I often get asked for a POV on this, so I guess a lot of people are asking the same question.

Recap: What are Priors?

In a Bayesian media mix model, priors are simply what you believe the results might be before you look at the data or do any modelling. Think of them as expressing real-world knowledge — like “it’s very unlikely that my sales go down when I spend more on advertising” — in mathematical language. [5].

My view, in summary: Priors introduce risks of bias which can undermine model integrity — particularly when the priors are derived from last‑touch, platform‑reported metrics or poorly specified experiments. The core issues are that last touch platform reporting is biased to last touch and therefore inaccurate, and experiments are rarely undertaken with the levels of statistical rigour required to make them reliable.

Potential Advantages of Using Priors

Let’s look at the potential advantages of using priors.

1. The biggest argument for Bayesian Priors is that they reduce uncertainty in an MMM environment. Bayesian MMM doesn’t just give you a single point estimate for the impact of each marketing channel. It provides a probability distribution of possible impacts… This allows you to understand the uncertainty associated with the estimates, leading to more informed decision-making [6].
2. Model outputs (e.g. CPA by channel) look “right” and are relatable in the real world: To a large extent, priors fix the model outcomes into sets of values that seem to match the marketing team’s expectations and experience.
3. Priors can help when the data available for modelling is limited: Sometimes there isn’t sufficient spend data to accurate estimate coefficients for media channel performance. Spend may be too low, or the spend in one channel is drowned out by higher spends in other channels at the same time, or the number of weeks with spend may be very low. In these situations, priors can help by preventing coefficients from collapsing toward zero because the data is weak.
4. Priors help when spend variables are highly correlated: Many media campaigns feature multiple channels running at the same time campaigns often take place over 2–8-week periods with media channels being combined to extend reach or delivery efficiency frequency. This means input variables are highly correlated with each other which is a problem for regression models – they can’t isolate the effects of correlated channels. Using priors in a model helps the model find clarity by providing guides to what channels effects should be.

Key Risks and Limitations

1. Priors are beliefs, not data, so they risk injecting bias into your model: A key issue is that Priors are beliefs about media performance. Because they are beliefs about media performance e.g. “our Last Touch Social CPA is always between £10 and £20, so let’s fix that range into our MMM results” – we risk using those findings to inform the model outcomes – clearly this is a form of confirmation bias [7].
2. Priors often rely on inaccurate last‑touch or platform‑reported ROI Data: In practice, most priors come from platform dashboards rather than controlled experiments. The problem with platforms is that they generally provide last touch attribution (sometimes, first etc, but never full funnel i.e. including trend and seasonality an marketing mix variables for example).  This means that from the “get-go” Last Touch reporting is inaccurate – our industry knows that- so why use these results as priors in MMM?  But obviously using Last Touch platform sourced priors is only going to transfer that uncertainty directly into your marketing mix model. This is ironic because most MMMs are commission to provide an alternative view to Last Touch reporting.
3. Priors can pull MMM results toward last‑touch, and away from more statistically reliable findings: One of the big arguments for using priors is that they keep MMM results grounded in reality. But if you are using priors, you are creating an artificial result. If the prior is biased (e.g. towards last touch reporting), the model output will also be biased. Whilst this can make the MMM appear more aligned with platform reporting, but that alignment is artificial and the results are erroneous.
4. Priors can destabilise other coefficients: Let’s go back to high school maths. We know that equations have to balance on both sides – output e.g. Sales on the left, inputs e.g. media spend on the right. If we fix one or more of the input components of our equation on the right, other parts of it will have to move in order to accommodate that prior fix. This means your model will almost certainly inflate some media channels, suppress other, channels especially upper‑funnel media and produce inaccurate results
5. Priors reduce transparency: Stakeholders often struggle to understand how much of a coefficient is “data‑driven” versus “prior‑driven,” which can undermine trust.
6. Priors can mask genuine performance shifts: If a channel’s true effectiveness changes (e.g., due to creative fatigue, privacy changes, or market dynamics), a strong prior can prevent the model from detecting it. Equally performance might improve – dramatically. Let’s say you launch a new social media campaign with a new offer and new creative. Your MMM Social CPA falls from between £10 and £20 to £5. That means your social campaign has become much more effective. But MMM priors would likely exclude that result – or at least suggest it is unrealistic.
7. Priors Do Not Include Adstock or Diminishing Returns: There is a common belief that priors somehow incorporate adstock or saturation assumptions. They do not. Adstock and diminishing returns are structural modelling choices — they define how media works overtime and at different spend levels. They are often subjectively judged but they can and should be extracted from the MMM dataset itself using a grid search loss minimisation technique [8].   This can’t be circumnavigated by experiments as Adstock and diminishing returns are critical parts of advertising evaluation [9]. They can only be determined from  long time-series datasets, not short‑term experiments.
8. Priors can come from experiments, but these are notoriously difficult to get right: Priors can be defensible when grounded in rigorous, repeatable experiments. However, this is a technically demanding area. Getting marketing experiements right requires:
   1. Clean treatment and control regions
   2. No spillover
   3. Stable delivery
   4. Repeatability
   5. Sufficient spend, time and statistical power

