What to Know about Spam Filtering

Any business sending SMS or MMS messages in the US or Canada must comply with certain industry guidelines, mobile operator policies, and regulations. To help ensure compliance and to avoid fees and potential litigation associated with non-compliance, OpenMarket and other mobile service providers have implemented filters to identify and block spam.

Starting in April 2021, OpenMarket will filter most messages sent from 10DLCs. Initially this applies to customers using the v3 and v4 SMS products. Our filter will look only at one-way MT messages, and will analyze them based on country-specific policies.

To understand the regulations relevant to messaging in the US, we encourage you to be familiar with these documents: 

  • CTIA Messaging Principles and Best Practices
  • CTIA Short Code Monitoring Handbook
  • CTIA Short Code Monitoring Handbook Addendum
  • AT&T Code of Conduct
  • AT&T A2P Code of Conduct
  • T-Mobile Code of Conduct

You can find links to all of these on the Global Coverage section of the docs site.

What is spam?

Spam is generally defined as content the end user has not consented to receive. Spam includes spoofing and smishing.

Spoofing is changing an originator to hide the sender's true identity and trick a consumer into thinking a message is from someone they know or a legitimate commercial entity. Smishing is a combination of spam, spoofing and social engineering techniques aimed at misleading the end user to gain access to online systems, accounts, or data. Smishing is considered a criminal activity. Typically spam is sent in bulk to a large number of end users.

Mobile network operators in the US usually consider the following types of content to be risky:

  • Debt reduction programs
  • Fast cash promotions
  • Work from home offers
  • Gift card promotions
  • Win free stuff offers
  • Adult content

Other indications that a message could be spam include:

  • Bad syntax, grammar, spelling in the message content or SenderID/Originator
  • Commonly used strings in bad URLs
  • Sender IDs prohibited by Carriers/Operators
  • Traffic patterns and volumes
  • No actual opt-out mechanism

As the incidence of spam has increased, some mobile operators have implemented their own spam filters, or they've required messaging service providers to implement them, which means your messages may be analyzed by multiple filters as it travels over the networks. The goal of filtering is to prevent unwanted content from reaching end users, who may in turn file complaints with government regulators (e.g., the FCC in the US) or with their mobile operator.

How filtering works

If you are using either the v3 or v4 SMS HTTP API or SMPP, you should assume that your messages will pass through one or more filters as they travel to the destination handset. Most messages sent from 10DLC numbers will be analyzed by the filter we operate.

OpenMarket's filter operates on a set of policies that reflect country-specific regulations. For SMS, the filter analyzes one-way MT messages based on various factors, including:

  • Message text, especially call-to-action text
  • URLs
  • Traffic patterns and irregular volumes
  • Source numbers that are already blacklisted

Any message that's identified by our filtering system as spam will be blocked. Our system will return one of these error codes: 

607

The message was blocked by a spam filter and will not be delivered to the end user's handset. If you believe your message was falsely identified as spam, please submit a ticket to OpenMarket Support.

You cannot re-try blocked messages.

561

The message was blocked by AT&T's spam filter.

You cannot re-try blocked messages.

You'll be able to use the standard SMS Summary and Detailed reports to look for instances of these response codes.

What to do if your message is blocked

It is possible that a message is falsely flagged as spam. We call these false-positives. If you think a message you sent has been falsely identified as spam, you should immediately submit a ticket to Support. We will investigate the message and if we conclude that it was a false-positive, we will update the filter policy and your message will not be subject to further filtering.