Research Methodology
Click on each channel to learn more about our research methodology
A2P SMS forecast methodology
Our data is based on extensive research over an on-going period, starting back in 2010, and has continued ever since. Between July 2022 to June 2023 we interviewed/surveyed/received data from almost 300 companies (including 160 mobile operators and mobile operator groups, and over 140 messaging ecosystem companies, including aggregators, CSPs, solutions providers, firewall providers, independent consultants and regulators). Throughout 2022/3, we connected with 200+ companies, with the main research for this report kickstarting in 4Q 2022. By the time Mobilesquared closed the research we had connected with 225 companies, based on the companies listed above.
Mobilesquared applies a three-pronged research strategy for A2P SMS.
Online surveys
Mobilesquared developed online surveys for each business messaging channel, running continuously via the company website. For A2P SMS, we have developed two online surveys, one for mobile operators monetising A2P SMS traffic, and a separate survey for the broader A2P SMS messaging ecosystem. In both surveys, companies not monetising A2P SMS traffic, or have no intention of monetising A2P SMS traffic, are screened out.
1-2-1 and 1-2-many interviews
Mobilesquared uses its extensive reach throughout the messaging ecosystem to interview companies operating within, or whom have knowledge and insight of, the A2P SMS landscape.
Data sharing
Mobilesquared shares existing data with clients and non-clients. We have over 70 A2P SMS clients and the majority provide feedback, data, trends, growth and insight on our (at the time) existing A2P SMS data and projected growth, based on their view of the industry. For clients we shared key data points from 2022 in order to receive their market data for 2022, and expectations for 2023. For non-clients we shared data points from 2021 and asked for their market data for 2021 and 2022, and their expectations for 2023.
Where requested, Mobilesquared has entered into NDAs with companies as part of a data exchange.
All data and insight is then processed, anonymised and aggregated (where there is input from multiple sources), modelled and sense-checked.
Mobilesquared updates its business messaging forecasts on an annual basis, with smaller updates occurring throughout the year.
Our industry research is supplemented by consumer research, updated on an annual basis, and enterprise research which is updated on an ad hoc basis. Mobilesquared last conducted consumer research in 1H2022, and this data will feed into the A2P SMS engagement data to be included in the 2023 updates. The consumer research was based on 10,000 people across 10 markets (Australia, Brazil, Colombia, France, Germany, Malaysia, Mexico, Singapore, UK, and USA). Consumer research forms the basis of our opt-in user data. Mobilesquared has supplemented consumer research with actual market data from partners to cover additional markets. Where we do not have national specific data we have applied a regionalised average per market.
A2P SMS forecast methodology
The data in our global A2P SMS messaging forecasts is based on data modelling (traffic or pricing), whereby Mobilesquared takes an average where two or more sources of actual market data were available. Across the 200 markets tracked by Mobilesquared, forecasts by market are based on an average of 18 data points. As a result of the greater input, we have been able to model each of the 200 markets individually for market size, split of international and domestic traffic, and growth over 2019, 2020, 2021, 2022, and into 2023, and throughout the forecast period up to 2027.
The starting point for these A2P SMS messaging forecasts is our mobile operator subscriptions database. We research the top 200 markets worldwide collating information from regulators, mobile operators and mobile operator groups. Data-only contracts and M2M accounts have been excluded (where available). Subscription forecasts are modelled based on previous growth records, overall mobile penetration rates and expected population growth.
Average A2P SMS pricing and pricing forecasts for the 200 markets covered (termination and wholesale rates) are based on country-by-country information obtained from mobile operators, mobile operator groups, aggregators, cloud service providers, and hubs.
Traffic and traffic growth forecasts for the 200 markets covered are based on our ongoing interviews, discussions and online surveys, looking at total volumes of authorised and unauthorised traffic, and the number of messages received per subscriber per month. The level of unauthorised traffic has also been applied according to our research into the per mobile operator deployment of SMS firewalls and SS7 protection, because of Information provided from firewall vendors and solutions providers on the rollout and adoption of SMS firewalls. This has been further segmented from 2023 research (and historic research between 2018-2021) to breakout the new traffic types (subsets of white traffic and grey traffic) based on differing fraud types: blue, orange and red.
Mobilesquared has made the assumption that approximately 5% of each operator’s A2P traffic will always be grey-route due to the nature of companies looking to expose and exploit loopholes, and approximately 10% will always be white-route due to direct agreements between mobile operators and the enterprise.
We have assumed that where no SMS firewall is in place that at least 80% of traffic is grey. We have also assumed there will be varying rates of protection where SMS firewalls are in place due to differences in managed solutions, hosted solutions, on-premise solutions, or blended solutions. Mobilesquared research reveals SMS firewalls typically block 80%-90% of all traffic, and this has been applied as the basis when modelling fraudulent traffic.
We have assumed that no exclusivity agreements of preferred partner deals are in place per mobile operator, unless Mobilesquared has received data from industry through the research process of such a deal. At the time of publishing the 2023 update, over 200 known deals were in place, and these have been taken into account when looking at pricing, and the varying types of traffic activity that can associated with such deals.
Domestic and international traffic volumes, messages by sector and associated forecasts are based on our ongoing interviews, discussions and research (including detailed survey information conducted throughout 2021, 2022, and 2023). Sector traffic, and sector traffic split out by use case, has previously only been attributed to white route messaging, but from 2023, this will be applied to white/grey, international and domestic, and retrospectively applied back to 2017 using historic data.
Pricing data for each mobile operator covered in this Databook was received from multiple sources – both discount and premium messaging providers, as well as mobile operators and operating groups.
Pricing
The basis for all international spend are the latest international termination rates (as of 2Q2023) for every mobile operator. Where we have received the latest domestic rates, these have been applied. Consequently, we have used three separate rate cards for our pricing:
A discount international rate for large-volume providers – with an approximate 10-20% mark-up on the cost of purchase from mobile operators in 2023 (we applied an average 15% mark-up).
A premium international rate – with an approximate 30-40% mark-up on the cost of purchase from mobile operators in 2021 (we applied an average 35% mark-up)
A domestic rate (where available) – typically 20-33% of the relevant international rate for discount or premium providers with the same mark-up applied.
We have assumed approximately 90% of non-direct traffic between mobile operators and brands is discounted, and the remaining 10% is premium.
We have also assumed that 5-15% of all traffic delivered to a mobile operator will be the result of a direct relationship with brands, priced without the above mark-up.
Large mobile operating groups, and mobile operators in mature markets, are likely to see more direct traffic than smaller, independent mobile operators, in developing markets.
Input on the cost of grey route traffic varied from 15-70% of the discount rate. We have applied a rate of 46%.
All monetary values show the actual income for each part of the value chain – mobile operator or aggregator. Total spend is what a brand pays for messaging with the income from that then split (where appropriate) between mobile operator and aggregator.
Mobile operator white route revenues are calculated based on varying small degrees of direct spend, according to our ongoing interviews, discussions and research (including detailed survey information conducted in the summer of 2021). The majority of mobile operator revenue is based on the average wholesale rates paid by aggregators.
Aggregator white route revenues are calculated based on the difference between the average termination and wholesale rate per market. Mobilesquared has allocated aggregators 100% of grey route revenues, even though a proportion could be attributed to mobile operators from, for example, the sale of SIM cards used in SIM farms. SIM farm traffic is included as grey route traffic in our data.
Artificially Inflated Traffic
Artificially Inflated Traffic (AIT) (aka labelled blue traffic by Mobilesquared) has been modelled using a spend-down approach, and these have been calculated on a country and mobile operator basis. Please note, only the country data will be released by Mobilesquared for the foreseeable future because of the sensitive nature of the traffic. AIT spend per market has only been applied to international traffic (based on research), and has been calculated on the following:
Mobilesquared’s understanding of the traffic split/messaging spend by the top 30 senders of international traffic;
the trusted assumption that these top 30 companies account for 80% of total traffic;
spend data from Twitter, and other OTT providers, based on insight shared with Mobilesquared.