In practice, achieving this level of scientific discipline in marketing experiments is difficult. Geographical regions can be more fluid than they look on a map – we live in a mobile world when people can move from one region to another in very short periods of time. [10].

Real‑World Example: Identical Creative, Different Results

There have been cases where a team ran an A/B test using the same creative in both treatment and control. Despite being identical, the experiment produced different lift results for each “creative.”  This highlights how algorithmic targeting, user heterogeneity, and data aggregation conspire to confound the magnitude, and even the sign, of ad A/B test result [11]. If identical creatives can produce different results, it shows how fragile and noisy marketing experiments can be — and why they often lack the stability required to serve as robust priors.

Conclusion:

I conclude with these points:

1. Avoid using Bayesian priors if you want a high integrity MMM.
2. Not using priors may produce more challenging results – but isn’t that what you want? You are building your MMM to get a different perspective on your marketing performance; to identify hidden opporutunities and to improve marketing ROIs. Why gloss over that valuable insight?
3. This may result in a more challenging conversation with the C-suite – but it’s safter to report a challenging result from a high integrity model than any result from a flawed model.
4. If you use priors to shape the answer, then the answer will look better. That doesn’t mean it’s the right answer, in as much as any model can produce the right answer.
5. Priors introduce confirmation bias into your model. If you believe the enemy of good modelling is bias, you must question the use of priors in your MMM project. They may be acceptable to some, but not to others.
6. If you are going to use priors, you must be very careful about where you source them. There are three main sources – Experience, Platforms and Experiments, but all have flaws and all run the risk of introducing bias into your model.

Implications for Marketers – checklist

• If you are going to use priors you must be certain that they are accurate.
• Be aware that in most marketing contexts the risks of priors being inaccurate are high.
• Don’t rely on last touch data for priors.
• If you use experiments, ensure they are properly specified, and even then, use them with care.
• In cases where you have too little data for full MMM, be careful about using Bayesian priors to overcome this problem
• In data light situations, consider reviewing your data, accepting lower granularity or looking at a different attribution modelling technique like regularisation.

