02 · Success Model

What drives your sales, and what it's worth.

Econometric and time-series models that measure what each channel, price, promotion and season adds to sales, share and brand, with margins of error you can defend to a finance director, and the KPI benchmarks each market can fairly be judged on.

● Success Model · where the next pound works hardest

Synthetic example

SPEND →SALESTV · todayOnline video · today
TVOnline videoSocialSearch
See the numbers behind this chart ▾
ChannelReturn on the next £1Spend todayStatus
Search£2.60£150kStill rising
Online video£2.10£300kStill rising
Social£1.20£255kNear flat
TV£0.80£795kBelow the £1 line

→ Decision

Move £195k a year from TV and social into online video and search.

▲ +£140k sales a year · likely £80k to £200k

What it answers

The questions a board asks, answered with numbers.

Each one gets an estimate, its likely range, and the data it rests on.

  1. 01

    What did media earn last year, in sales and in profit?

  2. 02

    Where does the next pound work hardest, by channel, brand and market?

  3. 03

    What happens to sales if we cut TV by 20% and move it to online video?

  4. 04

    How do price, promotions and distribution interact with media?

  5. 05

    Which brands and markets should get more next year, and which less?

  6. 06

    What is our working media really returning, after fees and production?

How we model. Six steps, no black boxes.

01

Data

Foundations first. We check and score the inputs: two years of weekly media, sales, price, promotion, distribution and seasonality, by market and brand. Below 75 on readiness, we fix the data before we model it.

02

Model

Market response models: econometric and time-series methods that estimate each driver's effect with carryover and saturation. Several model forms, tested against each other, not one black box.

03

Validate

Every model is tested on weeks it hasn't seen, back-tested against known events, and reported with its error and its ranges. If it can't predict, it isn't used.

04

Benchmark

KPI benchmarks that are fair: markets and brands grouped by how they behave, so each is judged against its peers and its own history, plus published benchmarks with their sources named.

05

Simulate

Budget, price and media scenarios across channels, brands and markets, so you see the return and its range before you commit the money.

06

Translate

Results in the terms finance uses: working media, brand equity investment, revenue, third-party revenue, margin and payback. The model earns its keep when the finance director uses it.

What you get.

Model report

What each driver added, with carryover, saturation and the error on every estimate, by market and brand.

Response curves

Where each channel flattens, and where the next pound still works, as charts and as tables.

Budget simulator

In Excel or Power BI: move money between channels, brands and markets and see the sales and range change.

KPI benchmarks

The KPIs that drive growth for you, with fair targets by market and brand, and the definitions behind them.

Finance translation

The same results in working media, revenue, margin and payback terms, ready for the budget round.

Readout and refresh plan

A walk-through with your team, in English or Spanish, and the plan to keep the model current through the Growth programme.

Scope and price

Priced in your proposal.

A Success Model is scoped by the markets, brands and drivers in it, typically eight to twelve weeks from clean data to readout. Fixed price, fixed end date. Where the data isn't ready, a Health Check or Diagnostic comes first, and its fee is credited.

What we need from you

  1. 01

    Two years of weekly data

    Media spend and delivery, sales, price, promotion and distribution, by market and brand.

  2. 02

    The decision you need to make

    Next year's budget, a channel cut, a launch: the model is built to answer it.

  3. 03

    A finance contact

    So the results land in the terms your budget round uses.

Tell us the decision you need to make.

We'll tell you which data, KPIs and methods it needs, how long it would take, and what it would cost.