Market-by-market data shared by industry
Analysis of historic research reveals AIT has always been present but held a minimal subterranean presence in terms of threat level.
To generate the forecasts, based on research, Mobilesquared has assumed AIT occurs in every market with vastly differing degrees of activity. We have assumed that where the international termination rate is low, AIT activity is extremely restricted but is present nonetheless. Where the international termination rate is high, we have applied high levels of AIT activity.
Traffic volumes were derived from dividing the total spend per market by the international termination rate cost per mobile operator, or the average international termination cost per market. Please note, the methodology to generate AIT forecasts is focused on spend, with traffic levels an output of this spend total. Mobilesquared acknowledges that this approach does not take into account the discounts negotiated by the major brands sending bulk messages globally. Because Mobilesquared believes the spend figure is an accurate reflection of AIT spend throughout the forecast period, Mobilesquared clients could apply known discounted international termination rates to get (what could) be a more accurate reflection of traffic levels on a market-by-market basis.
Please note II, AIT (blue traffic) has a permanent presence throughout the traffic section of this report, and is clearly stated where it is featured. However, due to the uncertainty and inconsistency in the data shared by industry relating to where brand spend in AIT is (and should be) allocated in the value chain, Mobilesquared has only applied AIT to total spend on a market level basis, and its inclusion within the A2P SMS datasets and report has been restricted to market spend data associated with fraud, such as the “Harmful” view of the industry. Where Mobilesquared has split out spend by value chain, we have removed blue (AIT) spend altogether, i.e. section 4ii of the report that looks at spend on different traffic types across the value chain. Where AIT has not been applied in the spend forecasts, it will be clearly labelled as such.
For dark blue traffic (AIT delivered to a consumer), Mobilesquared has applied a percentage of total traffic from the light blue traffic (AIT blocked) based on data from industry. However, because of the miniscule number of messages delivered, Mobilesquared has sidelined this traffic type for the time being.
Potential/Lost revenue
Our calculations for potential income treat all traffic (both white and grey-route) as official white-route, with pricing as above.
The lost income is therefore the difference between the actual, and the potential, income.
RCS business messaging (RBM) methodology
Our RCS data is based on extensive research over an on-going period, starting back in 2010, and has continued ever since.
Mobilesqurared has been tracking RCS since 2011, and our first forecasts were in 2017. This is our sixth update.
Mobilesquared applies a three-pronged research strategy for acquiring and attaining actionable data and insight on RBM.
- 1-2-1 interviews and discussions.
- Online surveys
- Industry events
Where requested, Mobilesquared has entered into NDAs with companies as part of a data exchange.
The data and insight provided as part of the research is then processed, anonymised and aggregated (where there is input from multiple sources), modelled and sense-checked, and applied to our RBM forecast model.
Our industry research is supplemented by consumer research, updated on an annual basis, and enterprise research which is updated on an ad hoc basis.
Mobilesquared updates its RCS business messaging forecasts on an annual basis, with smaller updates occurring throughout the year.
Between October 2023 to September 2024, we interviewed/surveyed/received data from almost 300 companies (160 mobile operators and mobile operator groups, and 131 messaging ecosystem companies, including aggregators, CSPs, solutions providers, firewall providers, and regulators) covering all aspects of business messaging. While all of these companies have provided valuable insight into business messaging, one-third have specifically provided data and insight on RCS and RCS business messaging.
Networks & users
Live RCS networks and registered RCS users are based on ongoing research of the RCS ecosystem by Mobilesquared. RBM users are based on the reported launch of RCS Business Messaging services and device client availability in relevant markets, supplemented by RBM provider RBM device capability check data.
RCS and RBM network forecasts and MAU forecasts are based on Mobilesquared research encompassing expected mobile-operator group rollouts, the expansion of competition in more mature markets, the availability of device clients and the launch of business messaging services by relevant mobile operators.
Due to the discrepancy between those markets that have launched RBM, and those markets that have launched RBM and developing a commercial proposition, Mobilesquared has determined that markets generating less than $500,000 per annum are still in “experimental phase” and yet to commercially launch RBM. Markets that are generating brand spend in excess of $500,000 are commercially active.
All networks and user data supplemented by data from MEF Data, Google, and other key RBM companies.
Traffic
Our total RBM engagement forecasts are based on the 5 billable events introduced by Google:
- Basic Message
An application-to-person (A2P) message that is up to 160 characters and contains only text. Can also be called Branded Message. Conversational agents only: if a P2A message is delivered as a response to an A2P message within the next 24 hours, the Basic message becomes part of an A2P conversation; otherwise, the session terminates. A Basic message is always delivered from an agent to a user.
- Single Message
An application-to-person (A2P) message that contains multimedia or text with more than 160 characters. Conversational agents only: if a P2A message is delivered in the next 24 hours, the Single message becomes part of an A2P conversation; otherwise, the session terminates. A Single message is always delivered from an agent to a user.
- A2P Conversation
Applies to conversational agents only: An A2P conversation is initiated when a P2A message is delivered within 24 hours of an A2P Single message or A2P Basic message. Note that if a P2A message is delivered within 24 hours of multiple A2P messages, only the A2P message that immediately preceded the P2A message is used to create the conversation session. This A2P message, along with any messages delivered within the next 24 hours, are part of the new A2P conversation.
- P2A Conversation
Applies to conversational agents only: A P2A conversation is initiated when there is no active session (A2P Single message, A2P conversation, or P2A conversation) and a P2A message is delivered, and the business responds within 24 hours.
- P2A Message
1.1. Non-conversational agents: A P2A message sent from a user towards an agent with a billing category of single message or Basic message.
2.2. Conversational agents: A P2A message sent from a user towards an agent where there is no existing conversation, and the agent does not return a reply.
To ensure consistency across other channels (to allow direct comparisons and analysis with WhatsApp Business API, for example), Mobilesquared has also applied the four WhatsApp Business use cases of “authentication”, “marketing”, “utility”, and “service” to RBM and applied to the RBM billable event types outlined above.
The traffic has been split by billable events based on market research. It has then been further segmented by use case based on the same market research.
In addition to the traffic data and insight shared by industry, Mobilesquared has also modelled the traffic by use case by the underlying A2P SMS traffic that we believe is most at risk from migration to RBM. In essence, this is the authentication traffic pushed out over white routes – that is to say official, billable business messaging.
Volumes for both chat-based A2P and P2A sessions show only the actual number of engagements. In terms of billing, a session in 2023 has been modelled on an average of four messages (exchanged between brand and consumer) that will be charged to the brand or business at an average rate of 2 x RCS event message termination cost. By 2028 the average number of messages in a session has increased to six.
SPEND
As part of the methodology, Mobilesquared has based RBM spend forecasts on the following:
- Basic message = 1 x SMS domestic rate per market
- Single message = 1 x SMS domestic rate per market + 30% premium
- P2A message = 2 x SMS domestic rate per market
- Conversation = 2 x SMS domestic rate per market.
Basic message spend is based on basic message traffic per market multiplied by the cost of 1 x domestic rate per market.
Mobilesquared has applied one rate per billing event per market, i.e. we have not replicated the A2P SMS model by having different domestic and international rates, as is the case in a significant number of markets. Therefore, when we compare RBM rates with WhatsApp Business rates, the RBM rates are based on the domestic SMS rate for that market.
Conversational traffic as applied by Mobilesquared has been categorised as customer care.
Mobilesquared defines P2A traffic as the engagement or interactive component of RBM and can be applied in two ways. Firstly, when a consumer sends a message to a brand and potentially initiates a conversation. Secondly, when a consumer interacts with an A2P message from the brand. The majority of P2A traffic over the forecast period is based on the latter, when a consumer interacts with an A2P message from the brand.