References

1. J Martin and P Perez, Frequentists vs Bayesians and Marketing Science, Quantified Nation, July 2024 – https://open.substack.com/pub/quantifiednation/p/qn9-frequentists-vs-bayesians-and
2. Hits and Misses of Meridian – A Thorough Deep Dive, Aryma Labs Feb 2025 https://arymalabs.substack.com/p/hits-and-misses-of-meridian-a-thorough
3. Duncan Stoddard, Is Bayesian MMM worth the faff? DS Analytics Blog, February 2024, https://dsanalytics.co.uk/thoughts/is-bayesian-mmm-worth-the-faff
4. Two key problems that ail Bayesian MMM – Aryma Labs April 2024 https://arymalabs.substack.com/p/two-key-problems-that-ails-bayesian
5. Marty Sanchez, What Are Priors in MMM – And Why They’re Difficult to Get Right (But You Need To) Get Recast Blog June 2025 – https://getrecast.com/what-are-priors-in-mmm-and-why-theyre-difficult-to-get-right-but-you-need-to/
6. Rohit Nair, What is Bayesian MMM & why use it? Medium April 2025 – https://medium.com/@rohitnair.inft/what-is-bayesian-mmm-why-use-it-942d0193e7eb
7. Cognitive bias and data: how human psychology impacts data interpretation – Penn LLPS Features October 2025 https://lpsonline.sas.upenn.edu/features/cognitive-bias-and-data-how-human-psychology-impacts-data-interpretation
8. Nephade, D., A Predictive Modeling Approach to Multi Objective Marketing Mix Optimization: Balancing Performance, Acquisition, and Efficiency – in International Journal on Science and Technology (IJSAT) Volume 16, Issue 1, January-March 2025.
9. Gijsenberg et al., Understanding the Role of Adstock in Advertising Decisions, SSRN, 2011.
10. Tyler Buffington, Eppo, The Bet test – Spotting Problems in Bayesian A/B Test Analysis Dec 2024, https://www.geteppo.com/blog/the-bet-test-problems-in-bayesian-ab-test-analysis
11. Braun and Schwartz, Where A/B Testing Goes Wrong: How Divergent Delivery Affects What Online Experiments Cannot (and Can) Tell You About How Customers Respond to Advertising, Journal of Marketing, American Marketing Association, August 2024. https://journals.sagepub.com/doi/abs/10.1177/00222429241275886

The post Using Bayesian Priors in MMM: Do Better‑Looking Results Mean You Have a Better Model? first appeared on Marketing IQ.

A marketing Mix Model (MMM) can produce challenging results, especially if you have been relying on last touch( LT) attribution.  When presented with challenging results your first question may well be “can I trust the model?”.

The answer to this depends largely on the quality of the model as expressed through a set of diagnostics. The problem is not all diagnostics are shared.

Let’s explore MMM diagnostics in some more detail so you are better placed to form your own conclusions on model quality.

What you should get from your MMM partner

Let’s start with the diagnostics you should see from your MMM partner. They should present you with these test diagnostics:

1. Exploratory Data Analysis (EDA)

• Correlation Analysis (before modelling): Identifies highly correlated variables early, helping avoid multicollinearity in the model build and guiding variable selection – this is important because correlated variables can destabilise any modelling and results.
• Trend & Seasonality Checks: These checks confirm long‑term trends and seasonal peaks are properly captured before modelling media effects – this is important because if you don’t isolate these components, you risk incorrectly misattributing their effects to media.
• ACF (Autocorrelation Function): Ensures time‑based patterns are understood and removed so the model isn’t biased by serial correlation – this is also important. Your sales or revenue data can include patterns or “tidal effects” where your sales in the current weeks are partly related to sales in the previous week. Again, if you don’t control these relationships, your model will produce erroneous results.

2. Model Fit & Accuracy

• R²: This is a diagnostic that most marketers will be familiar with – it shows how much of the sales movement the model can account for by the model – but beware this is by no means a full diagnostic of the model.
• Adjusted R²: A fairer version of R² that penalises unnecessary variables, but again, not a full diagnostic.
• Actual vs Fitted plot which shows how well the model’s estimates match the actual data.
• MAE (Mean Absolute Error): The average size of the model’s prediction error in real sales units. For example, if a typical week delivers 1,000 sales and the model predicts values between 900 and 1,100, the absolute error is 100 units.
• MAPE (Mean Absolute Percentage Error): The model’s average prediction error expressed as a percentage of actual sales. Using the example above, an absolute error of 100 on 1,000 sales corresponds to a MAPE of 10%.

3. Residual Diagnostics

• Durbin–Watson test: Checks whether any time‑based pattern is left in the errors; values near 2 indicate no remaining autocorrelation
• Breusch–Pagan test: This test checks whether error variance is stable; instability can affect model coefficient reliability

4. Model Stability & Interpretability

• VIF (Variance Inflation Factor): Helps you measure the correlation between channels in a model and helps ensure channels aren’t too correlated to separate cleanly. If your model contains variables that are highly correlated with each other, it is not likely to be reliable.

