How Hong Kong Businesses Compare International SMS API Deliverability — Four Tests You Can Run Yourself
Published
On this page
Key takeaways
- Deliverability is not a static figure on a proposal; it shifts with routing, content type, number status, country and carrier.
- Real-number testing should use at least 10–20 test numbers, split into plain OTP digits and marketing copy with links.
- Many platforms return API success when the request is accepted, not when the handset receives it, so reconcile against DLR results.
- Pricing clearly below market, rewritten Sender IDs and DLRs rarely showing delivered are common symptoms of grey routes.
- Compliance is local, so when comparing vendors ask who handles Sender ID and template approval per region, not only unit price.
Searches like compare international messaging APIs for delivery rates in Hong Kong really ask: How do I know the 99% deliverability on a quote holds for my numbers, my content, and my target countries?
Deliverability is not a static figure on a proposal — it shifts with routing, content type, number status, country, and carrier. Here are four tests Hong Kong businesses can run before contract or in quarterly review.
Test 1: Real-number delivery by carrier and content type
Method:
- Prepare at least 10–20 test numbers across major Hong Kong carriers (and overseas segments where your customers concentrate)
- Two content groups: plain OTP digits vs marketing copy with links
- Send from the same provider in the same time window; record DLR results
What to watch:
- Delivered rate and average latency (OTP often needs under 10 seconds)
- Whether failed carrier error codes are interpretable
- Whether marketing lags OTP abnormally (possible filtering)
Test 2: API “success” vs DLR “delivered” reconciliation
Many platforms return success when the request is accepted, not when the handset receives the message.
Reconciliation formula:
Your API request count ≈ provider submit-to-network count ≈ (delivered + explicit failed + pending timeout)
Missing volume in the middle warrants questions on grey-route drops or internal throttling. UFOSEND commits to full submit — accepted requests are submitted to the network and reconcilable with your logs.
DLR meanings: SMS DLR status codes explained.
Test 3: Grey routes and Sender ID authenticity
Grey routes forward via third parties. Common symptoms:
- Pricing clearly below market
- Unstable or rewritten Sender ID
- DLR mostly submitted, rarely delivered
- Deliverability “mysteriously” drops over time
Self-check methods: Grey routes vs direct carrier connection — including asking for SS7 direct path and sampling carrier responses.
Test 4: Multi-country routing and local compliance
One API for the world is common; compliance is local:
| Region | Common requirements |
|---|---|
| Hong Kong (852) | UEMO, DNC, OFCA “#” Sender ID (high-trust scenarios) |
| Taiwan (886) | Real-name rules, enterprise numbers, unsubscribe — see SIM real-name impact |
| China (86) | Signatures, unsubscribe keywords — see Sending SMS to mainland China |
When comparing vendors, ask: Who handles Sender ID / template approval per region? Are failure codes broken down by country? — not only unit price.
Vendor comparison (for RFPs)
The table below is for evaluation; “Typical international self-serve API” summarises common platform types — confirm against contract, routing, and your live-number tests.
| Provider | Hong Kong direct | DLR granularity | Full submit reconcilable | Multi-country compliance support | OTP dedicated route | Dashboard / webhook |
|---|---|---|---|---|---|---|
| UFOSEND | SS7 direct to carriers; no grey routes | Carrier-grade delivered / failed with error codes | Yes; accepted requests submitted to network; reconcilable with client logs | 852 / 886 / 86 Sender ID and compliance path support | Dedicated OTP route, separated from marketing | Real-time dashboard; DLR webhook (HMAC supported) |
| Typical international self-serve API (e.g. Twilio-class) | Varies; some via aggregator / forwarding | Often submitted; carrier detail depends on route | Contract-dependent; submit count may not reconcile | Mostly documentation; OFCA “#” often self-service | Often shared with marketing traffic | Webhooks available; granularity and latency vary by route |
Beyond deliverability rate
- P95 latency (OTP scenarios)
- Failure code distribution (invalid number vs carrier reject vs content filter)
- Complaints and unsubscribe rate (marketing)
- Billing segment accuracy — see SMS billing and Unicode segments
Conclusion
International SMS API deliverability cannot rely on marketing copy alone. Real numbers, split content, reconcile DLR, inspect routing, split compliance by country — five steps put you ahead of most RFPs.
Need help designing Hong Kong and multi-country comparison tests? Contact us; technical access: SMS API.