Mobilesquared has not included Google in the revenue share for basic message revenue. Google’s revenue share kicks in with single message, P2A and conversational traffic.
For the time being, Mobilesquared has applied the following splits when calculating RBM spend forecasts:
For basic message, mobile operators will receive 100% of the cost per message, while aggregators will receive their mark-up, which varies from region to region.
For single message, P2A and conversational traffic spend, all mobile operators receive 79.8% of the cost of the message, Google receives 20.2%, while aggregators will receive their mark-up, which varies from region to region. Mobilesquared will adjust this revenue share split as and when additional information is shared.
We have assumed that mobile operators are not selling direct to brands, therefore, all RBM spend will be subject to a revenue share as outlined here.
Mobilesquared has also assumed that all markets will become carrier messaging markets, i.e. no markets will remain Google Guested. This also means Apple’s support of iOS will be applicable to every market. Mobilesquared will adjust this market view as and when additional information is shared.
Our spend forecasts are solely based on the delivery of RBM, and do not include any value-added services delivered by an aggregator, messaging or marketing platform – such as service notifications, reporting or integration with business applications, or distribution fees charged by an aggregator to a messaging or marketing platform.
Spend and income is based on commercial campaigns and does include trials where applicable and the relevant information has been shared with Mobilesquared.
Prior to June 2023, Mobilesquared has used the previous pricing models and rates.
APPLE
To understand the impact that iOS’s support will have on RCS, Mobilesquared has developed two sets of forecasts: Android only and Android + iOS. To differentiate the two throughout the copy, any data based on Android and iOS have been labelled “RCS-i” or “RBM-i”.
Apple’s support of RCS (RCS-i) will only be for “carrier messaging” where a mobile operator has partnered with Google (i.e. therefore offering RBM-i). In markets where carriers decide not to partner with Google, they will remain a Google Guest-only market, offering RBM as an OTT (Over The Top) service akin to WhatsApp Business. For the time being, Mobilesquared has assumed that Apple will support RCS in every market by the end of our forecast period.
Initially, (at the time of publication, September 2024), Mobilesquared research confirmed that seven markets will have RCS switched on by Apple by the end of 2024:
“The Seven” markets are: Belgium / Canada / France / Germany / Spain / UK / USA.
iOS’ support of RCS is expected to be extensively rolled out from 2025.
Mobilesquared has split out the markets that have launched RBM between markets that are actively generating revenues from repeat campaigns from brands, and markets that are still only attracting experimental budgets from brands. Mobilesquared has assumed that any revenues below $250,000 per annum per market should be considered as brand experimental spend, and therefore RBM has not commercially launched in that market.
One further consideration when looking at the impact on the Apple announcement on RCS, is that although iPhone users are increasingly more likely to update their device’s software to the latest version compared to Android users, Apple data reveals that 30% of iPhone users update to the latest software within 3 months, with approximately 65-70% of iPhone users within 6 months, and 80% within 9 months. This indicates that the impact Apple has on RCS reach for RBM will be staggered over a period of at least 12 months.
Given the slow adoption of RCS to date, and the lack of RBM information released by Apple regarding how it will support “RBM-i”, Mobilesquared has applied a conservative software update process to the forecast model. In developed markets, Mobilesquared has assumed 10% of iOS devices will update software in the first year, 40% in the second year, 80% in the third year, and 100% of devices by year 4. In less developed markets the iOS software update user numbers have been capped between 40% and 80%.
As there is no data based on how iOS users will behave differently over RCS, to add the iOS traffic uplift to RBM traffic has been calculated on the average number of messages per Android RCS user per annum per market, and this average per RCS user has been applied to Android and iOS users.
Mobilesquared understands that Apple will not be included in the RBM value chain, and therefore not receive a share of brand spend. However, that is not to say Apple will not enter into direct agreements with mobile operators providing “carrier messaging” to access the value chain via the side door.
iOS spend uplift is calculated in the same way Android-only spend is calculated, with traffic based on Android + iOS RCS (RBM) users.
As more information is shared with Mobilesquared regarding all aspects of RBM, we will update our RBM forecasts accordingly.
WhatsApp Business API methodology
Our data is based on extensive research over an on-going period, starting back in 2010, and has continued ever since.
Mobilesquared applies a three-pronged research strategy for all business messaging research.
- Online surveys
Mobilesquared developed online surveys for each business messaging channel, running continuously via the company website. For A2P SMS, we have developed two online surveys, one for mobile operators monetising A2P SMS traffic, and a separate survey for the broader A2P SMS messaging ecosystem. In both surveys, companies not monetising A2P SMS traffic, or have no intention of monetising A2P SMS traffic, are screened out.
- 1-2-1 and 1-2-many interviews
Mobilesquared uses its extensive reach throughout the messaging ecosystem to interview companies operating within, or whom have knowledge and insight of, the A2P SMS landscape.
- Data sharing
Mobilesquared shares existing data with clients and non-clients. We have over 70 A2P SMS clients and the majority provide feedback, data, trends, growth and insight on our (at the time) existing A2P SMS data and projected growth, based on their view of the industry. For clients we shared key data points from 2022 in order to receive their market data for 2022, and expectations for 2023. For non-clients we shared data points from 2021 and asked for their market data for 2021 and 2022, and their expectations for 2023.
Where requested, Mobilesquared has entered into NDAs with companies as part of a data exchange.
All data and insight is then processed, anonymised and aggregated (where there is input from multiple sources), modelled and sense-checked.
Mobilesquared updates its business messaging forecasts on an annual basis, with smaller updates occurring throughout the year.
Data and forecasts are based on Mobilesquared’s ongoing research into business messaging. It is worth noting that the use of “brand” and “business” is interchangeable by Mobilesquared throughout this methodology (and also the main report).
Research for the WhatsApp Business dataset and report are based on online surveys, 1-2-1 interviews and discussions, with 105 WhatsApp Partners, CPaaS providers, messaging specialists, and mobile operators. This research is supplemented with broader messaging data and insights conducted by Mobilesquared, covering areas such as CPaaS, and messaging traffic migration, with an additional 182 companies.
The latest WhatsApp Business research specifically for this report and dataset was conducted between March to October 2023, with additional information and insight from industry provided up to March 2024.
The WhatsApp Business datasets cover both API traffic and spend, and app traffic. The WhatsApp Business API is a commercial service for medium and large businesses only accessible via an API either directly with WhatsApp or with a WhatsApp Partner (aggregator). The WhatsApp Business App is a free service for small and micro businesses only, that have downloaded and are actively using the app.
Business size is based on micro (1-9), small (10-49), medium (50-249) and large (>250).
The markets operating WhatsApp Business API services, or projected to launch during the forecast period, are based on the information shared in the online WhatsApp research and 1-2-1 interviews with WhatsApp Partners during the research process.
The total number of businesses which are using WhatsApp Partner APIs is based on surveys of the WhatsApp Partners starting in the first quarter of 2020 ending October 2023. Where business-size information was unavailable, we estimated the total number of businesses based on a percentage of total registered businesses divided by medium and large businesses (this data also contains micro and small business information). This data is based on publicly available data from the World Bank, Organisation for Economic Co-operation and Development (OECD), and modelled by Mobilesquared based on additional third-party data to create a market breakdown of business size.
In markets with a low WhatsApp presence (typically less than 1 million users), or a low penetration of WhatsApp users as a percentage of total unique mobile users (less than 30%), Mobilesquared has made the assumption that WhatsApp Business API commercial services will not be launched; unless the research, specific market data, or WhatsApp Partner information, stated otherwise.
Users
Total WhatsApp user numbers, and total WhatsApp user forecasts, are based on various public announcements dating back to December 2013 – most recently crossing 1.5 billion in February 2018, and 2 billion in February 2020, and insight shared during the research process by WhatsApp Partners taking the total number of 2.7 billion.