These tests are all mission critical for evaluating both the data going into the model and the  model itself. But while they diagnose the quality of the model, they do not evaluate its predictive quality.  Predictive quality tests are further tests that assess the model’s ability to predict outcomes on previously unseen test data. In some ways these validation tests are the  ‘acid test’ in model diagnostics. When you combine the standard diagnostics (1-4 above) with the predictive quality tests you will see:

1. Good results in both mean you are dealing with a reliable model.
2. Good results in one but not the other should be cause for concern.
3. Poor results in both diagnostics and validation tests mean the model should be ignored and re-specified.

Unfortunately, the predictive quality tests are not always shared by MMM providers – I’ll leave you to figure out why.

So, what are the predictive quality tests?

MMM validation route 1: The holdout test

• How does the test work? In this test, we test the model on a subset of the data, usually towards the end of the sample. Let’s assume you have a model built on three years’ worth of weekly data – that’s 156 observations. You can ‘slice off’ the last 26 weeks and use the first 130 observations to “train” the model. You can then apply the trained model to the last 26 weeks, also known as the “test” period.
• What should you see?  You will have the actual sales data for the 26 week test period. The key ‘acid test’ question for you is how well the model’s prediction for those 26 test  weeks match actual sales in the test period. If you model is predicting within +/- 20% that’s a strong result. If it’s predicting between +/- 10% that is a very strong result.

MMM validation route 2: Cross Fold validation

• How does this test work? In this test we slice your 156 weeks of data into 10 consecutive time “blocks” of 15-16 weeks. In a similar way to the holdout test, we provide the explanatory variables for each block  and ask the model to predict what it estimates sales to be in each of these blocks based on the model coefficients. We then examine the prediction vs the actual and the size of the error in each block.
• What should you see?  We will have the actual sales data for each of the blocks. Again, the key ‘acid test’ question for you is how well the model’s prediction for those blocks matches actual sales in each of those blocks. If your model is predicting within +/- 20% that’s a strong result. If it’s predicting between +/- 10% that is a very strong result.  It’s worth noting that in some categories the 20% might be too tight, you could relax it to say 20-30%, but you need to be close to that range to have a validated model.

What do these results mean? If you see a model prediction within 20% of the actual in this holdout period, then you have a model that is working well. It is unlikely to be ‘overfitted’. Overfitting is a way of making the model look good within the sample, but it  does not perform well outside the sample and would therefore fail these tests.

The post How reliable is your marketing mix model (MMM)? first appeared on Marketing IQ.

Establishing incrementality is critical if you want genuine brand growth. Why? Because many performance platforms collect, report and even double-count sales from multiple sources, including those which might happen even if you didn’t run any activity. This means you are attributing to media spend sales that would have happened without media spend. This type of misattribution will mean you are using flawed data for budget optimisation and this in turn will lead to sub-optimal media performance. Misattribution makes your budget less efficient and less effective.

In order to detect incrementality we need to establish what would happen if your product or service didn’t have any marketing activity. There are three things – sometimes called “components” to look at here:

1. Trend – what is the underlying trend in your category an din your sales – are sales they in growth, decline or stable?
2. Seasonal cycles – What are the repeating patterns in the data – do sales increase or decrease in certain months, certain weeks on a regular predictable pattern?
3. Base brand equity – how many sales would you expect to see if you paused your marketing activity

These components can often account for more than 75% of your sales revenue. If your performance platforms are reporting 100 sales, it could be the case that 75 of these sales would have happened without any marketing activity. For many advertisers this is an “OMG” moment.

Imagine if you could identify the sales that would have happened without marketing or media support and then focus your marketing budget on activities that deliver genuine incremental growth rather than paying a platform “tax” for sales that were going to progress through your sales pipeline without any short-term marketing spend.

Let’s take a closer look at trend and seasonality and why it’s important. We’re going to use the “Bike Sales” dataset from Kaggle.

