ASBIS · Decision Intelligence
Confidential
Board Briefing Note

CEO Follow-Up Brief

From the Executive Briefing to a decision: should ASBIS run a pilot?
Prepared forSerhei Kostevitch, CEO · ASBIS Group
FollowsExecutive Briefing · decision.d1s.app
DateJune 2026
DecisionApprove / decline a 60-day pilot
01

Executive Summary

ASBIS already owns the two things this initiative needs: decades of its own commercial data, and a working intelligence platform that is already in production. What it does not yet have is the layer that turns those into forward decisions. This brief asks for one thing — a bounded pilot to test whether it does.

Why continue

The platform is not a concept. It is operational today, ingesting a full market panel automatically and surfacing signals that currently take analysts days to assemble. Continuing does not start something new — it points an existing, working capability at ASBIS's own decisions.

Why now

The cost and time to build and validate this kind of capability has fallen sharply — what required several teams and a year can now be tested in weeks. A working prototype already exists. Every distributor sees similar market data; advantage goes to whoever converts it into decisions first, while testing is still cheap.

Why the risk is limited

It is additive. It sits on top of IT4Profit and Power BI, changes no operational system, and is proven first in a 60-day pilot with a capped budget and an explicit stop point. If the pilot fails its criteria, it ends — the only cost is the pilot.

Why the upside is meaningful

The decisions in scope — how much to buy, which brands and categories to back, where margin leaks — are the decisions that move inventory, margin and growth. On high-value categories, even a modest improvement in decision quality is material.

Recommendation

Approve a 60-day Customer Opportunity Intelligence pilot — one customer segment, one sales team — with success criteria agreed in advance. Judge it on evidence at day 60 before any larger commitment.

Status
Prototype live
The ask
60-day pilot
Scope
Customers · one segment
Budget
Capped envelope
Decision point
Day-60 go / no-go
02

What Has Already Been Built

This initiative is past the idea stage. The following components exist and work today.

ComponentWhat it isStatus
Market-intelligence platformAutomatically ingests a full market panel — ~1.2M SKUs, ~30K brands, 50 categories, 18 months of history — and turns it into self-serve views. Work that took an analyst days now runs in under a minute.Live
Normalized data layerA single structured model — brand · SKU · category · channel · time — that any new source plugs into. The hard, reusable part, already done.Live
Opportunity-discovery conceptsWorking analyses that flag channel shifts, pricing / premiumization trends and brand concentration — concrete examples of signals surfaced automatically.Working
Attribute-intelligence researchEarly work reading the market by product characteristics (e.g. specification adoption), not just by brand — the basis for understanding why markets move.Research
Executive briefing portaldecision.d1s.app — the access-controlled, bilingual briefing you reviewed.Live

In plain terms: the engine runs. Today it runs on external market data; connecting it to ASBIS's own internal data is exactly what the pilot does.

03

What Problem Are We Solving?

Not a technology problem. These are the recurring commercial questions ASBIS answers every quarter — today mostly on experience and backward-looking reports:

The constraint is no longer collecting information or reporting it. It is converting what we already know into the next decision — reliably, and earlier than competitors.

04

What Makes This Different From Existing BI?

One framework answers most of the overlap concern. The business has three layers; ASBIS is strong in the first two and missing the third.

Layer 1
Operations
Running the business. e.g. IT4Profit. Not replaced.
Layer 2
Visibility
Seeing what happened. e.g. Power BI. Not replaced.
Layer 3
Opportunities
Knowing what to do next. The missing layer.

Why this is not another Power BI dashboard

Power BI reports what happened. This layer recommends what to do next and ranks opportunities — combining ASBIS internal data with external market signal that reporting tools do not hold. Different job, sitting one level up.

Why this is not another reporting project

Success is measured in better decisions, not delivered reports. A reporting project ends when the dashboard ships; this is judged only by whether a buyer or category manager made a better call.

Why this is not a replacement for IT4Profit

It reads permissioned, read-only copies of data. It writes nothing back, changes no operational process, and introduces no new system of record. IT4Profit remains the operational backbone, untouched.

05

Pilot Proposal

Deliberately small and measurable: prove the principle with one customer segment and one sales team, end to end, in 60 days. The first output of the Opportunity Engine — Customer Opportunity Intelligence: what else we can profitably sell the customers we already have.