Historical WhatsApp users were calculated for each year-end and benchmarked against total smartphone penetration to assess actual, and smartphone-penetration, growth trends. WhatsApp user and penetration numbers were then researched and collated for the individual countries and regions contained in our forecasts. Where country information was not available, estimates were benchmarked against regional WhatsApp smartphone penetration.
Country and regional users were then aligned with the global total and forecast throughout the period based on actual registered users, unique mobile users, smartphone-penetration, and growth trends.
Mobilesquared applied WhatsApp Partner/CPaaS industry research in 2020, 2021, 2022, and 2023, as well as enterprise survey data conducted during 2019-22, to ascertain the average size of the customer database for micro, small, medium and large enterprises. In addition, this information has been supplemented with ongoing research of brand database development and maintenance, as well as extensive consumer research conducted by Mobilesquared between 2018 and 2022, with the last round of consumer research based on 10,000 consumers across 12 markets.
Mobilesquared has used this information to create an average database opt-in by business size (i.e. medium and large), and applied the year-on-year growth data supplied by industry. Mobilesquared then applied this data to average population data to convert the opt-in database to a percentage of total population, and applied this percentage to the total population of each market.
Mobilesquared has therefore created an average Year 1 database size, a Year 2 database size, a Year 3 database size, and so on.
The WhatsApp database per brand is also referred to as user opt-in, or WhatsApp Business Consumers, to ensure clarity between a business user opt-in and a WhatsApp user that has not opted in to receive any business communications.
In addition, the extensive consumer research, in conjunction with the ongoing WhatsApp Partner research, has allowed Mobilesquared to further segment the opt-in database by sector.
WhatsApp Business consumers by sector have been calculated by applying the ongoing research data to the same principles applied to total population per market and projected growth, the consumer research allied to the WhatsApp Partner research has then been applied to break the opt-ins out by sector. Where country information was not available, estimates were benchmarked against regional WhatsApp smartphone penetration, users, total market opt-ins, and applied to WhatsApp opt-ins.
The WhatsApp Partner research also provided information on average messages sent per brand based on the databases size, broken out by medium and large brands, on an annualised basis. In effect the average database size per market represents the average number of messages sent per market.
Businesses using WhatsApp Business App
The total number of small and micro businesses which have downloaded and are using the WhatsApp Business App is based on various public announcements dating back to January 2018. By mid-2019, around 8 million small businesses were believed to have downloaded the WhatsApp for Business app and, as of January 25th 2020, Google Play stated that the app had been downloaded over 100 million times.
However, it was also announced that only 3 million businesses were actually using the app as of August 2018, rising to 5 million at the end of 2019. In July 2020, then WhatsApp (now Meta) announced it then supported more than 50 million WhatsApp Business App users. In June 2023, Meta announced that the WhatsApp Business App had 200 million monthly active users.
To ensure consistency in the methodology between the medium and large businesses using the API and small and micro businesses using the App, Mobilesquared has applied the number of businesses using both WhatsApp versions to registered businesses as outlined above. We have then reduced the total number of active WhatsApp Business App users with total micro and small registered businesses. Where the number of WhatsApp Business App users exceeds 100%, Mobilesquared has assumed that these are unregistered and unverified business user accounts.
Use case traffic
Data and forecasts contained in this report are based on the four use cases introduced by Meta to WhatsApp Business API in June 2023. The use cases have been retrospectively applied to historic data based on relevant and applicable industry data and insight. Use case forecasts are based on data and market expectations shared by WhatsApp Partners during the research process. Where previous WhatsApp Partner research is applicable in relation to use case usage, this has also been applied.
Total traffic
Total traffic is based on actual industry data provided by WhatsApp Partners to Mobilesquared up to end-2023. Additional data and information provided by WhatsApp Partners and non-Partner companies has enabled Mobilesquared to breakout the traffic data by market by the top 10 markets. Beyond the top 10 markets, where information provided was more generic, Mobilesquared has applied the WhatsApp partner data.
To create these traffic totals in the model, Mobilesquared has multiplied the number of businesses per market using WhatsApp Business API by the average database size, and applied the growth on an annualised basis.
The migration of A2P SMS traffic onto WhatsApp Business API has also been applied in the total traffic data. Mobilesquared has calculated using two models. Firstly, we created the “Alternative View” of A2P SMS when updating our A2P SMS forecasts in 2023, based on the industry sustaining a stable approach to traffic growth and price increases as demonstrated between 2017 and 2021 and then trending the growth throughout the forecast period to the end of 2027. The Alternative View of the A2P SMS market is to highlight how the market might look if the “three issues” had not arisen and the market maintained strong growth from the previous years. The three issues that have blighted the A2P SMS industry in recent years are exponential increases in international termination rates in a number of markets globally, the rise of exclusivity agreements, and the increase in fraud.
The Alternative View creates lost A2P SMS traffic, and based on the industry research, a percentage of this lost traffic has been modelled to migrate over to WhatsApp Business API traffic, and split out by market.
As a result of these three issues identified above, there has been a significant drop in international A2P SMS traffic in the second half of 2023, primarily one-time password (OTP) traffic and Mobilesquared has modelled a percentage of this dropped traffic onto WhatsApp Business API, specifically applying to markets where there are high international SMS termination rates.
Traffic breakdown
Mobilesquared’s WhatsApp Partner/CPaaS industry research in 2022 and 2023, provided a market breakdown of traffic, split out in multiple ways:
- By use case (authentication, marketing, service, and utility)
- By category (A2P, P2A, and conversations)
- By conversational service (customer care) & conversational commerce
- In addition, the data provided also provided a detailed view on interaction levels between a consumer and brand.
We also applied consumer survey data conducted during 2019-22, to ascertain the average number of times a consumer would message a brand each year (P2A traffic).
Taken together, the enterprise and consumer survey data together formed the basis of calculations for all A2P and P2A engagement over WhatsApp Business API users. We have assumed an average of five messages will be exchanged in both A2P and P2A conversations rising to 8 after the fourth year of commercialising WhatsApp Business API traffic per market.
We have assumed a response rate of 8% in 2021 to A2P messages which, in turn, lead to conversational sessions, rising to an average of around 25% in 2027, based on WhatsApp Partner growth expectations.
WhatsApp pricing
WhatsApp introduced use case pricing in June 2023. Mobilesquared has based the WhatsApp Business API spend on the published rates. The rates do not take into account discounts provided by Meta to brands based on direct connections, and by the WhatsApp Partners (aggregators).
Prior to June 2023, Mobilesquared has used the previous pricing models and rates.
Conversations between a brand and a consumer are charged at one message per day maximum. If the conversation continues over the 24-hour period an additional message fee will be charged to the brand.
As outlined by Meta, it is possible to have multiple open conversations between you and a customer. This can happen in the following situations:
An open marketing or utility conversation exists between the for 24 hours.
When a brand sends a message to a customer (which can only be done if a customer service window exists between you and the customer), Meta will check if an open conversation — of any category — already exists between the brand and the customer. If one exists, no new conversation is opened. If a conversation does not exist, a new service conversation is opened, lasting 24 hours.
For example (based on Meta wording):
Hour 0: You send a targeted promotion (marketing template) to a customer. No open marketing conversation exists between you and the customer, so a marketing conversation lasting 24 hours is opened.
Hour 4: The customer messages you. This opens a customer service window between you and the customer, allowing you to send them free-form messages for the next 24 hours.
Hour 5: You send a free-form message to the customer. An open conversation already exists between you and the customer (a marketing conversation in this case), so a service conversation is not opened.
Hour 24: The marketing conversation expires.
Hour 25: The 24-hour customer service window is still open, so you send a second free-form message to the customer. No open conversation exists between you and the customer anymore, so a service conversation is opened, lasting 24 hours.
