Skip to content
Aplora
eCommerce and retail
AI Workflow Automation

Client under NDA

Nurturing stopped being one campaign to the whole list

An EU retail business running multichannel nurturing: e-mail, messengers, CRM. The contact base had accumulated over years and stayed undifferentiated — with almost no segmentation in it.

Published: 2026-08-24 · updated: 2026-08-24

A shop assistant with a laptop taking a customer's order on the sales floor
Illustrative photo of the working context, not a screenshot of the client's system

  • Stage-to-stage funnel conversion

    +30–40%

  • Conversion to sale

    +20–50%

In short

  • Problem: an unsegmented base, identical nurturing for everyone, hypotheses taking weeks, and sales unable to read lead warmth.
  • Solution: data-derived segments, action-triggered nurturing per segment, warmth and next task in the CRM, and a fast hypothesis loop.
  • Result: stage conversion up 30–40%, conversion to sale up 20–50% across segments (client figures), at the same ad spend.
  • Systems: CRM, e-mail platform, messengers, analytics.

Context

  • A mid-sized retail business; name and category under NDA.
  • Region: EU.
  • Process volume: the path from a contact in the base to a lead handed to sales.
  • Team: marketing plus the sales team.
  • Systems: CRM, e-mail platform, messengers, analytics; some under NDA.
  • The constraint became visible when raising ad spend stopped producing a proportional rise in sales: traffic grew and conversion fell.

Baseline

Baseline metrics with their sources. Without them, any later result has nothing to be compared against.
  • Before the work started we fixed: stage-to-stage funnel conversion, the share of leads reaching a sale, and the time from campaign idea to launch and assessment.
  • Data source: the client's CRM, e-mail platform and analytics.
  • Ad spend did not change over the period — otherwise the conversion gain could not be separated from added reach.
  • The “+20–50%” range is not the error bar of one measurement but the spread between segments. They behaved differently, and averaging them into a single number would be untrue.

Diagnosis

Which hypotheses were considered, why this one was chosen, what was assumed, and the condition under which we would have stopped.
  • Three hypotheses were on the table: too little traffic, a weak offer, or communication missing the need.
  • The third was chosen: reviewing the base showed that groups with different purchase histories respond differently to the same campaign — and everyone got the same campaign.
  • The assumption: segments distinguishable in the data are also distinguishable in need. Tested against past campaigns — it did not hold for every segment, and two of them were later merged.
  • Stop criterion: if segmented nurturing shows no difference from the general one in the pilot, the hypothesis counts as unconfirmed and the work stops.

What we implemented

  • Data sources: the contact base, purchase history, acquisition channel, behaviour on site and in campaigns.
  • What was built: a data-derived customer profile and segments, contact enrichment, per-segment offers, action-triggered nurturing, warmth and next task in the CRM.
  • AI components: segmentation, offer matching, drafts of personalised messages.
  • Business rules: segmentation and nurturing rules are set by marketing and editable without a developer. A hypothesis scales only after the pilot confirms it — otherwise an unverified campaign goes to the whole base.
  • Hypothesis testing sped up by roughly an order of magnitude: assembling a per-segment campaign takes hours instead of weeks. That is a property of the process, not a business result, which is why it sits here rather than among the metrics.
  • Human checkpoints: the copy that goes to the base is signed by a person. Segmentation is automatic; a promise to a customer is not.

How the process changed

Before

5 steps
  1. One campaign to the whole base
  2. A customer profile from gut feel
  3. A campaign launched by hand
  4. Results assessed weeks later
  5. Leads go to sales with no readable warmth

After

7 steps
  1. The base and purchase history as the source
  2. Segments described from data
  3. An offer matched to the segment
  4. Nurturing triggered by the customer's action
  5. Warmth and next task in the CRM
  6. Handover to sales with a priority
  7. A confirmed hypothesis gets scaled

What was stuck

The customer profile was described from gut feel, nurturing was launched by hand and identically for everyone, and hypotheses took weeks to test. Sales received leads with no readable warmth and spent time on people who were not ready.

  1. 1The base and purchase history as the source
  2. 2Segments described from data
  3. 3An offer matched to the segment
  4. 4Nurturing triggered by the customer's action
  5. 5Warmth and next task in the CRM
  6. 6Handover to sales with a priority
  7. 7A confirmed hypothesis gets scaled
  8. Measured result

Results

Stage-to-stage funnel conversion

+30–40%

Reported by: Client figures: conversion trends after segmented nurturing went live

Conversion to sale

+20–50%

Reported by: Client figures: observed across different segments, hence the wide range — this is not a single measured value

Economic impact

  • The effect is additional revenue at unchanged ad spend: the same outlay, a different conversion.
  • It is the least reliable of the three effect types: conversion depends on more than the initiative — season, price and demand mix move on their own. Hence ranges rather than point values.
  • No money figure is published: average order value and margin are the client's commercial information.
  • Cost of ownership: segments age with the base. Without revision, a year on the nurturing runs against a customer profile that no longer exists.

Adoption

  • Marketing stopped depending on launching campaigns by hand.
  • Sales began trusting lead warmth — before that the CRM priority was ignored, because it rested on nothing.
  • Hypothesis testing became a routine operation rather than a separate project.
  • The solution owner is the head of marketing.
«Hypotheses used to take weeks and the funnel ran on gut feel. Now everything is segmented, nurturing is automated, and we quickly see what actually produces sales.»

Client under NDA — Head of marketing, retail (NDA)

What's next

  • Scaling: segments for repeat purchases, not only for the first one.
  • Next initiative: linking nurturing to stock levels — offering what is in stock rather than what has run out.
  • What we decided against: raising campaign frequency. More frequency buys a quick lift and a slow loss of the base, and the second half does not show up in a quarterly report.

Have a similar workflow? Let's check whether the hypothesis transfers

Another company's result is not a promise. It does show where to look.

All cases