Ecommerce SMS Statistics: Revenue, Conversion and Cart Abandonment

The key ecommerce SMS statistics

Sources: Omnisend, 2026 Ecommerce Marketing Report, Postscript, 2026 SMS Benchmarks and Baymard, Cart Abandonment Rate Statistics. The dollar symbols follow the original reports; the captured definitions do not specify an ISO currency. These are three separate evidence sources, with no combined market average.

Which SMS benchmarks can you compare?

Start with the unit being measured. A dollar figure per message, a merchant percentile and a popup signup rate answer different questions. Matching the label alone is not enough.

Use the benchmark to frame a test. Use your own eligible audience, consistent attribution and a comparison group to evaluate the result.

1. How much revenue does an SMS message generate?

In Omnisend’s 2025 data, automated SMS produced an average of $0.74 in revenue per message sent, compared with $0.15 for campaigns. Its report analyzed 321 million SMS messages and reported SMS volume growth of 40% versus the previous year. Read Omnisend’s report.

The practical question is which customer moments your program covers. A cart reminder reaches someone who has already built a cart. A campaign may reach a much broader group. That difference in intent is one reason to investigate flows separately, rather than expect the same return from every send.

Audit missing triggers, message eligibility, delivery and purchase suppression before simply increasing send volume. Then compare each flow against its own history using the same revenue attribution settings. A high revenue-per-send figure can coexist with a small audience, limited total revenue or heavy discounts.

Postscript’s separate merchant-level revenue-per-message distribution has a median of $0.98, with its 25th and 75th percentiles at $0.41 and $2.13. Its methodology describes these as merchant-level averages across 2025; the overall report window is 1 January to 15 December. Read Postscript’s methodology.

Do not blend Postscript’s median with Omnisend’s reported average or use their difference to rank platforms. The merchant samples, calculations and reporting windows differ. A merchant median gives each merchant a position in the distribution; Omnisend does not publish enough detail here to align its averaging method.

2. What is a good SMS conversion rate?

A useful answer starts with the event being counted and the denominator. Purchases divided by delivered messages, purchases divided by clicks and purchasers divided by eligible shoppers can all be called conversion rates. They produce different percentages from the same activity.

Postscript publishes the following message conversion percentiles. Its public benchmark page does not specify the conversion denominator or attribution window. Treat these as source-defined references. Ask for those definitions before comparing them with your own dashboard.

The table supports a diagnostic order, not an instruction to replace campaigns with cart reminders. Inspect high-intent flows for missing coverage and broken handoffs. Inspect campaigns for audience relevance and the action they ask the shopper to take. Judge both against the opportunity they actually address.

If a message receives a reply but no purchase, read the conversation before adding another discount. Identify unresolved delivery questions, product suitability questions or checkout friction. Those are investigation categories, not measured causes in these benchmark datasets.

3. What do the Health & Wellness figures show?

Postscript also separates Health & Wellness merchants. This is closer to the category context of many health and supplement brands, but it is still not a US supplement-only sample. The public page does not disclose the number of merchants in this subset.

Source: Postscript’s Health & Wellness table. Use the category as a closer reference to investigate, then segment your own data by product, order value, customer status and purchase cycle. The table does not establish the performance of an individual supplement, a medical claim or a result Zobi will deliver.

4. What phone signup conversion did updated popups show?

For merchants using Postscript’s updated popup editor, Postscript reports phone conversion rates of 2.1% at the 25th percentile, 3.6% at the median and 5.5% at the 75th percentile. Here the denominator is explicit: conversions divided by total impressions. Merchants with fewer than 1,000 popup impressions on either the previous or updated editor were excluded from this analysis. See the popup methodology.

Match the exposure rule before comparing your result. A popup shown to every visitor addresses a different audience from one shown only after a long product-page visit. Record display conditions, device mix, offer and impression counting alongside the conversion rate.

Then follow the cohort downstream. A signup rate alone does not tell you whether subscribers remain reachable, purchase, unsubscribe or contribute profit. Track list growth with subsequent subscriber value and the actual cost of the offer. Keep permission and preference handling in the operating process; these benchmarks are not a legal checklist.

5. Why 70% cart abandonment is not your SMS recovery audience

Baymard’s current compilation reports an average documented cart abandonment rate of 70.22%, based on 50 studies. It combines historical estimates from different sources and methods. It is useful context for checkout friction, but it is not a single-year global census. See Baymard’s study list.

