Courses - It's time to grow👇

Can we try Facebook Robyn?

You’re asked by your boss to run Robyn, Facebook’s new open-source MMM tool. Just one problem: you don’t know how to code! You decide to take on the challenge, because solving attribution would be worth it.

Completed in: 1 hour, 6 minutes

Can we try Google LightweightMMM?

After Apple killed user-level tracking in iOS14, Facebook went all in on Marketing Mix Modeling with Robyn, an open source library. Now Google is getting into the game with an *unofficial* library called LightweightMMM, and it couldn’t be more different.

Completed in: 28 minutes

Compare our attribution to MMM

After months drowning in data it’s such a relief to arrive at an accurate and plausible Marketing Mix Model, that sometimes I forget what I built it for in the first place: the Channel Multipliers!

Completed in: 27 minutes

iOS14 Broke Our Tracking!

You're an analyst at a promising FinTech app when the dreaded update lands – iOS14. How do you measure performance with broken tracking?

Completed in: 58 minutes

How to measure and track brand accurately without a vendor-filtered lens

While most brand marketing people talk a big game, Kantar data show 89% of companies don’t actually grow their brand each year. Which means there’s a lot of marketing teams out there who are likely being fed inaccurate data about their brand’s ‘growth’. Are you ready to learn the wrong from the right? The evidence-based from the anecdotal? The fluff from the fortune?

Completed in: 27 minutes

Can we try Uber Orbit?

I’ve built 20+ Marketing Mix Models, and it’s always the same thing that blows the model’s credibility with decision-makers. The assumption that marketing performance doesn’t change! Nobody believes all their hard work was for nothing. Uber’s ‘Orbit’ model solves that problem.

Completed in: 22 minutes

We need to set up GA4

By mid 2023 Google will sunset Universal Analytics (the Google Analytics we’re all used to). GA4 will become the replacement Analytics product and given it’s different on almost all fronts, it’s important you have it running months before the cut off.

Completed in: 17 minutes

What's the impact of TV ads?

As the Head of Growth of GoolyBib, you just bet big on TV as your next growth channel... but how do you measure the impact?

Completed in: 1 hour, 10 minutes

Finding Numbers for Greek Letters

Building a model is not just about finding the right parameters, but also the hyperparameters used to transform your data outside of the model. Finding the right numbers can be surprisingly tricky.

Completed in: 27 minutes

But isn't everything interacting?

You just explained to the client that your model doesn't capture interaction effects, and now they don't trust it. How do we deal with it?

Completed in: 18 minutes

How do I build a marketing mix model?

Imagine you’ve got an opportunity to prove yourself with your first marketing mix modeling project. It’s a big topic: what’s the minimum viable thing to do first?

Completed in: 37 minutes

Do Facebook Ads drive organic?

The Head of Performance wants to know: do Facebook ads drive organic app installs too? As a marketing analyst at an early-stage FinTech start-up, you’re the one tasked with finding that out. The ad budget hangs in the balance.

Completed in: 33 minutes

Why don't we try AI?

Your company, a FinTech startup, has started to expand internationally. How do you model performance in countries you know nothing about? Maybe AI can just ‘learn’ the right model for you…

Completed in: 56 minutes

How do I control for COVID?

External variables often affect your performance more than anything you did in marketing. Yet most marketing reports are missing the impact of macro factors like pandemics, the economy, and seasonality. Failing to control for significant variables means missing your targets in downturns and overcompensating your team in periods of high demand.

Completed in: 28 minutes

From what happened? to what if?

It’s amazing just how many companies forecast future performance by extrapolating in Excel. If you want to avoid unrealistic targets or to really know how much to spend to hit it, you need a model.

Completed in: 22 minutes

Is my data in the right format?

Before you can build a model, you need your data in the right format. Data cleaning is 60% of the job: time to learn how to do it!

Completed in: 1 hour, 11 minutes

How much more can we spend?

Your CEO asks you to double marketing spend, but you have no idea what that will do to CPA... how do you forecast realistic goals?

