Why Your Customer's Journey Is Not a Funnel
Digital Customer Journey
Digital Customer Journey
Why Your Customer's Journey Is Not a Funnel
The funnel was designed for a media environment that no longer exists. Academic research on the digital customer journey shows what has replaced it – and why marketing built around a straight line is structurally mismatched to how customers actually move.
Quick Answer: A systematic review of the customer journey literature by Tueanrat, Papagiannidis, and Alamanos (2021), published in the Journal of Business Research, found that digital customer journeys are non-linear – shaped by social influence and algorithmic mediation rather than moving through a fixed sequence of stages. Customers enter at multiple points, disengage and re-enter, and are influenced by peer networks and recommendation algorithms that operate largely outside a business’s direct control. Marketing structured around a linear funnel – built only for brand-controlled channels in a fixed sequence – is structurally mismatched to how the journey actually unfolds.
The Funnel Was Built for a Media Environment That No Longer Exists
The marketing funnel – awareness, interest, desire, action – has shaped how several generations of marketers think about customer behaviour. It is intuitive, easy to diagram, and easy to build campaigns against. It is also a model of a media environment that has substantially changed.
The funnel assumes a single, brand-controlled path: a business pushes a sequence of messages through paid and owned channels, and the customer is assumed to move through awareness, consideration, and decision in that order, driven primarily by what the brand chooses to communicate at each stage. This model maps reasonably well onto a broadcast media environment, where television, print, and radio gave brands meaningful control over sequence and narrative.
A systematic review of the customer journey literature, conducted by Tueanrat, Papagiannidis, and Alamanos and published in the Journal of Business Research in 2021, analysed the substantial body of academic work that has emerged on this topic – the review notes that customer journey literature grew sevenfold over the preceding decade. Their analysis identifies a consistent pattern across this literature: digital customer journeys are not the sequential, brand-controlled process the funnel model assumes. They are shaped by touchpoints distributed across owned, earned, and paid channels, by social interactions with other people, and by algorithmic systems that determine what a customer encounters and in what order – much of which sits outside the business’s direct influence.
The funnel assumes the brand sets the pace. The research shows the customer, their social network, and the platforms mediating their search jointly determine what happens next.
What 'Non-Linear' Actually Means in Practice
Non-linearity is sometimes treated as a vague descriptor – a way of saying the journey is complicated. The academic literature is more specific than that. The table below contrasts the assumptions built into the linear funnel model against the structural features the customer journey literature identifies in digital environments.
Dimension | The Linear Funnel Model | The Non-Linear Digital Journey (Tueanrat et al., 2021) |
Structure | Awareness → Interest → Desire → Action. A single path, moved through in sequence, controlled largely by brand-pushed messaging. | Multiple entry points, re-entries, and skipped stages. A prospect can discover a business through a peer recommendation, leave, return through search, and re-enter the journey at a different stage entirely. |
Who drives the journey | The brand. Marketing pushes messages through paid media and owned channels in a planned sequence the business controls. | The customer, shaped by algorithmic systems and social interactions the business does not control. The journey is co-produced by the customer, their social network, and the platforms mediating their search. |
Where research happens | Brand-controlled channels: the business’s own website, advertising, and sales conversations. | Distributed across peer review platforms, social media, AI-assisted search, and algorithmic recommendation systems – many of which the business has limited visibility into. |
What determines the next step | The brand’s sequenced campaign – the next ad, the next email in a fixed sequence. | A combination of social influence (what peers and influencers say) and algorithmic mediation (what a platform’s recommendation system surfaces) – both operating largely outside the brand’s direct control. |
Implication for marketing | Optimise the sequence: refine each stage of the funnel to improve conversion to the next stage. | Optimise for presence at the touchpoints customers actually use, in the order they actually use them – which requires monitoring real touchpoint data rather than assuming a fixed sequence. |
The practical significance of this contrast is not academic. A business that structures its marketing around a funnel – one sequence, one set of brand-controlled touchpoints, conversion measured only at the end of a fixed path – is building a system that does not match the structure of how its customers are actually moving. The mismatch produces specific, measurable consequences: prospects who re-enter the journey through an unmonitored channel are not recognised as returning. Prospects who are influenced primarily by a peer conversation are evaluated as if the brand’s messaging were the dominant factor in their decision. Touchpoints that exist outside owned and paid channels are under-resourced because they are not part of the model being used to plan marketing activity.
Fractional CMO South Africa
Two Forces the Funnel Does Not Account For: Social Influence and Algorithmic Mediation
The review identifies two specific mechanisms shaping the non-linear journey that have particular relevance for how a service business should think about its marketing structure: social influence and algorithmic mediation.
Social influence is a journey touchpoint, not a pre-journey input
A common assumption in funnel-based thinking is that social influence – word of mouth, referrals, peer recommendation – operates before the journey begins, bringing a prospect into the top of the funnel, after which the brand’s own messaging takes over. Hamilton, Ferraro, Haws, and Mukhopadhyay, writing in the Journal of Marketing (2021), challenge this assumption directly. Their research on what they term the social customer journey demonstrates that customers frequently travel through their purchase journey accompanied by other people, physically or virtually, at multiple stages – not only at the start. Group conversations, visible peer behaviour, and ongoing social input shape decisions throughout the journey, not just before it.
