth building in SMS, and treating the two channels as interchangeable is a common way to burn list goodwill fast. The flows that genuinely suit SMS share a pattern: they’re short, time-sensitive, and the shopper is glad to get a text about them specifically because a text arrives faster than they’d check email.

• Shipping and delivery updates. A shipped or out-for-delivery notification is exactly the kind of message a shopper wants immediately, not the next time they open their inbox. This is usually the strongest-performing SMS flow precisely because it’s not promotional: it’s information the customer asked for by placing the order.
• Back-in-stock alerts. Stock windows can close within hours for popular restocks, and SMS’s near-immediate delivery is a genuine advantage over email here, where the same alert might sit unread until the item sells out again.
• Cart abandonment, sent at a short delay. A cart reminder sent within the first hour or two, while purchase intent is still warm, is a reasonable SMS flow. The same cart reminder sent a day or two later starts to read as an ad rather than a nudge, and by that point email, which the shopper doesn’t experience as an interruption the way a text can feel, is usually the better channel.

Flows that don’t belong on SMS are the ones built for browsing rather than urgency: welcome series beyond the first message, education or how-to content, multi-touch win-back sequences, and anything that needs more than a line or two of copy to make its point. These work in email because a shopper can skim or skip them at their own pace; the same content sent as a text interrupts, and a long nurture sequence on SMS is one of the more reliable ways to drive up the unsubscribe rate that then triggers carrier throttling. The general rule: if the message content is time-sensitive and short, it’s a candidate for SMS; if it’s exploratory or lengthy, keep it in email and let SMS stay reserved for the messages a customer is actually waiting on.

List hygiene: suppressing subscribers who never engage

A subscriber who hasn’t opened, clicked, or replied to any message across an extended run of sends isn’t neutral. On SMS, an inactive segment is a liability in a way it usually isn’t in email, because carriers weigh engagement signals (or the lack of them, alongside complaint rate) when they decide whether a sender’s traffic looks legitimate. A list that’s grown wide and stayed unpruned looks, from a carrier’s perspective, similar to a list built through low-quality or non-consensual acquisition, even if every subscriber on it opted in cleanly at some point.

The fix is a standing suppression practice rather than a one-time cleanup: define what “never engaged” means for your programme, a subscriber who hasn’t interacted with any send across a defined recent window, and move that segment out of regular marketing sends. That doesn’t mean removing their consent record or treating them as unsubscribed; it means routing them out of the campaigns that drive volume and complaint exposure while leaving the door open for a lighter-touch re-engagement attempt, sent at low volume and watched closely for complaint rate before you’d ever consider folding them back into full sends. Skipping this because the list total looks worse afterward is a mistake in the other direction: a sender-reputation problem caused by an unpruned list depresses delivery for every subscriber, including the engaged ones, not just the inactive ones you declined to suppress.

Running SMS and email without messaging the same person twice

Most $3M+ programmes run Attentive SMS alongside an email platform against the same Shopify customer base, and the failure mode that shows up isn’t a technical sync error: it’s two platforms independently deciding the same shopper qualifies for the same message. A cart abandoned at 2pm can trigger an Attentive SMS flow and a separate email flow within minutes of each other if neither platform is told the other already reached out, and a shopper who gets both about the same cart within the same hour reasonably reads that as a company that doesn’t talk to itself.

There’s no default coordination between the two platforms; that has to be built deliberately, and the two common approaches are worth naming. One is channel priority: decide, per flow type, whether SMS or email fires first for a given trigger, and suppress the second channel for a defined window if the first one sent. The other is true suppression logic, where each platform checks a shared signal (a Shopify tag, a synced field, or a light integration between the two tools) before sending, so the second platform can see that the first already messaged this customer about this event. Channel priority is simpler to set up and is usually enough for the highest-volume flows like cart abandonment and post-purchase; true cross-platform suppression matters more for anything sent frequently to a broad segment, where the odds of an overlap compound across every send rather than showing up occasionally. Whichever approach you pick, decide it before launch. Retrofitting suppression logic after a list has already learned to expect duplicate messages is slower and less effective than building it in from the start.

What to check before a peak-season send, and what to do if delivery drops mid-send

Consent drift, segment lag and carrier throttling are ordinarily tolerable at everyday send volume and become acute exactly when volume spikes, which is what a peak-season send is. A pre-send check against each break point, done a day or two ahead rather than at send time, catches most of what would otherwise surface as a live incident:

• Consent. Confirm the segment you’re about to send to has been reconciled against Attentive’s own consent log recently, not just built from a Shopify tag; a peak-season send typically reaches the widest audience of the year, which is the worst time to discover a consent-scope gap.
• Segment freshness. Re-pull any time-sensitive exclusion (recent purchasers, recent unsubscribes, recent refunds) as close to send time as the platform allows, since the lag between a segment snapshot and live Shopify state matters more the bigger the send.
• Number and volume headroom. Confirm your sending number’s registered volume tier can absorb a send that’s meaningfully larger than a normal week’s traffic, and confirm that with enough lead time to register a second number or raise a volume tier if it can’t. This isn’t something to discover the morning of the send.
• Frequency stacking. Check what else is scheduled to hit the same segment in the surrounding day or two, on both SMS and email, so the peak-season send isn’t landing on top of an already-heavy cadence.

If delivery visibly degrades mid-send, meaning messages send from Attentive but confirmations, replies or complaint signals arrive late or not at all, the escalation isn’t to resend into the same conditions. Pause the remaining send queue for that campaign first, since continuing to push volume into a carrier that’s already filtering compounds the reputation damage rather than working around it. Check complaint and unsubscribe rate on the portion that did send before deciding whether to resume, since a spike there is the signal that the throttling is reputation-driven rather than a temporary carrier-side delay. From there, involve Attentive’s deliverability support with the specific number, time window and volume involved, since carrier-side filtering decisions aren’t something a sender can diagnose from campaign metrics alone. Splitting the remaining send across a longer window, rather than resuming at the same pace, is usually the safer path back once the queue reopens.

How do you verify consent, segments and delivery are actually in sync?

Verification here means checking the three break points directly, not trusting that “connected” in Attentive’s integration settings means “correct.” Pull a sample of recent subscribers and confirm each has a matching consent event in Attentive, not just a Shopify tag. Compare a time-sensitive segment’s current membership against a live Shopify query for the same criteria and measure the gap. Track unsubscribe and complaint rate by campaign and by number, and watch for a rise that isn’t explained by message content or timing — that’s usually carrier behaviour, not your copy.

Consent scope, segment freshness and carrier reputation checks are not one-time setup tasks. Each drifts over time as your list grows and your send volume increases, which is why this needs to sit on the same cadence as your other lifecycle checks rather than a launch-day checklist you never revisit.

Where this sits in your lifecycle programme

Attentive SMS doesn’t run on its own: it’s one channel inside a lifecycle flows programme that also includes email, and consent scope, segment lag and send-volume throttling are lifecycle flow problems before they’re SMS-specific ones. The same drift in tracked consent, live customer state and unsubscribe-tuned cadence will cause the same damage in any channel. If your Attentive programme is breaking in these specific ways, the fix belongs in how your lifecycle flows are built and audited, not in a one-off Attentive settings change — and if you want to see what a fix to segment timing or send cadence is actually worth, run the numbers through the flow revenue calculator before you commit engineering time to it.