Completed in: 48 minutes

Can we trust this model?

You've built a model... but can you trust it? There are many statistical tests you can run to validate your model, but they've all got confusing-sounding names. What's most important when checking if your model is accurate?

Completed in: 1 hour, 48 minutes

We need a tracking plan

Everyone wants to be more data-driven, but nobody wants to implement tracking. So the job falls to you to get all the stakeholders aligned on what needs to be tracked, and how it all joins together.

Completed in: 17 minutes

How does our agency make money?

Most people don’t know how agencies make money, or where it all goes. How is it the client pays $128 an hour and the person actually doing the work gets $40?

Completed in: 27 minutes

Do we sell more when it's sunny?

Does the heat sell hot dogs? Your new client at Vexnomics needs to know if it's the sun driving sales, or a new marketing campaign.

Completed in: 1 hour, 2 minutes

Let's try 10,000 models

It's tedious to try different diminishing returns manually to find the best one. Facebook's Nevergrad package can try thousands of models for us!

Completed in: 39 minutes

Our data is all over the place

As much as 60% of the work of a data scientist is collecting and cleaning data, before any actual modeling work can begin. Getting organized can give you a huge edge in finding insights quicker.

Completed in: 27 minutes

Do ads behave differently during a sale?

Previously you did some analysis that told you how much to spend on ads. However a sale arrived and behavior changed – you underspent, leaving money on the table. How do you update your model to account for sales?

Completed in: 23 minutes

How much will CPM rise on Black Friday?

The CEO is worried about performance ahead of the Black Friday Sale – how much will the CPM rise on Facebook? Being able to predict the future would help them plan for increased costs... but that's impossible, right?

Completed in: 49 minutes

Is this a real problem, or just an anomaly?

Before anyone starts to panic, you need to triage the problem and estimate the size of the issue. We only have a week's worth of data, could this just be an anomaly?

Completed in: 17 minutes

What's the source of this lead?

If it isn’t in the CRM it didn’t happen. Sales guys just don’t want to believe marketing drives all their leads, unless they see it with their own eyes, and you’re not going to get them to log into GA. Capture the marketing source with hidden form fields and thank me later.

Completed in: 22 minutes

Compile a report by channel

Almost nobody is as good at Excel as they pretend to be in job interviews. One thing I see is people manually finding/replacing row by row to compile simple channel reports.

Completed in: 27 minutes

What's the lagged effect of our ads?

Now you're running a new branding campaign on Facebook, you have to wonder if there's a delayed effect of advertising, like you'd see with TV ads. How do you measure it?

Completed in: 1 hour, 1 minute

What's the long term effect on brand?

Traditional brand marketers can’t justify the impact of their campaigns, and yet performance marketers have a short term bias that can wreck your company. They both fight over budget, so how do we reconcile the two?

Completed in: 22 minutes

What are my key performance drivers?

How do you know what factors are most responsible for your performance? Selecting features for your models is both an art and a science, but it's tedious work until you learn to automate it!

Completed in: 1 hour, 19 minutes

Where should I allocate spend?

How do you allocate marketing budget across channels? You can't just guess: leadership won't be happy without seeing justification for the numbers you send them. You need a budget optimization script.

Completed in: 38 minutes

Can we prove the ROI of Ads?

It's getting harder to prove the ROI of your Facebook ads, as privacy rules hamper your measurement efforts. Geo Lift Tests could be the silver bullet you're looking for, that works on any channel.

Completed in: 42 minutes

Maybe it's just seasonality?

Sales are down at a trendy fast-casual hot dog restaurant, and the client needs to know: is it seasonality?

Completed in: 20 minutes

How can Facebook drive negative sales?

Your model doesn't make sense: Facebook ads can't drive negative sales!? If we can 'tell' the model that result isn't possible with Bayesian priors, our model will be much more believable.

Completed in: 36 minutes

Do I have enough data?

You've been tasked with modeling a new channel, but do you have enough data? It can be hard to get enough data, but you can improve your odds with a geo model.

Completed in: 35 minutes