WeDiscover - homeTalk to us
ServicesTechnology

Data science

Attribution, incrementality and forecasts you can plan around

Overview

Data science at WeDiscover

Our data scientists build the models that show what your marketing adds

How we do it

How we run data science

We measure what each channel adds with more than one method, model the value of each customer, then use the answers to plan and bid.

01

Measure what each channel adds

Attribution, incrementality tests and marketing mix models answer different questions, so we use them together and check each against the others.

Custom attribution
We design attribution around how your customers buy, including long and offline journeys. The attributed data then feeds your bidding.
Incrementality testing
Geo tests and holdouts measure what a channel or campaign adds. We write down the hypothesis, the control and the success metric before each test starts.
Marketing mix modelling
Bayesian marketing mix models, calibrated with your test results, show the return from each channel across your whole budget.
02

Model customer value

We predict what each customer and lead is worth, so bidding and budgets follow value.

Lifetime value
We predict each customer’s lifetime value from your first-party data, so you can spend more to win the customers worth the most.
Propensity and value scores
Models score each lead or conversion on how likely it is to become a sale, and what it is worth. Bidding then follows value from day one.
Conversion lag and cohorts
Cohort analysis shows how conversions build over time, so you can judge new activity early and accurately.
03

Plan and forecast

We forecast the return on each option and show where the next pound earns the most.

Forecasts before you commit
We forecast the return on each budget option, so you choose with confidence.
Marginal returns
We show where the next pound earns the most, by channel, product and region.
Scenario planning
We model different budgets, seasons and prices, so you can see the likely result before you set a plan.
Results

Results in data science

FAQs

Questions about data science

Have a question that is not here? Talk to us.
Related

Related services

Partner with us

Tell us what you want your marketing to do next.

Join our team

Join specialists in media, data, engineering and creative, and learn from all of them.