Defining the use case
Set the goal, desired segment count, business context, and the KPIs your segments are measured against.
The Definition tab is where you tell Neuralift what you’re trying to achieve. The definition records your goal, desired segment granularity, business context, and KPIs; Neuralift uses it to frame the engagement and to shape how results are named, explained, and prioritized. A clear goal, the right segment-count range, and rich business context are what separate generic clusters from segments your team can act on. Neuralift’s team captures and refines the definition with you (see Roles & permissions), and everyone who can see the use case can read it.
Setting the goal
The Goal is the specific outcome this segmentation is designed to support, for example reduce churn, re-engage dormant users, or customer discovery. Keep it to one or two sentences and make it a business outcome, not a data task.
The goal frames every AI-generated output downstream: segment summaries, insights, segment tactics, and actions are all written against it. A vague goal (“understand our customers”) produces generic narratives; a sharp one (“reduce churn among first-year subscribers”) produces segments and recommendations your team can take straight into planning.
Choosing desired segments
Desired Segments sets the granularity of discovery, the range Neuralift aims for when partitioning your customers:
| Size | Segments | Best for |
|---|---|---|
| Small | 4–12 | Strategic overviews: a handful of broad segments leadership can hold in their heads. |
| Medium | 12–20 | Balanced portfolios: enough distinction for differentiated campaigns without overwhelming the team. |
| Large | 20–40 | Fine-grained targeting: precise audiences for personalization at scale. |
| Extra Large | 40+ | Micro-segmentation: maximum resolution for large customer bases and highly automated activation. |
Fewer segments means each one is bigger and easier to action but blends more behavior together; more segments means sharper distinctions but more work to activate each one. New use cases default to Small. If you’re unsure, start smaller; you can clone the use case and rerun at a finer size once you’ve seen the first landscape.
Adding business context
The Business Context section has three fields. None of them change your data. They inform how the AI interprets segments and writes narratives, so the outputs use your language and respect your realities:
- Products / Transactions: your products and services, the transactional nature of your business, and any offers, discounts, or subscriptions relevant to this use case. This helps the AI interpret purchase and usage features correctly when describing what makes a segment distinct.
- Channels / Activations: the campaigns, channels, and activations where these segments will be used for targeting and personalization. Actions and their plans are more usable when the AI knows whether you activate through email, paid social, in-app messaging, or a call center.
- Temporal / Seasonal: trends, seasonality, and tent-pole events relevant to your data or your activation window. Without it, a holiday spike can be misread as a durable behavior change.
Write these the way you’d brief a new analyst on your first day working together. Specifics beat generalities: product names, real channel mixes, actual seasonal peaks.
Managing KPIs
The KPIs card lists the metrics your segments are measured against, in priority order. Each KPI should map to a feature in your data or be derivable from existing features. To add one, click Add KPI and fill in:
| Field | What to enter |
|---|---|
| Priority | The KPI’s rank. Priority 1 is your headline metric. |
| KPI Name | The metric, named to match the feature in your data, for example revenue, churn_flag, ltv. |
| Definition | How the KPI is calculated, in a sentence or a formula. |
| Objective | What good looks like: maximize, minimize, or a specific target. |
| Has lift potential | Enable when this KPI can be used to estimate incremental value, impact, or opportunity for lifted customers. |
Your prioritized KPIs shape how results are profiled and presented: every discovered segment is measured and compared against them; they become the KPI lenses you use to recolor the segment landscape; and each segment’s detail panel ranks its KPI priorities so you can see where it over- or under-indexes. KPIs marked with lift potential additionally unlock what-if lift modeling in the landscape.
Note. A better definition means sharper results. Neuralift uses every field here to shape how segments are named, explained, and prioritized. Ten minutes spent on context typically saves hours of interpreting output that missed the point.