Hour 26: The 24-hour customer service window is still open, so you send a third free-form message to the customer. An open service conversation already exists between you and the customer, so a new service conversation is not opened.
(Source: https://developers.facebook.com/docs/whatsapp/pricing/)
Meanwhile, WhatsApp Partners all have different pricing strategies and price points for both A2P and P2A messaging, varying from a monthly license fee to a mark-up rate only. Some WhatsApp Partners charge a flat monthly admin fee, some do not. Some include a P2A message volume within the fee, some do not. Some charge per user, some per message, and most seem to charge differing message rates – although their rates apply to both P2A conversations and A2P templates.
Mobilesquared has modelled the following:
Aggregator spend throughout the dataset has only been applied to “actual billable” messaging traffic, whereby an average mark-up rate of US$0.005 per message sent (A2P or P2A) or conversation has been applied for the period 2019 to 2022, an average rate of US$0.0045 per message/conversation in 2023, and an average rate of US$0.004 per message from 2024 onwards. This mark-up is payable to the WhatsApp Partner.
CPaaS
Mobilesquared are messaging specialists, and that is why we have split out as many messaging channels as possible within our CPaaS modelling.
Our data is based on extensive research over an on-going period, starting back in 2010, and has continued ever since. Between September 2020 to October 2022 we interviewed/surveyed/received data from almost 330 companies (163 mobile operators and mobile operator groups, and 167 messaging ecosystem companies, including aggregators, CSPs, solutions providers, firewall providers, and regulators) covering all business messaging. In addition, specific CPaaS data and insight across all 14 CPaaS channels has been generated from CPaaS-only related research based on interviews, discussions and online surveys with 113 companies, a large % of which we had already spoken to for the messaging research, but also CPaaS specialists not from a messaging background.
Mobilesquared developed online surveys for each business messaging channel and CPaaS, running continuously via the company website, and promoted to our research database and relevant industry contacts, as well as garner the support from trade bodies such as the Mobile Ecosystem Forum. Across every business messaging channel and CPaaS survey, participants are asked to answer for all regions where they operate, and identify the key markets for them. Each survey typically asks the respondent to identify channels used, traffic levels, expected growth, traffic split by market vertical and use case, spend, unit pricing, and unit pricing change. It is worth noting, that each survey is also tailored to that specific channel or channels (in the case of CPaaS).
Mobilesquared updates its business messaging forecasts (and non-messaging channels included in CPaaS) on an annual basis, with minor updates occurring on an ad hoc basis throughout the year.
Our industry research is supplemented by consumer research and enterprise research, both of which are updated on an ad hoc basis.
The consumer research included in CPaaS is based on the views of 10,000 consumers in 10 markets (Australia, Colombia, France, Germany, Malaysia, Mexico, Singapore, Spain, UK, US), conducted between November 2021 and February 2022.
The brand data included in the CPaaS report is based on data supplied by CPaaS providers, which has then been aggregated and anonymised.
Once all of the research process is complete, the messaging data and insight is then processed, anonymised and aggregated (where there is input from multiple sources), modelled and sense-checked, and applied (where applicable) to the following business messaging channels; A2P SMS, RCS, MMS, WhatsApp Business, Apple Messages for Businesses, Facebook Messenger, OTT messaging apps, and Viber.
This process has been repeated for the non-messaging channels included in our CPaaS datasets, however, the data has greater reliance on information and insight shared during the interview process, and the data shared in our online CPaaS survey – which had 75 respondents.
CPaaS forecast methodology
Our forecasts are based on a top-down, bottom-up methodology. Revenues for the top 38 CPaaS providers (as identified by Mobilesquared) are based on published revenues for 2019/20 and 2021 from the CPaaS provider either directly or from a reputable third-party, up to the end of 2Q 2022.
Mobilesquared has modelled this data on our market growth projections to generate CPaaS provider revenues for the remainder of the forecast period based on each company’s channel offering, customer base and growth, geographical reach, and company data (where applicable). Mobilesquared has also applied market research, market information and insight shared during the 18-24 month research process when modelling each company’s projected revenues.
In addition, to complete our total view of the CPaaS landscape, we have assumed the 80:20 rule, with 80% of total CPaaS revenues being generated from these 38 CPaaS, and the remaining 20% of revenues coming from the long tail of smaller, national-based CPaaS providers appearing in markets around the world.
Given that Mobilesquared is a messaging specialist, we then segmented our CPaaS forecasting process into messaging-based, and non-messaging based. For the messaging channels, Mobilesquared has applied its existing market forecasts for both traffic and spend to the CPaaS data. Details of each messaging channel methodology is outlined below.
With our total global view of CPaaS spend and growth forecasts created, we then applied the relevant messaging data spend to the CPaaS global spend. To segment the remaining global spend, we applied our CPaaS research to break out the spend by non-messaging channel and by market. This process was based purely on the data provided via the online survey from the 75 respondents and sense-checked during the 1-2-1 interviews.
Where total non-CPaaS global market spend per channel has been included (i.e. Voice, Email, Video, Social media, WebRTC, and In-app push notifications), figures included in the report are based on third-party data from Statista, F&F Research, IAB, MMA, Brandessenceresearch, Data Library Research, as well as anonymised third-party data from research participants. Where there is multiple global forecast datapoints per channel, Mobilesquared has taken an average. Mobilesquared has then applied the CPaaS research to take a % of the global spend to apply to channel spend within CPaaS. The online survey data has then been applied to segment further by country.
Mobilesquared has calculated CPaaS traffic by applying a global average cost per action to each channel. “Action” has been used throughout the CPaaS report as an umbrella term to cover a voice call, video, social media brand advert, promotion or update, a single in-app push notification, an email, even a single message and so on. Voice traffic has been calculated by applying a global average cost per call to a contact centre to the total spend. For all other non-messaging channels, with the commercial model for these channels based on a licensing model encompassing a set number of actions sent to a pre-determined sized database per month, Mobilesquared has calculated an average cost-per action based on available market pricing data from the major CPaaS providers, and where required, from the major non-CPaaS providers, i.e. company specialists in each sector.
CPaaS users are based on how each user can access every service. Mobilesquared has assumed that the smartphone and feature-rich non-smartphones have the capability of accessing all 14 channels, and have formed the basis of our CPaaS users. In each market we have applied 100% of smartphone users + a % of non-smartphone users.
When calculating the average spend per channel per user, and average traffic (action) per channel per user, Mobilesquared has applied the CPaaS user total. It is important to note, that this figure will greatly differ compared to if the average spend per channel per user, and average traffic (action) per channel per user was Viber or RCS/RBM and based on Viber or RCS/RBM only users respectively, for example.
Please note, this is Mobilesquared’s first view of the CPaaS marketplace. While we are confident that our data and forecasts accurately reflect the data and insights shared during the insight process, we acknowledge that as we get more visibility into the CPaaS providers (those included in this report and those yet to feature) we will amend our forecasts accordingly. Mobilesquared views this as an on-going process, in much the same way we have done so with A2P SMS, RCS, and WhatsApp Business, and now viewed as the leading provider of business messaging (and CPaaS) intelligence.
A2P SMS forecast methodology updated for CPaaS
Based on Mobilesquared’s research, we have assumed that 62% of total A2P SMS global spend was via a CPaaS provider in 2020, rising to 71.7% in 2022, and 77.19% in 2026, for each year of the forecast period we have then weighted this accordingly to the size and projected growth of each market when applying SMS traffic and spend to our CPaaS data.
The A2P SMS data included in CPaaS is based on data modelling (traffic or pricing), whereby Mobilesquared takes an average where two or more sources of actual market data were available. Across the 200 markets tracked by Mobilesquared, forecasts by market are based on an average of 18 data points. As a result of the greater input, we have been able to model each of the 200 markets individually for market size, split of international and domestic traffic, and growth over 2019, 2020 into 2021, and throughout the forecast period. However, only total white route traffic has been included as CPaaS traffic, and is the total of the relevant domestic and international traffic terminated in that market.
