Briefing: Using Commission Based Agents for Marketing
Summary
SafeMotos is investigating the building of a group of physical agents to act as an ordering party for new users. This will be a way to avoid the limitations of low smartphone penetration and low data usage, while being an effective tool of customer acquisition and brand awareness.
Problem Resolution
Low smartphone / data usage penetration rate in Kinshasa
While Kinshasa is a mega city of ~15M people with a fully distributed 4G network, smartphone penetration and internet usage is nascent.
| City | Facebook Ad (Q3 2018) | Google Ad (Q3 2018) |
| Kigali | 330,000 | 228,000 |
| Cotonou | 530,000 | 354,000 |
| Ougadougou | 540,000 | 254,000 |
| Kinshasa | 800,000 | 167,000 |
| Douala | 920,000 | 583,000 |
| Bamako | 950,000 | 395,000 |
| Kampala | 1,600,000 | 815,000 |
| Dar es Salaam | 2,100,000 | 1,400,000 |
| Accra | 2,600,000 | 2,114,000 |
| Nairobi | 3,800,000 | 3,200,000 |
| Lagos | 5,900,000 | 5,300,000 |
| Jakarta | 35,000,000 |
By using a network of agents it will give access to SafeMotos services to users who don’t have smartphones or active data connections.
Tech Illiteracy
As SafeMotos has learned in Kigali, even if someone has a smartphone with data, they are often very new to using technology and are unfamiliar with installing applications, registration of services, user flows and city maps.
In Kigali, the highest retention of users were those acquired via a physical third party (SafeMotos street team) since they were able to answer questions in a level of detail that normal client acquisition channels are unable to specify.
Agents will work as a way to explain SafeMotos services to customers and onboard them in a sensitive way.
Speed to Scale
A distributed agent network will be a fast way to make initial scale progressing in Kinshasa and is hypothesized to be a rapid way to kickstart the network.
This spreadsheet shows a hypothetical but reasonable acquisition projections and how they can translate into trip volume.
| Month 1 | Month 2 | Month 3 | Month 4 | Month 5 | Assumptions | |||
| Agent Number | 20 | 50 | 100 | 150 | 200 | Daily new user acquisition | 5 | |
| Monthly Acquisitions | 3010 | 7525 | 15050 | 22575 | 30100 | Retention rate | 25% | |
| Total retained users | 752.5 | 2634 | 6396 | 12040 | 19565 | Weekly trip / retained users | 5 | |
| Total trips from agent users | 16179 | 56626 | 137519 | 258860 | 420648 | |||
| Daily trip volume | 532 | 1863 | 4524 | 8515 | 13837 |
(source)
Who Are Agents
Agents can be:
- SafeMotos drivers when idle as another revenue stream
- Local mobile internet (Vodacom guys) who want to have another revenue stream
- A secretary or employee at a B2B partner
- A SafeMotos customer who has hit some threshold of trips and applied to be an agent (think someone at a bar recruiting SafeMotos customers to pay for a next beer)
- Recruited individuals to fill coverage gaps (eg late night areas)
- Part of SafeMotos branding campaigns (like Redbull style street team)
Each agent location is also a verified location that a customer can be dropped off at night or in risky areas.
Compensation
Agent compensation will be linked to triggers such as:
When agent is idle:
- Guarantee engine (for certain times / certain locations with prior approval)
- List of slipping away users (recruited by this agent)
- List of slipping away users (recruited by other agents)
- List of users who never took a follow on trip (recruited by this agent)
- List of users who never took a follow on trip (recruited by other agents)
- Raise awareness actions in local community (eg fliers)
- Recruit more agents
- Manage recruited agents
When agent is with a customer:
- Install app on smartphone
- Register with digital payment
- Put in facial recog
- Share SafeMotos with friends
Payment rationale should be clearly communicated to agents and tied to business objectives.
Some agents will be able to pull money out as physical currency (super agents) where some agents will be able to use the funds as in app credit.
Interface
There will need to be a modification of the existing SafeMotos technology that results in a hybrid of driver functionality and savings functionality.
Login: There needs to be a login to setup shop at this specific location. Would assume Smile Identity facial recognition, as well as this being at a location that the agent has been approved for in the past by the SafeMotos team.
Homepage: Can be a summary of agent earnings data (based off of driver app). There can be a big NEW TRIP button for starting a new trip. Managing ongoing trips can be a swipe style interface (eg when ordering two drivers) and at the bottom of the home page is an infinite scroll dynamic box (like Uber News) under a heading “Make More Money”.
Make More Money Page: This will an interface actions the agent can do to increase earnings while being idle. The interface will be a series of lines listing the task and how much compensation the agent will get for accomplishing the task (perhaps on a per hour basis). When the agent clicks a job there will be a details page explaining what needs to be done and what the verification will be verify the task.
New Trip Page: When clicking new trip the first thing that will come is to enter the clients phone number (or facial recog) for two step verification. If the client is a returning customer it will go straight to ordering a SafeMotos service. If it is a new customer it will go to a registration flow for this customer. While the bike is enroute the agent app screen will switch to the ‘Make More Money Page’ where the agent will work to get additional details from the customer.
Manage Trip Page: This will be a simple page where the agent can track an ongoing trip. There can be a way to reach out to either the customer or the driver via chat to communicate in case there are any issues.
Agent Management / Recruitment
Agents can be recruited via the app, via physical advertising and via existing agents.
There can be techniques borrowed from multi level marketing where agents can be incentivized to recruit additional agents by getting some commission for their success and also have certain managerial responsibilities over the sub agent.
Business Intelligence
Dashboards will need to be set up to monitor agent efficacy. The dashboard will be borrowed from driver dashboards to see data such as:
- Agent location
- New registrations per agent per day
- Avg trips taken by acquired user (at network level abd per agent level)
- Total trips taken by acquired user per agent (at network level abd per agent level)
- Retention rate of acquired user per agent
- Average cost per retained user
- Agent satisfaction
- Agent earnings
FreeMotos for Agents
SafeMotos has developed a technology system that allows for subsidized trips to users when the user goes to a specific location (eg an event, a bar, a restaurant). The business owner pays SafeMotos for the customer acquisition.
This system would work well with the agent network.
A SafeMotos user could select to go to certain sponsored agents, but when the trip ends the agent needs to verify the reception of the user and is responsible to make sure the user going into the sponsored business.
This could also be inverted, where a business partner could run a promotion that would allow users to go home at a price subsidized by the business (IE buy 3 drinks go home for free). The business would have a series of promo codes they pay for that would be entered into the app to bill the trip to their corporate account.