“Customers often do not travel through the customer journey alone. The presence of companions – whether physically present or psychologically salient – shapes choices throughout the journey, not only its initiation.” – Hamilton, Ferraro, Haws, & Mukhopadhyay (2021, Journal of Marketing)”
Algorithmic systems determine what a customer encounters, and when
The second mechanism is structural rather than social: search engines, social platforms, and increasingly AI-assisted research tools determine which content a prospect encounters, in what order, and how prominently – through ranking and recommendation logic that is largely invisible to both the customer and the business being evaluated. Jeong, Kim, Li, Li, Choi, and Kim, writing in Sustainability (2022), studied this dynamic empirically using a Stimulus-Organism-Response framework, finding that algorithmically generated recommendations function as a distinct stimulus shaping customer behaviour and engagement – not simply as a passive filter on existing content, but as an active influence on what the customer’s journey actually contains.
For a South African service business, the combined implication of these two mechanisms is direct: a meaningful share of the journey is being shaped by forces the business has limited visibility into and cannot directly script. The marketing system needs to account for this rather than operate as if the brand’s own messaging is the dominant input at every stage.
Fractional CMO South Africa
Social and Algorithmic Influence Mapped to Practical Marketing Decisions
The table below translates the two influence mechanisms identified in the literature into specific implications for how a growing service business should think about its touchpoints.
Influence Type | How It Shapes the Journey | What It Means for a SA Service Business |
Social influence (peer and network) | Hamilton et al. (2021, Journal of Marketing) demonstrate that customers frequently travel through their purchase journey accompanied by other people – physically or virtually – whose presence shapes the choices made at each stage. Recommendations, group discussions, and visible peer behaviour all function as journey touchpoints, not just pre-journey inputs. | Referrals and peer recommendations are not a separate activity from the digital journey. They are touchpoints within it. A prospect’s LinkedIn network, WhatsApp group, or industry peer conversation is shaping their evaluation in real time, often alongside their independent online research. |
Algorithmic mediation | Search engines, social platforms, and AI-assisted tools increasingly determine which content a prospect encounters, in what order, and how prominently. The journey is shaped by ranking and recommendation logic that is largely invisible to both the customer and the business. | Visibility in algorithmically mediated environments – search rankings, AI Overview answers, social platform reach – is not a downstream output of marketing. It is a structural determinant of whether a business is even present at the touchpoints where the journey unfolds. |
Touchpoint re-entry | Tueanrat et al. (2021) identify that digital journeys frequently loop back – a customer may research, disengage, and return to an earlier stage after new information or a changed circumstance, rather than moving in one direction toward purchase. | A prospect who visited your website three months ago and did not convert has not necessarily exited the journey. They may re-enter it later through a different touchpoint – a referral, a piece of content, a renewed need. Marketing systems built only to capture first-visit conversions miss this re-entry behaviour entirely. |
Multiplicity of touchpoint types | The digital journey spans owned (website, email), earned (reviews, word of mouth, organic search visibility), and paid (advertising) touchpoints simultaneously, often interacting with each other rather than functioning independently. | A business that manages each of these touchpoint categories separately – different suppliers, different metrics, no coordination – is managing the parts of a journey that customers experience as a whole, in a structure that does not reflect how the journey actually works. |
Why the Funnel-Only Approach Produces Specific, Avoidable Costs
The cost of marketing built only around a linear funnel is not abstract. It shows up in three specific, observable patterns. The first is the prospect treated as lost after a single non-conversion, when the non-linear model suggests they may simply be between touchpoints, likely to re-enter the journey later through a different channel. This is the same dynamic identified in the earlier Rolland Digital article on the B2B buyer journey – most B2B buyers complete a substantial portion of their evaluation before ever making contact, often across multiple separated sessions, not in one continuous funnel pass.
The second is the under-resourcing of touchpoints that do not fit neatly into brand-controlled channels – referral systems, review management, organic search visibility – because a funnel model implicitly treats these as peripheral to the ‘real’ marketing activity of advertising and owned content. As the earlier article on disconnected marketing established, channels managed in isolation, with no shared model of how the customer is actually moving between them, consistently produce duplicated effort and orphaned leads.
The third is the absence of any system to recognise and respond to re-entry. A prospect who returns to the website eight weeks after their first visit, having been influenced in the interim by a colleague’s recommendation or a piece of content they encountered through search, is frequently treated by the business’s marketing systems as a new, unknown visitor – losing the accumulated context of their earlier engagement.
What to Build Instead of a Funnel
None of this means abandoning structure. It means building a marketing system whose structure reflects how the journey actually works, rather than how it is convenient to diagram. The table below maps four specific implications from the research to what a service business should build in response.