That abandonment percentage does not tell you how many carts you can contact or recover. Some shoppers return without a message. Others cannot receive the proposed contact, are no longer interested, or encounter a problem a reminder cannot solve. Start with the contactable, eligible audience rather than the total abandoned-cart count.

Measure recovery as orders per eligible shopper over a fixed outcome period, with the same identity and deduplication rules in each test group. Also report total orders and contribution. A vendor-attributed recovery number answers which sales received credit under its rules. A randomized holdout can estimate how outcomes differ when eligible shoppers do not receive the tested treatment, provided assignment, tracking and sample sufficiency are sound.

6. Turn a revenue benchmark into a profit question

A higher credited revenue figure is a starting point for an investigation. To decide whether a program is worthwhile, account for discounts, variable order costs, refunds, message charges, extra service costs and the sales that would have happened anyway.

This is why the measurement worksheet below separates credited revenue from outcome and cost data. The example is arithmetic, not evidence that a discount will increase orders. Extra fixed costs, sales-based fees, returns, a change in basket mix or limited eligible audience can change the hurdle.

A measurement worksheet for your own store

Before judging performance against these references, fill one row per flow or campaign family in the downloadable worksheet. Keep the reporting window and cost treatment consistent. Leave unknown fields blank rather than converting them to zero.

Define the job: signup, first purchase, cart recovery, replenishment or another specific customer action.

Record eligible shoppers, assigned shoppers, messages sent and delivered, and the purchase outcome window.

Save the conversion formula, revenue attribution rule, click filtering and identity/deduplication method.

Separate new and returning customers when that distinction changes the decision.

Record net sales after discounts and refunds, variable costs and extra program costs once, with no double counting.

For an incremental result, preserve treatment assignment and compare contribution per assigned eligible shopper with a randomized holdout over the same window. Report uncertainty and operational defects.

Download the measurement worksheet ↗

Methodology, sources and reuse

The article excludes SMS open-rate folklore, Omnisend click-rate growth affected by its tracking warning, and ambiguous industry unsubscribe percentile rows. It does not mix email automation statistics into SMS claims. For message conversion, the missing definition stays beside the table and chart rather than being filled in from another report.

The downloadable data includes each reported value, percentile or aggregation, source URL, scope, data period and checked date. Source excerpts and file hashes are retained in the internal evidence packet for review. Multiple rows from one report are observations, not independent studies.

Download all 27 observations and source details ↗

Source check: Sources checked 26 October 2026. The data file lists 27 reported statistics from three publisher pages. Reporting periods are stated beside the relevant evidence. Changes will be recorded after the affected source and claim have been checked again.

Frequently asked questions

Are these 2026 SMS performance results?

The Omnisend and Postscript pages are 2026 editions, but the selected performance data are from 2025. Baymard’s abandonment compilation spans different historical periods. The checked date describes our review, not when shoppers were measured.

Is there one universal SMS conversion benchmark?

No single figure in these sources has a universal denominator and population. Identify the message type, shopper eligibility, conversion formula and attribution window before making a comparison.

Does a high revenue-per-message figure prove incremental revenue?

No. Revenue credit under a platform’s attribution rules does not establish the sales caused by the program. A well-run randomized holdout can estimate an incremental difference in outcomes.

Can a supplement brand use the Health & Wellness figures?

Use them as a category reference for investigation. The source does not isolate US supplement merchants or provide the exact subset sample size, so the figures should not become a promise or forecast.

What should we test first?

Inspect missing or broken high-intent flows, then choose a measurable intervention tied to a customer problem. Compare the outcome and contribution under fixed definitions rather than optimizing the benchmark percentage alone.

Put the evidence to work

Choose one flow, complete its measurement row and identify the customer question or friction it should resolve. Agree on the outcome, cost treatment and test design before changing the message. That gives your team a decision it can learn from, rather than another percentage to chase.

Start with the measurement worksheet ↗

Publisher disclosure

Zobi compiled this article from public primary-publisher pages checked on 26 October 2026. The downloadable data file lists 27 reported statistics from three publisher pages. The original reports retain ownership of their research. Charts and explanatory analysis were created for this article; this is not a Zobi merchant study.