Jenner Studio application · Version 0.4.0

Marketing Mix: marketing mix modelling on your own data

An application for Jenner Studio, published by Jenner Analytics.

Marketing Mix takes your own sales, media, price and weather files to a budget plan in seven steps: the weekly data, the baseline, the drivers, your geo experiments, three models, their comparison and the plan. Every step writes a SAS program from what is on the screen and shows it under "The program", and every run lands in your Jenner Studio project with its log.

Needs Jenner Studio, which is part of Jenner. Plans are on the pricing page.

Marketing Mix, step by step

  1. 1

    Data

    Sources · Checks · Model table

    In 0.4.0
  2. 2

    Estimate the baseline

    In 0.4.0
  3. 3

    Choose the drivers

    In 0.4.0
  4. 4

    Read the experiments

    In 0.4.0
  5. 5

    Fit the models

    In 0.4.0
  6. 6

    Compare the models

    In 0.4.0
  7. 7

    Plan the budget

    In 0.4.0

§1 How it works

How does Marketing Mix work?

Nothing happens out of sight. Each step is a program you can read, and each run is kept.

Every step is a SAS program

Each step writes a SAS program from what the screen holds and shows it under "The program", so you can read exactly what will run.

Every run is kept in your project

Each run lands in your Jenner Studio project together with its log, next to the rest of your work.

Explanations come from your results

The explanations are sentence templates filled in from the run's own results.

§2 Step 1, Data

Step 1, Data: from your files to the model table

The first step has three tabs: Sources, Checks and Model table.

Tab 1

Sources

Add .csv, .xlsx or .sas7bdat files, or tables in a library. Each source is added as facts, a lookup, a calendar or an experiment.

Each source is profiled: rows, column types, missing values, ranges and distinct values. For dates, the profile says how far apart they are: a day, a week, a month or a quarter.

ColumnRoleTypeMissing
week_starttimedate, a week apart0
store_idplacetext0
units_soldoutcomenumber14
tv_spendspend, channel TVnumber0
shelf_pricepricenumber3

Example profile of one source (invented data)

Column roles

Marketing Mix guesses the role of every column, and you can change any of them.

What, where and when
place, product, time
Sales and media
outcome, amount, spend, activity with a channel
Prices and other measures
price, share, flag, index
Weather
temperature, precipitation, level
Keys and locations
lookup key, ZIP code, location

Tab 2

Checks

Marketing Mix looks for the problems that spoil a model. Each finding is marked "must fix" or "worth a look".

  • Values below zero
  • Prices of zero
  • Shares outside 0 to 1, such as 45 for 45%
  • Flags that are not 0 or 1
  • Temperatures out of range
  • Missing-value codes such as 999 or -99
  • Extremes more than five times the 99th percentile
  • A change of scale
  • Duplicate keys
  • Places with gaps
  • Numbers read as text, such as "$1,204.50"
  • Missing keys

Tab 3

Model table

The last tab writes the weekly table a marketing mix model starts from.

  1. 01Choose the product level and the place level.
  2. 02Choose the day weeks start on and how many weeks to hold back.
  3. 03Daily data is added up into weeks.
  4. 04Monthly and quarterly data is spread over the days, then added up.
  5. 05Each shop gets the weather from its nearest weather station.

Try it on the example data

Marketing Mix comes with example data for a simulated manufacturer: sales, products, stores, media, competitors, search, weather, weather stations, a calendar and two geo experiments, with the column roles already set. It is simulated, so it belongs to no real company.

Start with your own data

Version 0.4.0 takes your files from the first checks to a budget plan, and keeps every result as a table in your project.

§3 Steps 2 to 7

What do the other six steps do?

All seven steps are in version 0.4.0. Each result is a permanent table, kept in your project unless you choose another library, for a workflow or a notebook to carry on from.

  1. 02

    In 0.4.0

    Estimate the baseline

    Two methods, either or both: the classic baseline engine and a state-space model (PROC UCM). Checks on the baseline, and the lift each kind of promotion adds (PROC MIXED).

  2. 03

    In 0.4.0

    Choose the drivers

    Give each column of the model table a use. Marketing Mix writes the drivers table, with carry-over (adstock), lags, the price index, season terms and a dummy per place, and flags channels that run together and high variance inflation.

  3. 04

    In 0.4.0

    Read the experiments

    Read your geo tests: difference in differences, a permutation test and a synthetic control. The estimator is chosen on fake tests before any real test is read, and each test gives a lift and a return per dollar withheld.

  4. 05

    In 0.4.0

    Fit the models

    Three models: MMM 1, regression with carry-over (PROC AUTOREG); MMM 2, Hill curves and adstock (PROC NLIN); MMM 4, Bayesian and calibrated by your experiments (PROC MCMC). A long run shows its progress, and can be stopped and resumed.

  5. 06

    In 0.4.0

    Compare the models

    Fit, return on ad spend (ROAS) and marginal ROAS by channel, response curves, and each experiment replayed through each model. The pick follows a rule shown on screen before any model runs.

  6. 07

    In 0.4.0

    Plan the budget

    Each model that converged plans the same total spend across the channels, within bounds you set. A channel the models disagree on is held where it was, and a cautious alternative sits beside the recommendation.

Still to come: three more models, MMM 3, the two-layer model and Meridian.

§4 What you need

What do you need to run Marketing Mix?

Marketing Mix needs Jenner Studio, which is part of Jenner. It installs into Jenner Studio and shows up there. Plans and prices are on the pricing page.

Application
Marketing Mix
Publisher
Jenner Analytics
Version
0.4.0
Runs in
Jenner Studio, part of Jenner
Programs
SAS language, shown for every step

§5 Related

Learn the method, or have us build the models

Free course: Marketing Mix Modeling with SAS Code

Our free video course on the Marketing Analytics page takes a simulated retailer from raw store sales to a budget decision.

Watch the free course →

Marketing Analytics service

If you would rather have us build marketing mix models on your data and deploy them in your own infrastructure, that is our Marketing Analytics service.

See the Marketing Analytics service →

Frequently Asked Questions

Marketing Mix is an application for Jenner Studio, published by Jenner Analytics, for marketing mix modelling (MMM) on your company's own data. Version 0.4.0 has all seven steps: data, baseline, drivers, experiments, models, comparison and the budget plan.

Step 1, Data, in three tabs. Sources adds and profiles your files and guesses each column's role. Checks finds problems such as negative values, shares written as percentages and numbers read as text, each marked "must fix" or "worth a look". Model table builds the weekly table, with the weather from each shop's nearest weather station.

Version 0.4.0 fits three: MMM 1, a regression with carry-over; MMM 2, with Hill curves and adstock; and MMM 4, a Bayesian model calibrated by your geo experiments. MMM 3, the two-layer model and Meridian are still to come.

Yes. Every step is a SAS program written from what is on the screen and shown under "The program". Each run lands in your Jenner Studio project with its log, and the explanations are sentence templates filled in from the run's results.

.csv, .xlsx and .sas7bdat files, and tables in a library. Each one is added as facts, a lookup, a calendar or an experiment.

Marketing Mix needs Jenner Studio, which is part of Jenner. It installs into Jenner Studio and shows up there. Plans and prices are on the pricing page.

Get Marketing Mix

Marketing Mix runs in Jenner Studio, part of Jenner. Plans and prices are on the pricing page.

Questions about Marketing Mix?

Tell us about your data and what you want to measure, and we will get back to you.