What the Research Implies | Why a Funnel-Only View Misses It | What to Build Instead |
The journey has multiple entry points, not one. | A funnel assumes every prospect starts at ‘awareness’ and is unaware of the business. In practice, a referral can place a prospect directly into a late-stage evaluation mindset, skipping the stages a funnel assumes must happen first. | Map the most common entry points your actual clients describe – not the theoretical first stage of a textbook funnel. Build content and touchpoints for each entry point, not just for the assumed beginning. |
Re-entry and disengagement are normal, not failures. | A funnel treats a prospect who does not convert on first contact as a lost lead. The non-linear model suggests they may simply be between visits – present in the wider journey but temporarily inactive. | Build a re-engagement system – email, retargeting, periodic relevant contact – that assumes prospects will return rather than treating every non-conversion as final. |
Social and algorithmic touchpoints are part of the journey, not separate from it. | Marketing structured around brand-controlled channels (website, ads, email) ignores the touchpoints a funnel does not account for – peer review platforms, social recommendation, AI-mediated search. | Treat referral systems, review management, and search visibility as core journey infrastructure – measured and managed with the same rigour as paid advertising, not as peripheral activities. |
The journey is co-produced, not brand-controlled. | A funnel implies the business sets the pace and sequence. In practice, the customer, their network, and the platforms they use jointly determine what happens next. | Build systems that respond to where the customer actually is – using behavioural and CRM data to identify journey stage – rather than systems that assume every prospect is following the brand’s planned sequence. |
The common thread across all four rows is the shift from a sequence-based model to a touchpoint-based model: rather than assuming every prospect moves through the same stages in the same order, the marketing system identifies where a specific prospect actually is – based on behavioural and CRM data – and responds accordingly. This is precisely the function a well-structured CRM and segmentation system performs, and why building that infrastructure early in a business’s growth, as outlined in the article on email and CRM discipline, pays dividends as the journey complexity increases.
A practical starting point: Trace the actual path your last five clients took – not the funnel you assume exists, but what they describe when asked how they found you and what influenced their decision. Most service businesses discover that the real path involved multiple touchpoints, at least one social or referral element, and at least one re-entry after an initial period of inactivity. That gap between the assumed funnel and the actual path is the single most useful diagnostic for understanding where your current marketing system is structurally mismatched to your customers’ real behaviour.
Frequently Asked Questions
What is the difference between a marketing funnel and a customer journey?
A marketing funnel is a model that assumes customers move through a fixed sequence of stages – typically awareness, interest, desire, and action – driven primarily by the brand’s own sequenced messaging. A customer journey, as analysed in the academic literature reviewed by Tueanrat, Papagiannidis, and Alamanos (2021), is a more realistic model that accounts for multiple entry points, re-entry after disengagement, and influence from sources outside the brand’s control, including social networks and algorithmic recommendation systems. The funnel describes an idealised sequence; the journey describes what the research shows actually happens.
What does ‘non-linear’ mean in the context of the digital customer journey?
Non-linear means that customers do not necessarily move through journey stages in a single, fixed sequence. They can enter at different points (a referral might place someone directly into late-stage evaluation), disengage and later return to an earlier stage after new information, or be influenced simultaneously by multiple touchpoints rather than moving through them one at a time. The academic literature on customer journeys consistently documents this pattern, in contrast to the assumed sequence of the traditional marketing funnel.
How does social influence affect the customer journey?
Research by Hamilton, Ferraro, Haws, and Mukhopadhyay (2021), published in the Journal of Marketing, found that customers are frequently accompanied through their purchase journey by other people – whether physically present or psychologically salient through remembered conversations and social context. This influence operates throughout the journey, not only at its outset. For a business, this means referral and peer recommendation activity functions as an ongoing touchpoint within the journey, not a separate pre-journey input that brand messaging then takes over from.
What is algorithmic mediation and why does it matter for marketing?
Algorithmic mediation refers to the role that search engines, social media platforms, and AI-assisted research tools play in determining which content a customer encounters during their journey, in what order, and with what prominence. Because this ranking and recommendation logic operates largely outside both the customer’s and the business’s direct control, a business’s visibility within these algorithmically mediated environments functions as a structural determinant of whether it is present at the touchpoints where the journey actually unfolds – not simply a downstream marketing output to optimise separately.
How should a service business restructure its marketing to account for a non-linear journey?
The core shift is from a sequence-based model to a touchpoint-based one: rather than assuming every prospect moves through the same stages in the same order, the marketing system should identify where a specific prospect actually is, using behavioural and CRM data, and respond accordingly. This means building re-engagement systems for prospects who disengage rather than treating non-conversion as final, treating referral and review management as core journey infrastructure rather than peripheral activity, and ensuring marketing channels share data so a returning prospect is recognised rather than treated as new. A Marketing Audit and ROI Review is a practical starting point for mapping how your customers’ actual journey compares to the funnel assumptions your current marketing structure may be built around.
Want to map your customers' actual journey against your current marketing structure?
A Rolland Digital Marketing Audit and ROI Review identifies where your marketing system is built around funnel assumptions that do not match how your customers actually move. Book a consultation at rollanddigital.co.za/contact. No pitch. No pressure.