Average A2P SMS pricing and pricing forecasts for the 200 markets covered (termination and wholesale rates) are based on country-by-country information obtained from mobile operators, mobile operator groups, aggregators, cloud service providers, and hubs.
Traffic and traffic growth forecasts for the 200 markets covered are based on our ongoing interviews, discussions and online surveys, and data shared by respondents, looking at total volumes of authorised traffic (white route traffic), and the number of messages received per subscriber per month.
Pricing data applied to A2P SMS traffic used in the CPaaS dataset was received from multiple sources – including both discount and premium messaging providers, as well as operators and operating groups.
All monetary values show the actual income for each part of the value chain – mobile operator or aggregator. Total spend is what a brand pays for messaging with the income from that then split (where appropriate) between mobile operator and aggregator.
White route revenues are calculated based on varying small degrees of direct spend, according to our ongoing interviews, discussions and research (including detailed survey information conducted in 2021 and 2022).
Aggregator white route revenues are calculated based on the difference between the average termination and wholesale rate per market.
RCS Communication Services (RCS) business messaging (RBM) forecast methodology updated for CPaaS
In the CPaaS data Mobilesquared has assumed that 100% of the traffic and spend is via a CPaaS provider.
RCS modelling scenarios are based on our understanding of the expected rollout of RCS services and estimated monthly active users (MAU) from the GSMA. Mobilesquared aligned rollout and MAU information with the level of smartphone penetration per market, the smartphone operating system market share and expected availability of RCS clients in relevant devices.
Mobilesquared has only applied the RBM total traffic and total spend to the CPaaS data.
The starting point for Mobilesquared’s RBM traffic forecasts is the underlying A2P SMS traffic that we believe is most at risk from migration to RBM. In essence, this is the marketing & promotion traffic pushed out over white routes – that is to say official, billable business messaging.
Marketing & promotions constitute roughly 28% of all white route A2P SMS traffic, and it is this type of traffic that we have initially modelled to migrate to RBM – where mobile operators have launched RCS, to network subscribers that have an RCS device client, and where brands and businesses have initiated RBM programs.
We have also presumed that the remaining 72% (predominantly alerts, reminders, and security messages) will largely remain on A2P SMS over the forecast period.
Our engagement and spend data does not include the expected movement of voice-based and in-app customer care onto RBM. This will be available in 2023.
Live RCS networks and registered P2P users are based on ongoing research of the RCS ecosystem by Mobilesquared. A2P users are based on the reported launch of RCS Business Messaging services and device client availability in relevant markets.
P2P and A2P RCS network forecasts and MAU forecasts are based on Mobilesquared research encompassing expected mobile-operator group rollouts, the expansion of competition in more mature markets, the availability of device clients and the launch of business messaging services by relevant mobile operators.
The Google impact on RBM methodology
The biggest change to our previous forecasts is a result of changes in the way Google is now addressing RBM. Firstly, Samsung is migrating its RCS messaging client to Googles Android messages. This will happen by the end of 2021. Although this removes some of the device fragmentation which had previously impacted interconnection, Google will no longer support interconnection of Android Messages users with Open RCS MAAPs – that is to say MAAPs provided by the likes of Mavenir and WIT. Notable exceptions are T-Mobile, Vodafone, and the Telefonica properties in Europe. Any other operators using an Open RCS MAAP will no longer be able to connect to Google Android Messages users by the end of 2021, other than inviting them to download and use the relevant client.
Where operators are using an Open RCS MAAP, and have either launched A2P services, or plan to, we have reduced relevant user numbers in line with Google’s new stance.
Moreover, in markets where Google has launched Guest A2P services, it could be actively competing with the MNOs. Guest A2P was launched in nine markets at the end of 2020 – namely Brazil, Canada, France, Germany, India, Mexico, Spain, the UK, and USA.
We have also presumed that, in future, Google will activate Guest A2P services in markets with more than 2 million registered users that achieve more than 25% penetration.
Income from Google Guest RBM should be split between Google and Aggregators, rather than the MNOs and Aggregators.
Our total RBM engagement forecasts are made up of four components, although only total events and total sessions have been included in the CPaaS data. Nevertheless, in the background our RBM engagements are made up of:
- Event-based A2P – one-off pushes to an opted-in marketing database, including promotions & marketing and customer service updates
- Session-based A2P – a business-originated message that leads to a conversational flow of messages between consumer and business which may ultimately involve replacement of other A2P SMS traffic (e.g. customer service updates and notifications)
- Event-based P2A – a consumer-originated message that leads to a single reply from business which may ultimately involve replacement of other A2P SMS traffic (e.g. customer service updates and notifications)
- Session-based P2A – a consumer-originated message that leads to a conversational flow of messages between consumer and business which may ultimately involve replacement of other A2P SMS traffic (e.g. promotions & marketing, customer service updates, and notifications).
As stated previously, traffic volumes for both session-based A2P and P2A messaging show only the actual number of messages sent by a brand or business to the consumer. However, in terms of billing, we have assumed the brand or business will be charged for all messages contained in any given session at a ratio of 2:5 – that is to say that the brand or business will be charged the cost of terminating an average of two RCS messages for every five exchanged, averaged up or down depending on the number of messages exchanged.
A single push message to consumers would be charged in the same way as SMS – at 1 x the cost of terminating an RCS message.
Mobilesquared has assumed that the average RCS termination rate has now increased from SMS x 115% to SMS x 130%, although there are different pricing models adopted by different operators. We have used actual pricing where we are aware.
We have forecast RMB spend for the key players in the value chain – not revenues. For example, if an RBM termination rate was US$1.00 – a mobile operator would retain 100% of the spend if selling direct to a brand or business, and the wholesale cost if selling to a messaging provider or aggregator. The messaging provider or aggregator would retain the difference between termination cost and wholesale cost – plus a mark-up.
Our spend forecasts are solely based on the delivery of RBM, and do not include any value-added services delivered by an aggregator, messaging or marketing platform – such as service notifications, reporting or integration with business applications, or distribution fees charged by an aggregator to a messaging or marketing platform.
Spend and income is based on commercial campaigns and does not include trials.
WhatsApp Business forecast
In the CPaaS data Mobilesquared has assumed that 100% of the traffic and spend is via a CPaaS provider.
For the purposes of our CPaaS data, we have only used the WhatsApp Business API model, which is based on predominantly medium and large enterprises which are using APIs provided by WhatsApp Business partners. All of our CPaaS data excludes traffic from the WhatsApp Business App, which is a free service for predominantly micro and small businesses.
Mobilesquared researched and found business-size information for 127 of the 195 countries covered by WhatsApp in WhatsApp Business report and dataset (separate to CPaaS) – split between micro (1-9), small (10-49), medium (50-249) and large (>250). Where business-size information was unavailable, we estimated the total number of businesses based on regional averages per head of population and then split by the average percentages for micro, small, medium and large businesses. In the CPaaS data, we have only used the medium and large business data.
The total number of businesses which are using WhatsApp Partner APIs is based on surveys of the Partners starting in the first quarter of 2020, and last updated in 1Q 2022. By the end of 2020, we estimate almost 2,000 companies had engaged with partners, rising to almost 3,000 in 2021, and over 10,000 in 2025.
Total WhatsApp user numbers, and total WhatsApp user forecasts, are based on various public announcements dating back to December 2013 – most recently crossing 1.5 billion in February 2018, 2 billion in February 2020, and 2.7 billion in December 2022.
Historical WhatsApp users were calculated for each year-end and benchmarked against total smartphone penetration to assess actual, and smartphone-penetration, growth trends. WhatsApp user and penetration numbers were then researched and collated for the individual countries and regions contained in our forecasts. Where country information was not available, estimates were benchmarked against regional WhatsApp smartphone penetration.
