Segmentation & Data Strategy
Clean data and smart audiences, so the right message reaches the right customer at the right time. We cover data models, lifecycle segments and SQL audiences through to CDPs, real-time profiles and AI-powered predictive segmentation.
A data foundation that every campaign can rely on
Great automation runs on great data. Before building a single segment, we work out what data you have, where it lives, what you are allowed to use and what your marketing actually needs. Then we design an architecture that fits your business, whether that is your marketing platform alone, a CDP, or your data warehouse at the centre.
- Data audit & use-case mapping: every source (CRM, e-commerce, web, app, POS, loyalty, support) mapped to the campaigns and segments it powers
- First-party & zero-party data: capturing data customers share directly, through preference centres, quizzes, sign-up and loyalty programs, as third-party cookies lose value
- Architecture choice: marketing platform only, a packaged CDP, or a composable, warehouse-native setup, chosen on cost, speed and team skills
- Data dictionary & standards: shared definitions, naming conventions and a single source of truth for each attribute
- Consent & governance: consent captured with every record, purpose-based access, and GDPR, CCPA and DPDP-ready retention policies
A data model built to scale
A good data model makes segmentation fast and personalisation simple. A poor one leads to duplicate contacts, slow queries and workarounds. We design structures that match how your business works and grow with it.
- Subscriber & contact keys: one stable ID per customer across every system, so profiles never split or duplicate
- Relational Data Extensions: customers, orders, order lines, products and events in SFMC, linked in Contact Builder attribute groups
- Platform-native models: Braze custom attributes, events and Catalogs, Marketo person, company and custom objects, and Emarsys contact fields with sales and product data
- Sendable vs reference data: lean sendable tables for journeys, with history and lookup data kept separately for speed
- Data feeds & refresh: scheduled imports, API and streaming updates, field validation and retention rules that keep tables clean
Audiences based on what customers do
We start with proven segmentation models and add behavioural signals, so every audience reflects where a customer is in their relationship with you and what they are likely to do next.
- Demographic & firmographic: location, age group, language, and for B2B, industry, company size, role and account
- Lifecycle stages: prospect, new, active, loyal, at-risk, lapsed and win-back, with automatic movement between stages
- RFM & value tiers: recency, frequency and monetary scoring to find your best customers and those slipping away
- Behavioural & intent signals: browsing, product views, cart activity, app usage, search and content engagement
- Engagement tiers & lead scoring: email and app engagement levels for B2C, and behavioural and demographic scoring for B2B lead handover
Automated audiences that refresh themselves
Drag-and-drop segment builders cover simple audiences. For complex logic across orders, products and events, we write well-tested SQL and automate it, so your audiences are always up to date without manual list pulls.
- SFMC SQL Query Activities: joins across Data Extensions and data views, scheduled in Automation Studio
- Braze SQL Segment Extensions: SQL over event and purchase history for audiences the standard filters cannot build
- No-code segment builders: Marketo Smart Lists, Emarsys segments, Braze segments and Data Cloud segments, set up so marketers can self-serve
- Calculated attributes: lifetime value, average order value, days since purchase and favourite category, computed once and reused everywhere
- Performance & QA: query tuning, row-count checks, failure alerts and documentation for each audience
One profile per customer, available in real time
When customer data is spread across many systems, a CDP or a warehouse-native setup brings it together into one profile and sends audiences to every channel. We help you choose, implement and get value from it.
- Identity resolution: deterministic and probabilistic matching across email, phone, device, loyalty and CRM IDs
- Unified profiles: CRM, web, app, commerce, service and offline data combined into one customer view
- Real-time & streaming segments: audiences that update the moment a customer acts, ready for instant journeys
- Zero-copy & composable data: segmenting directly on Snowflake, Databricks or BigQuery without copying data, using Salesforce Data Cloud zero-copy or tools like Hightouch
- Audience activation: the same segments sent to email, SMS, app, web personalisation, paid media and sales teams
Audiences built on what customers will do next
Traditional segments describe the past. AI helps you predict the future and find patterns people would miss. We set up and govern the AI in your platform and build custom models where they add value.
- Predictive scores: churn risk, purchase propensity, predicted lifetime value and engagement likelihood for every customer
- AI clustering & lookalikes: machine learning that finds natural customer groups and new prospects who resemble your best customers
- Natural-language segment builders: describe an audience in plain words and let AI build it, using tools such as Salesforce Agentforce, Adobe AI Assistant and BrazeAI
- Next-best-action & real-time decisioning: AI that chooses the best offer, message or channel for each customer at each moment
- AI agents for data operations: agents that monitor data quality, suggest new segments and explain why audiences change
- Privacy-safe collaboration: data clean rooms and consent-aware AI, so you can use more data without putting customer privacy at risk
Platforms we work with, and where each one leads
There is no single best platform for segmentation. Salesforce leads on relational data and the CRM-connected profile, Braze on real-time behavioural audiences, Adobe on B2B and enterprise real-time CDP, and SAP Emarsys on ready-made retail lifecycle segments. Composable, warehouse-native CDPs are growing fastest for companies that already invest in Snowflake, Databricks or BigQuery. We are platform-agnostic and recommend what fits your data, team and budget.
