What E-Commerce Gets Wrong About the Customer Journey (And What SaaS Can Learn)
Published:
Reading time:
Category:
Content
The linear funnel model of the customer journey, meaning awareness, consideration, purchase, and loyalty, has shaped product and marketing design for decades. It is clean, intuitive, and wrong.
Research consistently shows that the large majority of consumers use multiple channels during their shopping journey before making a purchase decision. Google's research team coined the term messy middle for the space between initial trigger and final purchase, where customers loop between exploration and evaluation, returning to earlier stages, switching devices, consulting peers, and revisiting options they had already dismissed.
And yet most product teams still design for a linear path. Onboarding flows assume users arrive with full context and move forward in sequence. Marketing pages assume users read from top to bottom and convert at the CTA. Analytics are built around funnels where users fall out at predictable stages. The result is a product experience designed for a user who does not exist. The research methods behind mapping the real journey are covered in the guide to user research methods.
What the Research Actually Shows About How Journeys Work
The classical Engel-Kollat-Blackwell model, developed in the 1960s and still widely referenced, describes consumer decision-making as a five-stage linear sequence: problem recognition, information search, evaluation of alternatives, purchase, and post-purchase evaluation. It remains influential because it provides a useful analytical scaffold. The problem is that it was developed for a world without smartphones, social media, or peer review infrastructure.
Non-linearity. Users revisit earlier stages repeatedly. A user who has evaluated and nearly committed to a product returns to information search when a friend shares an alternative, or when a review raises a new concern.
Multi-channel. Most consumers use multiple channels during their journey. A user discovers a product on TikTok, researches it on Google, checks reviews on the platform, asks a friend via WhatsApp, and purchases on desktop two days later.
Community-integrated. My thesis research found that social feeds, friend recommendations, and influencer content function as discovery triggers, but almost no users (2 of 15) purchased directly from these triggers. They moved to an analytical validation phase, checking reviews, UGC photos, and rating distribution, before deciding. The journey is not linear, it loops through social and analytical modes. This is covered in detail in community evidence as a conversion tool.
Emotionally influenced. Purchase timing is shaped by mood, salience, and social context as much as rational evaluation. A product discovered in a social context may be purchased days later when a delivery discount appears, or abandoned when a competing product appears in a review thread.
These are not edge cases, they are the dominant pattern. The marketing teams reporting campaign performance problems are largely failing because their campaigns were built for the linear model their users had long since abandoned.
Implication 1: Every Touchpoint Must Function as an Entry Point
In a linear model, users arrive at the homepage first, progress to category pages, then product pages. In reality, a substantial share of search results for major brands link directly to user-generated content. Users arrive on product pages from social ads, influencer links, review aggregators, and friend shares, with no context about where they came from or what they saw before.
Every product page must function as a self-contained trust experience. It cannot assume the user has already read the homepage value proposition. Reviews, seller reliability signals, and key product evidence must be visible without navigation.
Implication 2: The Evaluation Stage Happens on Your Product Page, Always
Regardless of how a user arrived, the evaluation stage happens on the product page. This is where community evidence determines whether they proceed. In my thesis research, 14 of 15 participants used reviews and UGC photos as their primary evaluation mechanism, regardless of how they discovered the product.
Designing the evaluation stage means designing the product page for the specific tasks users perform there: accessing review volume and distribution, filtering for UGC photos, checking negative review ratio, and finding seller reliability indicators. These are not secondary features, they are the evaluation infrastructure.
Implication 3: Social Discovery and Analytical Validation Must Be Connected
The gap between a social discovery trigger, such as a reel or a shared link, and the analytical validation phase, meaning reading reviews on the product page, is where most non-linear journeys break. If a user taps a social ad and arrives at a product page where reviews require three scrolls to reach, the social trigger generated attention that the product page failed to convert. Journey mapping that treats social and evaluation as separate stages misses this connection, because they are sequential tasks in the same moment of the user's experience. The integration side of this is covered in how to design a social commerce feature that actually gets used.
Implication 4: Post-Purchase Is a Journey Stage, Not an Afterthought
Journey mapping that ends at purchase is incomplete. Post-purchase engagement determines whether a customer generates the community evidence, meaning reviews and UGC photos, that the next buyer relies on. In my research, 13 of 15 participants consumed reviews extensively but only 4 contributed regularly. The journey is circular: post-purchase contribution feeds the evaluation infrastructure that drives the next user’s decision. The contribution design problem is covered in the post-purchase contribution problem.
