What is personalisation marketing and how does it differ from segmentation?+
Segmentation divides a customer base into groups and serves each group a shared message the same email to all members of the "lapsed customers" segment, the same homepage to all visitors from paid search. Personalisation serves each customer a distinct experience based on their individual attributes, history, and behaviour not the shared characteristics of the group they belong to. AI personalisation uses machine learning to predict what each individual customer is most likely to engage with and purchase, replacing the segment-average approximation with individual-level precision. The commercial difference: personalised recommendations consistently outperform segmented ones by 20-50% on engagement metrics, because the individual-level model captures preference patterns that segment-level generalisation misses.
How much data do I need to start personalising?+
Data requirements depend on the personalisation approach. Segmentation-based personalisation (serving distinct experiences to rule-defined segments) can be effective with modest data volumes enough to define meaningful segment criteria and collect sufficient behavioural signals to assign each user to a segment. Collaborative filtering recommendation models typically require at minimum 1,000 users with 10+ interactions each to produce reliable recommendations below this volume, the model has insufficient signal and falls back to popularity-based recommendations. More sophisticated personalisation (deep learning models, real-time behaviour-driven personalisation) requires larger data volumes. We assess data readiness at the start of every personalisation engagement and recommend the approach appropriate for the current volume.
What personalisation platforms do you work with?+
We implement personalisation across the major platforms: for e-commerce email personalisation, Klaviyo and Salesforce Marketing Cloud; for website personalisation, Optimizely, Dynamic Yield, and Mutiny (B2B); for product recommendations, custom-built systems (Python/MLflow) or managed platforms (Recombee, Barilliance) depending on the scale and integration requirements; for SaaS in-app personalisation, Intercom, Customer.io, and Appcues; for advertising personalisation, Meta Dynamic Product Ads, Google DCO, and LinkedIn's Dynamic Ads. We also build custom personalisation systems where the use case requires capabilities beyond what managed platforms provide.
How long does it take to see results from personalisation?+
Simple personalisation interventions (predictive send-time optimisation, basic product recommendation widget deployment) typically show measurable lift within 2-4 weeks of deployment. More sophisticated personalisation (collaborative filtering recommendation models, use-case-segmented onboarding flows, website experience personalisation) typically requires 4-8 weeks to deploy and 4-6 weeks of post-deployment data collection before the statistical significance required for reliable lift measurement is achieved. Complex personalisation programmes (multi-channel, multi-touchpoint personalisation with A/B testing across multiple variants) typically produce the most reliable lift measurements after 3-4 months of operation sufficient time for the model to have learned from the live environment and for the statistical confidence to accumulate across the test variants.
How do you handle personalisation and data privacy?+
Personalisation based on individual behavioural data must comply with GDPR, CCPA, and applicable privacy regulations. We implement privacy-compliant personalisation: using only data collected with appropriate consent and disclosed in the privacy notice, implementing cookie consent that respects the user's preferences for personalisation cookies, providing opt-out mechanisms for personalisation-based tracking, and designing personalisation systems that function within the data that users have consented to provide rather than depending on tracking methods that exceed the consent obtained. For markets where privacy regulation is strict (EU, UK, California), we design personalisation systems that are effective within the first-party data the business has legitimately collected.
Can personalisation work for B2B companies or is it primarily for e-commerce?+
Personalisation is commercially valuable for B2B companies the applications are different from B2C but equally impactful. B2B personalisation applications: account-based website personalisation (serving industry-specific content and case studies to visitors from target companies), personalised email sequences calibrated to the buyer's role and stage in the evaluation process, personalised SaaS product experiences calibrated to the user's role and use case (as in the onboarding personalisation case study above), and content recommendation for B2B content hubs (serving the specific whitepapers, case studies, and webinars most relevant to each subscriber's role and topic interests). B2B personalisation is often more immediately impactful per customer than B2C because the individual transaction values are higher and the buying process is more research-intensive.
How do I measure the ROI of personalisation?+
We establish personalisation ROI measurement frameworks before deployment. Standard metrics by personalisation type: for product recommendation systems, revenue attributed to recommendation interactions and average order value lift for recommendation-assisted orders; for email personalisation, revenue per email, open rate lift, and click-to-purchase conversion lift versus non-personalised control group; for website personalisation, conversion rate lift and revenue per session versus non-personalised control; for onboarding personalisation, trial-to-paid conversion rate and time-to-first-value. We run A/B tests comparing personalised and non-personalised experiences for each personalisation implementation, enabling clean attribution of the revenue lift to the personalisation programme.
How do I get started?+
Book a free personalisation assessment. We review your current personalisation capability (what data you have, what tools you use, what personalisation is already in place), identify the specific personalisation gaps producing the largest commercial cost (the highest-ROI personalisation interventions for your specific business model), and produce a prioritised implementation roadmap. No commitment required at the consultation stage.