Salesforce Marketing Cloud & Data Cloud
- Contact Builder & Data Extensions: the most flexible relational data model among the major marketing platforms
- SQL & Automation Studio: unlimited audience logic across Data Extensions and system data views
- Data Cloud (Data 360): identity resolution, unified profiles, calculated insights, streaming segments and zero-copy access to your data warehouse
- Einstein & Agentforce: engagement scoring, predictive insights and AI-assisted segment creation from plain-language prompts
- Where it leads: complex B2C data and businesses already on Salesforce CRM. The trade-off: SQL skills are needed, and classic SFMC segmentation runs in batches
Braze
- Real-time segments: audiences that update instantly from app, web and server events
- Custom attributes, events & Catalogs: rich behavioural profiles with product and content data for personalisation
- Segment Extensions & Cloud Data Ingestion: SQL-based audiences and direct sync from Snowflake, Databricks, BigQuery and Redshift
- BrazeAI predictive audiences: predictive churn and purchase scores used directly in segments and Canvas journeys
- Where it leads: real-time behavioural segmentation for mobile apps and digital brands. The trade-off: it uses one profile per user rather than a relational model, so complex data is best kept in a warehouse
Adobe Marketo Engage & Real-Time CDP
- Smart Lists & segmentations: rules-based audiences on person, company, activity and custom object data
- Lead & account scoring: behavioural and demographic scores, lifecycle stages and MQL handover to sales
- Account-based marketing: target account lists, account profiling and buying-group segmentation
- Adobe Real-Time CDP & AI: real-time B2B and B2C profiles, predictive audiences and Adobe AI Assistant
- Where it leads: B2B lead and account segmentation, and enterprise real-time CDP when paired with Adobe Experience Platform. The trade-off: the full Adobe stack needs a larger budget and team
SAP Emarsys
- Smart Insight: ready-made customer lifecycle and RFM segments, generated from your sales data
- Predict & AI segments: AI-driven product affinity, purchase likelihood and churn-risk audiences
- Commerce data integration: product catalogue, sales and web behaviour data from SAP Commerce, Shopify and other platforms
- SAP data connections: links to the SAP Customer Data Platform and wider SAP business data
- Where it leads: retail and e-commerce teams that want lifecycle segmentation working quickly. The trade-off: less freedom for custom data models than SFMC or a warehouse-native CDP
Other platforms & CDPs
We also work with Adobe Real-Time CDP, Twilio Segment, mParticle, Tealium, Bloomreach, Klaviyo, HubSpot, Iterable and MoEngage, and with composable, warehouse-native tools such as Hightouch and Census on Snowflake, Databricks and BigQuery. Twilio Segment and mParticle lead on clean event collection, Tealium on real-time consent-aware data, Klaviyo on built-in data for growing e-commerce brands, and composable CDPs on flexibility and cost for data-mature teams.
Core capabilities we deliver
From the basics to AI, these are the building blocks of a data and segmentation strategy that works. We make sure each one is in place.
Data audit & roadmap
A clear picture of your data sources, gaps and quality, with a prioritised plan linked to business goals.
Scalable data model
Contact keys, relational tables and attribute design that keep segmentation fast and accurate as you grow.
Identity resolution
One customer, one profile, matched across email, phone, device, loyalty and CRM records.
Lifecycle & RFM segments
Proven segmentation models that show who your best customers are, who is at risk and who to win back.
SQL & audience automation
Tested, documented queries and calculated attributes that refresh automatically on schedule.
Real-time segmentation
Streaming audiences that react the moment a customer browses, buys or goes quiet.
Predictive & AI audiences
Churn, propensity and lifetime value scores, AI clustering and lookalikes that target customers by what they will do next.
Composable & warehouse-native
Segmentation directly on your data warehouse with zero-copy and reverse ETL, without duplicating data.
Consent & data governance
Consent tracked with every record, role-based access, data quality rules and GDPR, CCPA and DPDP-ready retention.
Segment performance measurement
Holdout groups and incrementality testing that show which segments actually drive extra revenue.
How we work
A step-by-step approach that fixes the data foundation first, then builds audiences that get smarter over time.
- Data & segmentation audit: sources, data quality, current model, audiences and how they perform
- Strategy & architecture: use cases, target data model, platform or CDP choice and governance rules
- Model & integrate: contact keys, tables, data feeds, identity resolution and validation
- Build audiences: lifecycle, RFM, behavioural and SQL segments with calculated attributes
- Add AI & real time: predictive scores, streaming segments and next-best-action
- Measure & refine: holdout tests, segment reporting, data quality monitoring and team training
What you get
- A documented data strategy, architecture and data dictionary
- A clean, scalable data model with one profile per customer
- A library of lifecycle, behavioural and value-based segments
- Automated SQL audiences and calculated attributes that refresh themselves
- Predictive and AI-powered audiences, set up with clear guardrails
- Consent and governance in place, and a team trained to self-serve
Ready to turn your customer data into smarter audiences?
Tell us about your data sources, platform and goals. We will come back with a clear data strategy, timeline and estimate.
Book a Free Data Audit