First let’s look at the sales data itself:

Here we can see bike sales from July 2017 to July 2022 over a total of 260 weeks.  We can make some initial observations. There is an underlying growth trend. We can also see that there are a number of peaks and troughs in the data. We see that the highest sales weeks are around 110k and the lowest sales weeks are around -30k so the weekly sales have a range of c. 140k.

Now let’s extract the trend component from the dataset:

We can see the underlying trend in the data, quantified using a moving average. We can see there is a strong upward trend from 50k sales to almost 85k sales.

Next, let’s extract the seasonality component from the data set:

It’s important to note here that “seasonality” doesn’t mean “seasons” as in Spring, Summer, Autumn and Winter. Here seasonality refers to any repeating cycles in the data. We can see there is  clear pattern of repeating cycles. These repeating cycles range from +10k to -20k.

And finally we are left with the Random component:

The Random component represents sales that are not explained by trend and seasonality. You can see that these random sales i.e. not explained by trend or seasonality, range from about +35k to -30k.

This random data is the data we test for contributions from media spend.  More on that model and its outputs in the next post.

The post What is incrementality in marketing – extracting trend, seasonality and brand equity first appeared on Marketing IQ.

Before we look at the 4Ps in detail let’s summarise the difference between strategy and tactics:

Strategy: Sets out which direction you have selected to achieve the macro marketing objectives your organisation has set. In marketing terms this might be to increase share by depositioning weaker competitors or to increase sales by increasing market penetration into new audiences. Think of strategy as the journey you need to make to get to your destination relative to everything else that is going on in the economy and in your category. Strategy is about the management of your resources in your business environment. Strategy is delivered over the medium to long term – it usually takes time to deliver, months and sometimes years. Strategy is what you’re going to do to achieve your objectives.

Tactics: Sets out the individual actions you will undertake in order to deliver the strategy. In marketing this might mean increasing revenue by increasing prices and using advertising to drive preference and reduce sensitivity to price.  Think of tactics as the individual decisions you have to take to complete your journey. Tactics can happen quickly – days, hours or even minutes.

Against this background the 4Ps are not exclusive to strategy or tactics, they can contribute to both. Let’s examine how each of the 4Ps works in a bit more detail.

P1 – Product

• What do we mean? Products have attributes which can confer advantage, or mean that the product lags behind market trends. If the product is ahead of demand trends, it should perform well in market. If it’s behind, it will do less well. The development of attributes is referred to as NPD – New Product Development – the more NPD generally means better, more competitive product and vice versa.
• Examples –
  • Apple used product technology to revolutionise the mobile phone market. Apple’s iPhone set totally new standards in mobile technology by combining a phone, a music player and a web browser, not to mention developing its associated app marketplace. Today Apple still retains around 24% of the mobile market.
  • Toyota led the way in hybrid auto technology development and retains the dominant share in this category.
• Timescale – All products (and most services) have to be researched, designed and tested before they can be launched. Product development generally takes time, it can be months and, in some cases, it can be years.
• Strategic or tactical? Such are the costs, resources and timescales required product development it has to be regarded as a strategic issue.

P2 – Price

• What do we mean? Price is what we pay for goods and services. There is no question that price can change consumer behaviour. As a general rule the lower the price of a good, the more units it will sell and vice versa. However, a high price can also be used to assert and reinforce superiority in a category. Discounts and sales promotions fall under the Price element of the 4 Ps.  These can be used tactically to change price for short time periods and increase sales for price sensitive goods and services.
• Examples –
  • Aldi commits to no frills value and prices itself as being seen as the lowest price supermarket.
  • John Lewis used to guarantee that they were “Never knowingly undersold”. Recently, this mantra was dropped. Since then, the company’s fortunes have changed suggesting this price promise had a positive impact on consumer behaviour.
  • Stella Artois is positioned as reassuringly expensive.
• Timescale – Price changes and promotions can be activated quickly, by day in retail and in real time in online / e-commerce environments. however, there can be longer term commitments to price vs category average. The Stella example shows a long-term commitment to upholding a price premium to position a brand. We could say that supporting a premium price is a longer-term initiative, reducing price is a short-term initiative.
• Strategic or tactical? Price reduction and discounting can be tactical in the short -erm but maintaining a long-term low or premium price relative to a category average usually requires a longer-term strategic commitment. In the case of Aldi, the whole business – from supply chain to checkout is structured around delivering a low-price, this is a long-term strategic initiative to secure a market specific position.