Touchpoint Optimisation: The Non-Linear Journey in Practice
Based on both industry data and qualitative findings from my thesis, here is how community features and design decisions map across the real touchpoints of a non-linear shopping journey.
Touchpoint | Channel mix | Community role | Common failure | Optimisation |
|---|---|---|---|---|
Social discovery | Short-form video, reels, friend shares via messaging | Inspiration trigger; exposes users to products outside intended search | Discovery surfaces link to homepage, not the PDP with evidence visible | Deep-link to PDP with reviews in first viewport; include review count in OG metadata |
Search and browse | Platform search, category browse, algorithmic feed | Filtering shortcut; ratings and sold counts narrow the option set | Search results show only star average, no count or popularity signal | Display review count in every result tile; surface sold-this-month badges on high-volume SKUs |
Product page | PDP reviews section, UGC gallery, Q&A | Decision engine; reviews, photos, distribution determine whether to proceed | Reviews below two scrolls; no photo filter; rating shown without count | Reviews within first scroll; photo filter tab; rating histogram; seller reliability adjacent |
Checkout | Cart, payment, order confirmation | Trust confirmation; reliability indicators and clear policies reduce abandonment | Engagement mechanics appear at the highest-intent moment; returns policy buried | Suppress attention-seeking mechanics; surface returns, delivery estimate, response rate |
Post-purchase | Email, SMS, in-app notification | Contribution trigger; generates the evidence the next buyer needs | Generic rate-your-order email 5 days after delivery; no reciprocity framing | Prompt within 24–48h; photo-first minimal format; reciprocity framing |
Re-engagement | Push notification, email, retargeting | Loyalty and advocacy; repeat buyers sustain the evidence system | Promotional rather than community-connected; ignores review history | Surface new reviews on previously purchased items; acknowledge contributor status |
Sources: Emon Datta MSc thesis research, Tampere University (n=15), alongside published industry journey research.
What SaaS Products Can Take From This
SaaS customer journeys exhibit exactly the same non-linear dynamics. A B2B buyer might discover a product through a social post, research it via G2 reviews, watch a competitor comparison video, ask colleagues in a Slack channel, attend a webinar, and sign up for a trial three months later. The linear funnel model is even less accurate for SaaS than for e-commerce, because the consideration cycle is longer and involves more stakeholders.
Every touchpoint functions as an entry point: pricing pages, feature pages, and blog posts are all potential first contacts, and none of them can assume prior context. Evaluation happens wherever the user is: G2 reviews, peer recommendations, case studies, and trial experiences are all evaluation-stage touchpoints, not separate from the purchase flow. Trust evidence must appear at the decision moment: case studies belong near the pricing table, not on a separate customers page, and review scores belong near the CTA, not in the footer. Post-activation is a journey stage: users who activate and experience value are more likely to generate reviews, referrals, and community contributions, which makes activation design a form of journey design.
The onboarding redesign that moved a Morphic client from 34% to 61% activation in 90 days was partly a journey mapping exercise. Users were arriving from multiple sources, paid social, organic search, and referral, with different levels of product context, and the onboarding flow had been designed for a single assumed journey. Redesigning it to function as an entry point for multiple journey states was the structural fix. The patterns behind it are in the 10 mobile app onboarding patterns guide, and the retention consequences in SaaS churn and UX.
The Bottom Line
The linear funnel model produces linear product designs. Modern customer journeys are not linear, they are messy, multi-channel, community-integrated, and non-sequential. Designing for the actual journey means building product pages that function as entry points, connecting social discovery to evaluation infrastructure, and treating post-purchase as the start of the next cycle rather than the end of this one.
Morphic maps the real journey rather than the assumed one, and designs the product and evaluation layers that convert users regardless of how they arrived, through its SaaS design and user research work. Every plan starts with a free 3-day trial before your first invoice. Book a 30-minute call, or see the pricing and recent projects first.
Key Takeaways
The linear funnel is a useful analytical scaffold and a damaging design assumption, because it produces products built for a user who moves forward in sequence and never does.
Users loop repeatedly between exploration and evaluation in the messy middle, returning to earlier stages when a friend shares an alternative or a review raises a concern.
Only 2 of 15 shoppers purchased directly from a social discovery trigger; the rest moved to an analytical validation phase before deciding.
Every product page must work as a self-contained trust experience, because users arrive from social ads, influencer links, and shares with no prior context.
The journey is circular, not linear: post-purchase contribution generates the evidence infrastructure that drives the next buyer's evaluation.
Emon Datta
Founder, Morphic Agency
Morphic Agency is a Melbourne-based UX design agency. We've shipped 50+ products across 15+ countries.