Country and regional users were then aligned with the global total, and forecast throughout the period based on actual registered users, unique users, smartphone-penetration, and growth trends.
Mobilesquared applied enterprise survey data conducted during 2019-21, to ascertain the average size of the customer database for medium and large enterprises. We also applied consumer survey data conducted during 2019-22, to ascertain the average number of times a consumer would message a brand each year.
In addition, 1-2-1 interviews with WhatsApp Business partners, have provided Mobilesquared with traffic levels on a per country basis, dependent on where each company operates.
Taken together, the data together formed the basis of calculations for all P2A engagement, as well as A2P broadcast messages sent by WhatsApp for Business app users, and A2P template messages sent by Partner API users. We have assumed an average of five messages will be exchanged in both A2P and P2A conversations – 2.5 each from consumer and enterprise. We have assumed a response rate of 30% (starting in 2018) to A2P messages which, in turn, lead to conversational sessions. However, please note, only total traffic and total sessions have been included in the CPaaS data.
Our survey of WhatsApp Partners indicated that approximately 10% of total API traffic constituted A2P templates at the end of 2020. We have assumed this will rise to around 23% by the end of the forecast period due to the apparent interest from enterprise in sending updates and reminders verified by WhatsApp, and relaxation of rules on A2P messaging in September, 2021.
WhatsApp increased the rates it charges for A2P template messages sent via its APIs in February 2022. Simultaneously, WhatsApp also started charging brands per conversation, regardless of whether the conversation is initiated by the brand (A2P), or the consumer (P2A). Each key territory (as identified by Meta) has a “business-initiated rate” (A2P), and a “user-initiated rate” (P2A), which Mobilesquared has applied to each market.
The cost of a consumer-originated conversation is less than a brand-originated conversation. Meanwhile, WhatsApp Partners all appear to have different pricing strategies and price points for both P2A and A2P messaging.
Some WhatsApp Partners charge a flat monthly admin fee, some do not. Some include a P2A message volume within the fee, some do not. Some charge per user, some per message, and most seem to charge differing message rates – although their rates apply to both P2A conversations and A2P templates.
We have based our WhatsApp Business methodology on the guidelines and regulations determined by Meta, which can be accessed here https://developers.facebook.com/docs/whatsapp
Viber forecast methodology
For the purposes of the CPaaS data, Mobilesquared has assumed that 100% of Viber’s business messaging activity is applicable to CPaaS and delivered via CPaaS provider. Only total traffic and spend have been included in the CPaaS data.
The data and assumptions contained in the Viber dataset (separate to CPaaS) were compiled from exhaustive research, interviews and surveys conducted from September 2021 to June 2022.
The addressable market is based on monthly active users – not unique IDs as reported by Rakuten, Viber’s parent company. Monthly active users (MAUs) were modelled on public statements from Viber over the past five years.
We have assumed a significant reduction in the volume of 2022 business messaging traffic in the Ukraine (one of Viber’s largest markets) as a result of the Russian invasion and ensuing war. The subsequent refugee crisis has also resulted in a reduction in the number of MAUs in the Ukraine.
Rakuten itself reported a ¥1.4 billion (US$10.23 million) reduction in Q2-2022 profit for Viber reflecting the situation in Ukraine. However, this incorporates all Viber business activity – including advertising and sale of stickers, as well as traditional business messaging.
The total market size for Viber business messaging traffic volumes, and regional split of business messaging, is based on public statements from Viber. Traffic volumes are the result of multiple inputs across the main regions for Viber business messaging, showing actual business messages delivered up to 2021, and resultant growth trends throughout the forecast period.
We have primarily attributed business messaging to the individual regional markets where Viber has at least, approximately, 20% penetration of the total smartphone user base. Mobilesquared has Viber business messaging active in 48 markets throughout the forecast period.
Facebook Messenger forecast methodology
In the CPaaS data Mobilesquared has assumed that 100% of the Facebook Messenger traffic and spend is via a CPaaS provider. For the purposes of our CPaaS data, we have only used the billable API model, which is based on predominantly medium and large enterprises. Mobilesquared research understands that Meta will start charging for Facebook Messenger messages and sessions based on its WhatsApp Business model, from early 2023, and our forecasts, and associated forecasts included in CPaaS data, reflect this.
Based on this understanding, Mobilesquared has modelled the following:-
An average fee of US$300 per brand per month – payable to the Facebook Messenger Partner
The advertised Facebook Messenger template fee will be applicable from early 2023 – payable to Meta
An average rate of US$0.005 per message sent by enterprise – payable to the Facebook Messenger Partner.
Facebook Messenger will charge brands for messages and conversations regardless of whether the conversation is initiated by the brand (A2P), or the consumer (P2A). Each key territory (as identified by Meta) has a “business-initiated rate” (A2P), and a “user-initiated rate” (P2A), which Mobilesquared has applied to each market as part of the modelling process.
Mobilesquared research estimates that Facebook Messenger was available as a commercial business messaging channel in 94 markets in 2020, this increased to 96 markets in 2022, and will continue to increase to 106 markets by 2026.
Total Facebook Messenger user numbers, and total Facebook Messenger user forecasts, are based on various public announcements dating back to December 2013 – most recently crossing 1.3 billion in September 2017, and 1.4 billion in December 2022.
Historical Facebook Messenger users were calculated for each year-end and benchmarked against total smartphone penetration to assess actual, and smartphone-penetration, growth trends. Facebook Messenger user and penetration numbers were then researched and collated for the individual countries and regions contained in our forecasts. Where country information was not available, estimates were benchmarked against regional Facebook Messenger smartphone penetration.
Country and regional users were then aligned with the global total, and forecast throughout the period based on actual registered users, unique users, smartphone-penetration, and growth trends.
Mobilesquared applied enterprise survey data conducted during 2019-21, to ascertain the average size of the customer database for medium and large enterprises. We also applied consumer survey data conducted during 2019-22, to ascertain the average number of times a consumer would message a brand each year.
In addition, 1-2-1 interviews with Facebook Messenger partners, have provided Mobilesquared with traffic levels on a per country basis, dependent on where each company operates.
Taken together, the data formed the basis of calculations for all P2A and A2P messages and engagement.
We have assumed an average of five messages will be exchanged in both A2P and P2A conversations – 2.5 each from consumer and enterprise. We have assumed a response rate of 30% (starting in 2018) to A2P messages which, in turn, lead to conversational sessions. However, please note, only total traffic and total sessions have been included in the CPaaS data.
Mobilesquared has assumed that Facebook Messenger Partners will charge the same flat monthly admin fee (or not), that they do for WhatsApp Business. Similarly, we have applied the same assumption that some Facebook Messenger Partners include a P2A message volume within the fee, and that some do not. Some charge per user, some per message, and most seem to charge differing message rates – although their rates apply to both P2A conversations and A2P templates.
We have based our Facebook Messenger methodology on the guidelines and regulations determined by Meta, which can be accessed here https://developers.facebook.com/docs/messenger-platform
Apple Messages for Businesses forecast methodology
In the CPaaS data Mobilesquared has assumed that 100% of the Apple Messages for Business (AMB) traffic and spend is via a CPaaS provider. AMB is an inbound customer care model, targeting large of medium-large enterprises with a platform that supports asynchronous messaging with live agents and basic automation support.
AMB can only be accessed via an Apple partner (or Messaging Service Provider – MSP as Apple refers to them as). The MSPs charge a monthly licensing fee to use the platform based on a cost-per-user model, and a message fee of $0.005 once the business has exceeded their monthly allocation of 500 messages.
The data and assumptions contained in the AMB dataset (separate to CPaaS) were compiled from exhaustive research, interviews and surveys conducted from September 2021 to June 2022.
