
There is a hard limit on how much information a human brain can absorb and retain in a given window of time and insurers are consistently bumping against it without realising that is what is happening.
80% of workers now experience information overload, up from 60% in 2020. The average screen-based attention span has dropped from 2.5 minutes in 2004 to just 47 seconds in 2024 and 2025. When you put a new agent through ten product training modules in sequence, you are not giving them ten units of knowledge. You are giving them a cognitive pile-up where each new module competes with and partially overwrites the one before it.
Multitasking during training climbed to 70% in 2025, up from 58% in 2024. Seven in ten agents are checking phones, messaging colleagues or mentally drifting during training sessions and that is not a discipline problem. It is what happens when the format of content delivery asks more of the human brain than the human brain can sustainably give.
Sales teams forget 70% of information learned within a week of training and 87% forget it within a month. For an insurer with a ten-product portfolio, that means an agent who completed their full training programme in January is working from fragmented, partially incorrect product knowledge by February and the problem compounds with every new product launch, regulatory update or feature change added on top.
The issue is not that agents are incapable of learning ten products. The issue is that dumping all ten products into their heads in sequential modules is the worst possible way to make that knowledge stick.


The course-per-product model feels logical to build because it mirrors how products are organised internally. The health team owns the health module, the motor team owns the motor module and so on. Each team ensures their content is comprehensive, accurate and compliant. The result is ten well-built courses that collectively overwhelm the agent trying to learn from all of them.
Beyond cognitive overload, this model has three other structural problems that compound over time.
The first is the update cycle. When a product changes — a new exclusion clause in the health policy, a revised premium structure in the motor product, a regulatory update affecting the ULIP — every relevant course needs to be updated, re-reviewed by compliance, re-distributed and re-completed by agents. For a ten-product portfolio this creates a near-permanent content maintenance burden that most L&D teams handle reactively rather than proactively, which means agents are frequently working from slightly outdated product knowledge without knowing it.
The second is relevance mismatch. An agent whose book of business is 80% health and motor renewals does not need to spend equal time on every module in the product library. Sitting through a comprehensive pension product course when their entire pipeline is health renewals is not just inefficient — it actively dilutes the learning time available for the products they actually sell.
The third is the cross-sell gap. At some of the largest agency organisations, accounts are only 3 to 6% cross-sold across their platform. One of the primary reasons agents do not cross-sell is not that they lack the motivation to do it but that they do not know enough about the adjacent product at the moment of the conversation to recommend it with confidence. A customer who mentions their child is starting college is a natural opportunity for an education protection conversation. But if the agent's knowledge of that product is a half-remembered module from four months ago, the moment passes.
The second principle is that repetition over time beats intensity in a single session. An agent who receives a three-minute reel on the key differentiators of a critical illness product every month for four months retains more of that information than an agent who sat through a two-hour module once and has not revisited it since. Microlearning boosts retention rates by 50% compared to traditional training methods and the mechanism is straightforward — repeated short exposures to specific content move knowledge from short-term to long-term memory in a way that a single comprehensive session never can.
The third principle is that cross-sell training works best when it is triggered by customer context rather than scheduled in advance. An agent who is about to call a health insurance customer for a renewal conversation needs a two-minute reel on how to open a motor insurance conversation with that same customer. Not a full motor product course — just enough contextual knowledge to spot the signal, ask the right question and know what to say next. That is the kind of training that actually produces cross-sell behaviour.
Is built through ongoing personalised training tied to the agent's actual portfolio. An agent who is primarily selling motor and health receives regular, progressively detailed reels on those two products — objection handling, pricing logic, claim scenarios, regulatory disclosures — until their knowledge is deep and confident. The depth layer is different for every agent based on what they actually sell.
Delivers cross-sell prompts tied to specific customer moments. Before a renewal call with a health customer who also owns a car, the agent receives a brief reel on how to open a motor conversation naturally. Before a call with a customer approaching retirement, the agent receives the relevant pension product framing. This layer turns cross-sell training from a scheduled event into a continuous, contextual habit.
Ensures that when any product changes, every agent who sells that product receives a short update reel immediately — not a revised module they need to re-complete but a 90-second summary of exactly what changed and what it means for their conversations with customers.
Amplispot's AI Personalised Reels is built for exactly this kind of layered, contextual multi-product training approach. The platform generates personalised short-form training reels for each agent based on their actual book of business — which products they sell, which customers are coming up for renewal, which cross-sell signals are visible in their portfolio data and which product knowledge gaps are most likely to be costing them conversations.
An agent with a motor-heavy book receives a different training sequence than an agent whose portfolio is primarily life and pension products. A new agent in their first month receives foundational product overview reels across the full portfolio. A senior agent who has not sold a travel insurance policy in twelve months receives a short refresh before a customer meeting where it might be relevant. All of this happens automatically based on data, without requiring the central L&D team to manually design and assign a different training path for each of the hundreds of agents in the network.


The commercial case for fixing multi-product training is straightforward and significant. The probability of selling to an existing customer is 60 to 70%, compared to 5 to 20% for new customers. Cross-selling is most successful in banking, retail and insurance, with success rates above 70% when executed well. Customer lifetime value can increase by up to 40% through strategic cross-selling and 90% of cross-selling success depends on understanding customer needs.
Every one of those figures assumes the agent has enough product knowledge to spot the opportunity and execute the conversation. When agents are carrying incomplete, partially forgotten knowledge across a ten-product portfolio because they went through ten courses once and have not revisited them since, the opportunity is present but the capability is not. And the gap between the two is exactly the cross-sell revenue that sits uncaptured in almost every multi-product insurer's existing customer base.
Advanced analytics tools that help sales agents identify products that best suit existing customers based on their needs, portfolio and purchase history are already in use at leading insurers for identifying cross-sell opportunities. The missing piece in most implementations is the training infrastructure that ensures agents can act on those signals confidently when a customer conversation presents the moment.
Yes. Training paths are configured based on agent profile data including tenure, product portfolio and performance metrics, so a new joiner receives a different sequence than a senior agent being upskilled on a new product line.
Central content updates go live immediately and reach all relevant agents in the same distribution cycle, removing the lag that typically sits between a product change and agents actually knowing about it.