PilotCustomer Opportunity Intelligence
CountryOne country — chosen for the cleanest customer data and a willing sales director
Customer segmentOne segment — e.g. mid-tier resellers, where whitespace is largest and reps have capacity to act
Sales teamOne team (~6–12 reps), under one sales director
Duration60 days
FocusThree play types only: cross-sell, customer whitespace versus comparable accounts, and reactivation of lapsing accounts.
GoalValidate whether opportunity discovery — on ASBIS's own customer and sales data — produces better commercial decisions, measured in incremental gross profit, than current practice.
ScopeUse existing internal sales, customer and product data to produce ranked per-rep opportunity lists and a back-test of recent decisions. Internal data only — no external market panel required for the pilot.
MethodTest versus control: a matched set of reps/accounts receives recommendations, a comparable set does not, so the lift is cleanly attributable.
ResourcesOne initiative lead (owner); part-time, controlled data-access support from IT; an executive sponsor; read-only permissioned data; modest tooling and compute.
CostA defined, capped envelope — a fraction of a traditional BI engagement — to be finalised with Finance. Components: lead time, limited data engineering, tooling.

Expected outputs

Success criteria — agreed before we start

Stop point

If the criteria are not met at day 60, the pilot ends. The cost is bounded and there is no further commitment — by design.

06

Governance & Risk

The risks are real and worth stating plainly. None of them require accepting them unmanaged.

AreaPosition & mitigation
Data ownershipASBIS owns all data. The layer reads permissioned, read-only copies and writes nothing back. No new system of record is created.
SecurityData remains within an ASBIS-controlled environment; external market data is used under its licence terms; everything aligns to ASBIS's existing data policy.
Access controlRole-based, least-privilege access. Pilot data is scoped to the one segment and one team only — nothing wider is touched.
Continuity & key-person riskThe real risk. Speed has come from a small, focused build, which concentrates knowledge in few hands. Mitigation: document the data model and method; pair a second person during the pilot; keep all artefacts and IP in ASBIS-owned repositories; make an explicit ownership-handover plan a condition of scaling. We do not pretend this risk is zero — we contain it.
07

Financial Impact Framework

This brief deliberately contains no invented revenue projections. At pilot stage the honest position is that we measure decision quality and opportunity size, and build a bankable business case on that evidence — not on a forecast.

Value leverHow it would be measured
Cross-sell & customer expansionIncremental gross profit from selling more to existing customers; euro-sized share-of-wallet gaps closed.
ReactivationGross profit recovered from lapsing or declining accounts brought back into regular purchasing.
Better portfolio decisionsBrand/category onboarding and exit hit-rate; reduction in dead or slow-moving SKUs.
Inventory risk avoidedOverstock and stockout exposure — before versus with system-informed decisions.
Faster validationTime and cost to test a new commercial initiative, before versus after.
The logic

Value = the improvement in each decision × the money at stake in that decision, summed across the decisions in scope. The pilot establishes those improvements on a small, measurable scope — so a full business case rests on evidence, not on projection.

08

Questions We Expect

Why now?

Because the cost and time to build and test this has collapsed, a working prototype already exists, and the underlying asset — our data — is already paid for. The advantage is moving while testing is cheap.

Why this team?

Because it has been built by someone who understands the distribution business and can also execute — a combination that is rare. The pilot deliberately adds a second person and full documentation to reduce dependence on any individual.

What happens if the initiative stops?

Nothing operational. Because it is additive, IT4Profit and Power BI are untouched; stopping costs only the capped pilot spend. There is no migration to unwind.

How does this scale?

By adding data sources and categories onto the same normalized model — each addition compounds value without rebuilding. Scaling is gated on pilot evidence and a continuity / ownership plan, not assumed.

How does it coexist with IT4Profit and Power BI?

It sits above them and consumes their outputs. It is the Opportunities layer — it replaces neither the operations system nor the reporting standard.

What evidence should precede a larger investment?

The day-60 read-out: a back-test with quantified deltas, a sized opportunity list, and at least one real decision changed. No larger investment is requested — or warranted — without it.

In one line

This is a narrow, low-risk request: approve the pilot, and judge it on evidence in 60 days.