The addressable market is based on monthly active iPhone users, modelled on public statements from Apple over the past five years. Mobilesquared has based the availability of AMB on where Apple has a dominant or strong smartphone market share and the availability of Apple Pay. To this end, Mobilesquared believes AMB was available in 26 markets in 2020, and has increased to 29 and is expected to stay at this number for the duration of the forecast period.
Mobilesquared has applied the total market size traffic volumes based on data shared by industry, and supplemented by online surveys, and where applicable, third-party data. Spend is based on the average prices applied by MSPs. Mobilesquared research reveals that Apple has received a very low take-up of the service, and this has been reflected in the very low traffic and spend forecasts for the channel by Mobilesquared.
OTT messaging apps forecast methodology
In the CPaaS data Mobilesquared has assumed that 100% of the OTT messaging app traffic and spend is via a CPaaS provider. Given the limited reach in markets outside of their domestic dominance, Mobilesquared opted to create an OTT messaging catchall category to cover the likes of WeChat (China), LINE (Japan), and KakaoTalk (South Korea), and the smaller emerging channels (from a business messaging perspective) such as Telegram.
To generate traffic and spend for each messaging channel in each market, Mobilesquared applied company information and data, supplemented with publicly available information and data from third-parties, as well as data and insight from Mobilesquared’s research and online surveys.
As pricing data for these channels was not widely available, Mobilesquared has applied a generic price point based on average price of rich messaging channels (RCS, WhatsApp and Viber), weighted against the average cost of an A2P SMS for the limited number of markets where business messaging is available over these channels.
MMS forecast methodology
In the CPaaS data Mobilesquared has assumed that 100% of the MMS traffic and spend is via a CPaaS provider. Traffic (and service availability) is based on data shared during Mobilesquared’s 1-2-1 interviews and online surveys. Based on this information, Mobilesquared believes MMS is used as a business messaging proposition in 75 markets, although it is dominated by only a handful of markets.
Spend is based on available market pricing by MMS providers, and where not available, Mobilesquared has applied a weighted average.
Messageverse generic data
The starting point for MessageMap IQ data is our mobile operator subscriptions database. We research the top 200 markets worldwide on an annual basis, collating information from mobile operators and mobile operator groups, regulators, and where applicable GSMA Intelligence.
Data-only contracts and M2M accounts have been excluded where available. Subscription forecasts are modelled based on previous growth records, overall mobile penetration rates and expected population growth. Population (global and national) numbers and national demographic data based on data from worldometer.com, population.un.org, worldpopulationreview.com, and populationpayramid.net.
Mobilesquared updates smartphone numbers and OS breakdowns on an annual basis to create a smartphone database, using device manufacturer data and quarterly updates, mobile operator data, national regulator data, and GSMA Intelligence data where applicable. In addition, when required, this data is also supplemented by publicly available third-party data. Smartphone forecasts are modelled based on previous growth records, overall smartphone penetration rates and expected growth rate as percentage of total unique mobile users and total population growth.
User numbers based on Mobilesquared data, and where applicable, Mobilesquared’s smartphone database. OTT user numbers based on company data, consumer research, and publicly available third-party data.
Engagement data based on existing Mobilesquared data, extensive consumer research (last updated 4Q 2021), and case studies shared by companies operating within the business messaging ecosystem privately to Mobilesquared, and publicly available case studies also. Additional third-party consumer research data applied where it provides supplementary and incremental consumer behaviour knowledge.
Consumer research conducted by Luc.id on behalf of Mobilesquared. UK-based Luc.id conducted multiple consumer research campaigns for Mobilesquared between 2018 and 2021, using its online panel of double opt-in consumer participants. Data in Messageverse is based on consumer research of >20,000 respondents based across the Americas, Europe, Middle East & Africa, and Asia Pacific. Where Mobilesquared has actual market consumer data, this has been applied to Messageverse. Where actual consumer data is not available for a particular country, Mobilesquared has applied the regional average from the consumer data.
Engagement Methodology
In 1H2022, Mobilesquared undertook a series of consumer engagement surveys covering 10,000 individuals across 10 markets – Australia, Brazil, Colombia, Mexico, France, Germany, UK, Malaysia, Singapore and the USA.
Results from Brazil, Colombia & Mexico were averaged to provide indications for other Latin American markets while, similarly, France, Germany & UK were averaged to provide indications for other European markets. New Zealand was aligned with results from Australia and, similarly, Canada was aligned with results from the USA.
In total, we have extrapolated results for a further 65 markets giving total engagement coverage of 75 markets. Later this year, we will repeat the process to include coverage of markets in Eastern Europe, the Middle East and North Africa and elsewhere in Asia.
Specifically, we wanted to know the following:
- Which communications channels do you use on your mobile phone?
- How do brands and businesses mainly contact you?
- How often are you contacted by brands and businesses on the channels selected previously?
- Of the brands that you have opted-in to receive communications from, which sectors do these operate in?
- Do you find the messages from brands and businesses on the channels selected previously useful?
- What is an acceptable number of messages that you would like to receive on a weekly basis from the brands/ channels that you have opted-in to?
- What is the preferred time of delivery for messages from the brands/ channels that you have opted-in to?
- How many messages from brands will you read?
- How many messages from brands will you click on to find out more?
- How likely are you to make a purchase based on a message you receive from a brand, assuming that the offer is targeted and relevant to you?
- What would be your preferred payment option when making a payment in a messaging channel?
The survey answers enabled us to model engagement behaviour based on gender, age and channel and were applied to our unique user research covering individual messaging channels.
Where applicable we have used older consumer data from 2019 and 2018 to provide more granular information. This was based on research of 2,000 and 3,000 users respectively, on engagement issues based on the usefulness of different message types, and the timing of when to most effectively send a message.
All of the above will be updated in our 2022 consumer research of 40+ markets.
Disclaimer
© 2025 Mobilesquared Ltd. All rights reserved. The Data from the Messageverse™ included in this document is protected by international copyright laws and other intellectual property rights. The owner of these rights is Mobilesquared Ltd.
This data may not be: (i) copied or reproduced; or (ii) lent, resold, hired out or otherwise circulated in any way or form without the prior permission of Mobilesquared Ltd. Any use of this data must be attributed to Mobilesquared.
Whilst reasonable efforts have been made to ensure that the information and content of this publication was correct as at the date of publication, neither Mobilesquared Ltd nor any person engaged or employed by Mobilesquared Ltd accepts any liability for any errors, omissions or other inaccuracies. Data and assumptions relevant as of Jan 2024. Users should independently verify any facts and figures as no liability can be accepted in this regard – users assume full responsibility and risk accordingly for their use of such information and content.
Without good data, you’re just another person with an opinion. Our business messaging data is the best available, enabling you to make the right decisions at the right time. Buy our data as a subscription, pay-as-you-go, or bespoke research or consultancy.
As seen on TV (and elsewhere)
If our name is familiar, that’s because we’re the go-to messaging experts for leading media including BBC, CNBC and the FT. We’re also the official research partner of the Mobile Ecosystem Forum and keynote presenters at global messaging events.
“Tell me you’re good without saying it…”
We supply business messaging intelligence to industry leaders in various sectors and our forecasts shape the messaging strategies of some of the world’s biggest brands. See who we work with already, and how we can add value to your messaging strategy.
Without good data, you’re just another person with an opinion. Our business messaging data is the best available, enabling you to make the right decisions at the right time. Buy our data as a subscription, pay-as-you-go, or bespoke research or consultancy.
As seen on TV (and elsewhere)
If our name is familiar, that’s because we’re the go-to messaging experts for leading media including BBC, CNBC and the FT. We’re also the official research partner of the Mobile Ecosystem Forum and keynote presenters at global messaging events.
“Tell me you’re good without saying it…”
We supply business messaging intelligence to industry leaders in various sectors and our forecasts shape the messaging strategies of some of the world’s biggest brands. See who we work with already, and how we can add value to your messaging strategy.