AI 1-to-1 Delivery

AI 1-to-1 Delivery is the AI-powered planning path for large-scale personalization. In the current phase, it focuses on the Coupons Bank use case: the team provides a bank of coupon or discount IDs, selects the business goals, and Loyal Guru builds the campaign structure, audiences, coupon distribution, budget allocation, and forecast review automatically.

Use AI 1-to-1 Delivery when the business wants to scale many personalized coupon campaigns without manually creating every audience and action. The operator still reviews the generated campaigns and forecast before execution.

info

This page has moved

AI 1-to-1 Delivery is now documented as its own Campaign Personalization group. Use AI 1-to-1 Delivery for the complete guide, including business goals, prediction models, Coupons Bank, Forecast KPIs, and ROI KPIs.

Configuration

Area	Current capability
Plan description	Name, availability dates, printing range, non-printable coupons, and optional ROI calculation schedule
Business goals	Up-selling and Cross-selling are active. Churn and Increasing Frequency are visible as future goals but disabled in the current flow
Coupons Bank	Select discount/coupon IDs manually or import them from a CSV whose first column contains IDs
Budget / contribution	Budget used to size the audience and distribute coupon assignments when budget-based allocation is enabled
Restricted Audience	Optionally limit the whole AI Plan to one existing Audience
Up-selling period	Date range used by the PBS model to identify existing category buyers and split them by purchase behavior
Cross-selling periods	Recency windows used to target Customers who have not bought a category recently: 1, 3, 6, or 12 months, or never purchased
Cross-selling filters	Optional flags for Customers who bought in the last 12 months and active Customers
Advanced configuration	Plan-level overrides for up-selling percentile splits and cross-selling Customer selection threshold

How It Works

1. Create a new AI 1-to-1 Delivery from Campaign Personalization → AI 1-to-1 Delivery.
2. Fill in the plan description: name, availability, printing range, and optional ROI calculation schedule.
3. Select one or both active business goals: Up-selling and Cross-selling.
4. Select or import the Coupons Bank. The current importer accepts pasted IDs or a CSV with discount IDs in the first column.
5. Optionally select a restricted Audience for the whole plan.
6. Configure the goal-specific inputs: up-selling period and/or cross-selling periods.
7. Optionally open Advanced configuration to override default distribution thresholds.
8. Click Validate discounts bank. The plan is saved and the coupon bank is validated.
9. When validation passes, click Save and continue. Loyal Guru generates AI campaigns, calculates forecast, and opens the AI Plan detail.

During generation, the Owner App shows a loading screen while the platform analyzes purchases, maps coupon affinities, optimizes redemption probability, analyzes Customer affinities, and distributes coupons. The app polls until generation and forecast are ready, with a timeout guard.

Business Goals

Goal	Current status	What the AI does
Up-selling	Enabled	Uses Purchase Behavior Score (PBS) to target Customers who already buy the category and assign different coupons by behavior tier
Cross-selling	Enabled	Uses the product recommender / cross-selling model to target Customers who have not bought the category recently
Churn	Disabled	Future goal; visible in the UI but not active in the current flow
Increasing Frequency	Disabled	Future goal; visible in the UI but not active in the current flow

Prediction Models Used

AI Plans use predictive models to decide which Customers should receive which coupon, but the Owner App does not expose model formulas. The operator sees the business inputs, generated campaigns, audiences, forecast, and budget impact.

Model / signal	Used for	How it affects the plan
Purchase Behavior Score (PBS)	Up-selling	Identifies Customers who already buy the affected category and ranks them by affinity or purchase behavior.
Product recommender / cross-selling model	Cross-selling	Finds Customers with high probability of buying a category they have not purchased recently.
Coupon affinity and distribution rules	Coupon assignment	Matches coupon strength to Customer affinity: stronger Customers usually receive weaker incentives; lower-affinity Customers receive stronger incentives.
Forecast estimates	Forecast and budget sizing	Estimate delivery, redemption, cost, incremental sales, and ROI for generated actions.

Coupons Bank Validation

The Coupons Bank must be validated before the plan can generate campaigns. Validation checks whether the selected IDs can be used for AI distribution.

Validation outcome	What it means
No valid coupons	Blocking issue. The plan cannot be built if no selected IDs map to valid coupons
No product/category affectation	Blocking issue. The plan needs coupons tied to product category or product affectation
Cart-level affectation	Warning. Coupons with full-cart affectation are ignored for AI campaign generation
More than 5 coupons in one category group	Warning. Only up to 5 coupons are used for up-selling segments in that category
Duplicate discount and minimum purchase condition	Warning. When several coupons have the same discount and minimum condition, the backend keeps one and ignores the rest

In the Owner App, validation results are shown in a modal with the coupon and the warning or ignored reason. Save and continue remains disabled until validation passes for the current form state.

Coupons Bank Loading

The Coupons Bank is the set of coupons or discounts the AI can distribute. Operators can paste IDs manually or import a CSV. The importer reads discount IDs from the first column and caps the import size to keep validation and generation manageable.

Before generation, the bank is normalized by product/category affectation. Coupons with compatible affectation are grouped together so the platform can create campaign actions per category or product group. Coupons that do not provide usable product/category scope are ignored or blocked during validation, depending on the issue.

Coupon Distribution Logic

Coupons are grouped by product category or product attribute. Each group becomes an AI campaign, and the selected business goals determine how actions are created inside that campaign.

Up-selling

Up-selling targets Customers who already buy a category. The platform uses PBS tiers to split the audience and assign incentives.

