While we describe how to get started with two-way SMS in our recent blog, there are a number of things to consider before embarking on that journey. We’ve put together our top ten list of considerations below. Let us know whether these resonate with you or whether there are things that you’ve noticed from your own experience setting up such systems and we can update this list.

1. Ensure that SMS is an ‘appropriate’ channel in the first place

“Let’s do a bulk-SMS campaign! It’s easy”

But what if the population don’t have access to mobile devices? Before choosing your channels for communication, you’ll need to undertake an assessment of the communities’ information and communication needs. This includes which channels they (used to) use, which they would like to use, and how they would like to communicate with humanitarian organisations. If SMS isn’t regularly mentioned, or only used by select groups, then an SMS campaign may only have limited impact. It might not be the best choice for an intervention. At worst, it could actively frustrated and/or confuse the communities you’re targeting. 

2. No messaging is ever ‘one-way’.

“We want to send messages to people. But we don’t want to get any back. We can’t / don’t want to respond”

So frequently the urge to use SMS as a tool comes from a desire to share information with a population – information ‘from us to them’. This precludes information being shared ‘back’. When a communication channel is opened – particularly with a population who feel isolated and unable to communicate with humanitarian organisations – people will take advantage of a new, open channel and start reaching out. Even if you try to manage expectations or write ‘DO NOT RESPOND’ there will nonetheless be inbound messages. As humanitarian’s we cannot leave what could potentially be urgent messages seeking help and protection unanswered. We need to understand that systems and protocols need to be put in place to respond to (and refer) inbound messages appropriately.

3. Be realistic about your capacity as an individual and as an operation

“We’ll just add it to Simon’s Terms of Reference”

Before deciding on if / how to start using SMS as a tool, it is important to look at your own internal capacity to design, deliver and maintain the channel. The resource requirements for different solutions vary and the needs cover technological, content and administrative components. As such, it is rarely one person’s job to implement the whole system. Before starting consider if you have you been able to engage the different staff in your operation who will need to be involved. Do they have time to engage? These are questions to address as they will help determine the solution you opt for. For instance, if you need to handle a lot of messages and simply don’t have capacity, maybe it’s better to consider an external service provider who can do the heavy lifting from a technological and administrative standpoint. 

4. Work out your workflow

“We’ll just send the messages out”

It is important to develop a full understanding of the entire workflow of information around both outbound and inbound messaging, ensuring that all aspects are mapped. As humanitarian responses are often undertaken by a complex array of humanitarian, government and other actors, you will also need to make sure you have referral pathways mapped into the workflow. This will ensure that any action required as a result of the messages received from communities can be managed by the appropriate organisations. This will help ensure that nobody’s requests and feedback falls through the cracks.

To help with this you might want to map out the information and referral flow using a online flowchart tools such as draw.io, lucidchart, or Visio.

5. Inter-operability is important!

“We’ll put the information into one big excel file”

Many requests to use SMS systems may require the use of another technological system. Is this a registration database, an incident management system? Perhaps you’ll require a system which catalogues red flags on specific issues such as SGBV, or allows you to link to a broader feedback management system or database. When undertaking the workflow exercise (above), include other platforms that you’ll link to. For example, do you need to pull data out of the system for analysis in a separate tool? This should be considered during planning stages so that no surprises emerge later – when it may be more costly, and complicated to implement.

6. Think about how you can be inclusive: Bridge the digital divide by design

“We’ll only contact the youth; they’re the ones with the phones ”

When undertaking an information campaign, or soliciting feedback by SMS – are we only targeting phone owners? In many contexts we see great disparities between different ages and genders regarding phone ownership. When considering using SMS, think about if and how these groups can be brought into and benefit fomr the the initiative. Perhaps your SMS system could be integrated into community structures and information passed on via word-of-mouth to those groups that don’t have access to a device. Perhaps additional device access is required – could this be addressed with targeted phone distributions? While each context will be different, always keep in mind that those who are excluded and the implications on their wellbeing. Those ‘left in the dark’ may even feel threatened due to their lack of inclusion.

7. Coordination or yet another phone number?

“X agency has this service, but we need our own!”

Like all channels, use of SMS isn’t reserved for one humanitarian organisation. It’s a technology that can be operated by numerous actors , simultaneously. Often organisations tend to set up their own separate systems, which can lead to confusion and duplication. As humanitarian responders, we have a collective responsibility to ensure that individuals are not overwhelmed by inbound messages. The experience should be coherent and systematic – with clear, actionable information sharing. Humanitarian organisations need to coordinate and cooperate. How do we do this? Firstly, having a forum to discuss such matters i.e. a working group on communicating with communities is often the first step. Then, when solutions around SMS are brought forward to the forum it is possible to avoid reinventing the wheel, and leverage investments already made in others’ systems. Even if it is the case that a separate system needs to be built, the coordination is still vital to ensure coherency for the community. 

8. Make sure you have procedures for developing the information that you want to share

“I’ll just write it and send it out myself”

Given the broad scale of communication and the requirements above relating to coordination, having Standard Operating Procedures for the content that needs to be shared with communities can make the process much easier. Defining who should be involved, who has ‘sign-off’ and how you will test the effectiveness of the SMS is critical. Before sending 1,000s of messages, test your content with a couple of community members/groups to check comprehension and identify any mis-interpretations. Having a content creation ‘process’ defined (at agency/inter-agency level as appropriate), will save time in the long-run. A more predictable system can also respond more effectively when timelines are ‘squeezed’ and you may need to develop and share information in an emergency. 

9. Airtime costs! See what Mobile Network Operators have on the table

“We’ll just buy $1000 airtime from the local shop and send to 50,000 refugees”

Airtime doesn’t come for free (usually). You’ll need to develop a clear understanding of the airtime costs involved in your project by determining the number of people you want to reach, how many messages to/from you’ll be sending and what the cost is per message. If the sums don’t add up, you’ll need to go cap-in-hand to your donor / budget team. Ideally, engage Mobile Network Operators (MNOs) to try and broker a special deal. Negotiations on cost, systems and so forth can take place to see how there could be a mutual benefit for humanitarian responders and MNOs. Consider how creating a ‘competition’ between MNOs could work in your pricing favour! 

Certain organisations such as Human Network International have invested a lot of effort in brokering good relationships with MNOs to support their informational services. In fact, research shows investment in such information provision schemes can actually enhance customer retention and have an impact on the MNO’s revenue. When such potential exists, having this frank discussion can leverage outside resources and help facilitate the delivery  of a two-way SMS solution.

10. Start small and test your solutions before deciding whether you want to DIY, or contract a service provider

“So we’ll just give this company $200,000 and it will happen…”

When you’ve developed an understanding of the capacities of staff and partners and the workflows of inbound and outbound messaging, you’ll start to get a feeling of what kind of solution you’re looking for. To start, rather than making big commitments or investments, test the platforms out. Our video shows you how you can get started with Frontline SMS in 2 minutes. You may also want to approach a service provider for a ‘trial period’ of their service, which they often provide to potential customers. This will help you reject options quickly that obviously don’t fit your requirements. It’s also an opportunity to take into account the less tangible aspects like ‘ease-of-use’ etc. 

Following this testing, you can work out whether you will need a Do-It-Yourself (DIY) solution, or to contract a service provider, or something in-between. You can take the results of your testing, along with your earlier mapping of workflows and see how you can map these workflows to the specifications of the various different systems trialled (they will have differences in their features / specifications.


When you get to this stage, start to test out things that might fit your specific use case. Check out our blog on designing solutions for some more tips on how to go about delivering these solutions.