Rule	Behavior
Audience model	PBS model for Customers who already purchase the target product category
Coupons used	Up to 5 coupons per category group
Segmentation	One segment per PBS tier, based on configured percentiles
Incentive strategy	High-propensity Customers receive weaker coupons; lower-propensity Customers receive stronger coupons
Plan override	Advanced configuration can override company default percentile splits for 2, 3, 4, and 5 coupons

The default UI example uses these company-level splits unless overridden for the plan:

Coupons per campaign	Default split
2	20% / 80%
3	20% / 20% / 60%
4	10% / 20% / 20% / 50%
5	10% / 10% / 20% / 20% / 40%

Cross-selling

Cross-selling targets Customers who have not bought the category recently and ranks them by purchase probability.

Rule	Behavior
Audience model	Product recommender / cross-selling model
Coupons used	One best coupon per category group
Segmentation	One action per selected recency period
Periods	1 month, 3 months, 6 months, 12 months, or never purchased
Customer selection threshold	Advanced configuration can override the company default percentage of highest-probability Customers to include

Budget-Driven Allocation

AI Plans can work from a budget or contribution assigned to the Coupons Bank. Instead of starting with a fixed audience and calculating the cost afterwards, the allocation works backwards: the budget helps decide how many Customers can receive coupons and how those Customers are distributed across goals, categories, and coupons.

At a high level, budget allocation follows this cascade:

1. Business goal split — if the plan has more than one goal, budget is split between goals, such as up-selling and cross-selling.
2. Affectation split — within a goal, budget is distributed across product/category affectations based on their business weight, such as sales weight in the selected period.
3. Coupon split — within each affectation, Customers are distributed across coupons using the configured distribution percentages and coupon priority.
4. Audience sizing — the platform estimates how many impacted Customers fit in the allocated budget, using expected delivery, redemption, redemption frequency, and average coupon cost.

The percentages configured for coupon distribution are Customer distribution percentages, not budget percentages. The platform keeps the intended Customer mix and uses the estimated average cost of that mix to size the audience.

Allocation Rules

Rule	Behavior
Best-affinity Customers first	Customers are ranked by model affinity. If budget is limited, lower-affinity Customers are left out first.
Lower incentive for stronger affinity	In up-selling, the weakest coupon is assigned to the highest-affinity Customers, and stronger coupons are reserved for Customers who need more incentive.
Budget left over	If there are fewer eligible Customers than the budget can cover, all eligible Customers can be included and the remaining budget is shown as unused.
Control group	Control-group Customers are excluded from delivery and do not consume coupon budget.

Budget Assumptions and Limits

Budget allocation is an estimate. In the current model, Customers assigned to the same coupon are treated with an average expected cost for that coupon. Real cost can vary because Customers do not all redeem with the same probability or spend the same amount.

This is especially sensitive for percentage coupons without a minimum purchase, where the cost depends on Customer spend. Future improvements may make the cost curve more granular by affinity percentile instead of using one average cost per coupon.

Generated Campaigns and Forecast

After validation and generation, the AI Plan detail currently exposes:

Tab	What it shows
Campaigns	Generated AI campaigns, business goals, Customers, discount cost, incremental sales, ROI, approval status, and actions to review Audience or Forecast details
Forecast	Plan-level forecast totals and per-campaign forecast rows

The Campaigns tab aggregates:

KPI	Meaning
Customers	Total Customers expected to be impacted by generated campaigns
Discounts distributed	Total generated coupon assignments across the plan
Discount cost	Forecasted coupon/discount investment
Inc. Sales	Forecasted incremental sales
ROI	Estimated return based on incremental sales over discount cost

The Forecast tab groups KPIs into volume and efficiency:

KPI	Meaning
Impacted Customers	Customers included in generated actions
Distributed Promotions	Coupons expected to be delivered
Coupons per Customer	Average number of coupons assigned per impacted Customer
Customers who redeem	Expected unique redeemers
Promotions Investment	Forecasted discount investment
Investment per Customer	Forecasted investment divided by impacted Customers
Inc. Sales per Customer	Forecasted incremental sales per impacted Customer
Total Incremental Sales	Total forecasted incremental sales
Incremental per Customer	Incremental value per impacted Customer
Estimated ROI	Estimated return ratio for the plan

Current Scope and Future Phases

This page documents the current AI Plans implementation and the Phase 0 Coupons Bank scope.

Area	Current status
AI Plan creation/editing	Implemented
Coupons Bank import and validation	Implemented
Automatic campaign generation	Implemented through asynchronous generation after Save and continue
Plan-level Campaigns and Forecast tabs	Implemented
Campaign-level Audience and Forecast detail screens	Present in the app, but still evolving; use the plan-level Campaigns and Forecast tabs as the main operator reference
Budget-based audience sizing	Documented as the allocation model for Coupons Bank. UI/activation details may depend on rollout configuration
ROI tab	Planned for a later phase; route is not active in the current AI Plan detail
Analysis / pre-post comparison tab	Planned for a later phase; route is not active in the current AI Plan detail
Automatic budget enforcement after launch	Future phase; budget allocation forecasts expected spend, but live enforcement and later automatic budget control remain separate rollout scope
Additional recipes	Future phases may include churn, frequency, automatic coupon creation, coupon scoring, and microsegmentation

Configuration

Go to Campaign Personalization → AI 1-to-1 Delivery.

Before creating an AI 1-to-1 Plan, make sure:

1. The company has Intelligent Delivery and AI Plans enabled.
2. The Coupons Bank contains valid coupon/discount IDs.
3. Coupons are active, not archived, and have product or category affectation.
4. Company-level AI Plan defaults are configured for up-selling percentiles and cross-selling threshold.
5. The data required by PBS and cross-selling models is available for the company.
6. If the plan should be limited, the restricted Audience already exists.