P3 – Place

• What does this mean? Place means Distribution. It’s where and how consumers are able to buy your product. For many years, distribution was simply about retail, but since commerce has migrated to online, distribution has now had an online manifestation. This could be the more generic impact of e-commerce such as wider access to product through much reduced impact of distance, but it’s also about how consumers assess distribution quality. Quality can be measured through speed of delivery, ability to try and buy and the returns policy.
• Examples –
  • Traditionally, retailers would sell more products if they increase their number of stores and vice versa.
  • Banks continued to close branches as more and more of their customers transition their banking activities from the counter to online.
  • Amazon revolutionised distribution by creating a massive and accessible e-commerce platform.
  • Apple revolutionised how music is distributed and bought.
  • ASOS revolutionised the distribution of multi brand clothing and fashion items.
  • Netflix has revolutionised how we consume movies – and had effectively killed off other physical formats such as DVD.
  • In the e-commerce world, delivery times, costs and returns policy all form part of the distribution characteristics of a company or brand.
• Strategic or Tactical? Traditional retail distribution networks are a strategic asset but they can be leveraged in a tactical way. They are strategic because they involve the use of a lot of capital and are slow moving. They can be leveraged tactically through localised incentives. Digital channels e.g. ecommerce are distribution channels but they are much more flexible and can therefore be used both strategically and tactically.
• Timescale – changes in traditional retail distribution are generally slow moving although the opening and closing of retail stores can have a significant impact on short term revenue. Changes in e-commerce distribution policy can have a quick effect. Increasing delivery costs or free delivery thresholds can have an immediate effect on consumer behaviour.

P4 – Promotion

• What do we mean? Promotion means marketing and advertising communications. In the marketing mix, promotion does not mean price promotion. Price promotion sits under the price element of the marketing mix. Long term commitment to advertising spend can confer competitive advantage and a long-term commitment to investing on a share of category spend (Share of Voice or SOV) that is greater than your market share (Share of Market or SOM) has been shown to drive growth.  This is called excess share of voice or eSOV. See a post on this topic here. Commitment to advertising consistently and at scale is a core component of consumer goods marketing where prices are generally low, decision making is as much emotional as it is rational and consumer purchase decisions are made quickly on System 1 ‘autopilot’ decision making. To enable this high mental availability is required, and that in turn requires always on advertising which is efficient at reaching mass or large segment markets.
• Examples –
  • Examples of large scale “always on” advertisers include Unilever, P&G, Sky, McDonalds and Tesco – these brands represent over £500m in adspend – seems a lot, but for these mass market brands, they are investing less than £10 per person per year to maintain high mental availability and high brand preference.
  • Of course, these brands are not representative and there is a long tail of advertisers who use much lower spends to deliver targeted communications to build online traffic, clicks, leads and sales.
• Strategic or Tactical? Clearly promotional communication activity can be both strategic and tactical. We talk about ‘brand building’ and we talk about ‘performance’ media. There is little doubt that strategic activity is about

The post What are the 4Ps of the Marketing Mix first appeared on Marketing IQ.

Forbes recently analysed the carbon footprint of agencies in the US and found that the key drivers are travel, commuting, office supplies, marketing collateral, utilities, shipping and computer services [1].

Within this, we can take a closer look at the marketing and media campaign activities that generate CO2. For example, did you know that:

• One piece of direct mail costs 205g of CO2?[2].
• The internet including mobile phone use had a bigger carbon footprint that the airline industry [3].
  
• A typical online ad campaign with a spend of £100k could generate as much as 5.4 tonnes of CO2 [4].
  
• Digital advertising accounts for 7.2 million metric tonnes of CO₂ annually [5].
  
• A single ad campaign generates 70 tons of CO2 equivalent emissions: the same as what 7 people on average release into the atmosphere in a year [6].
  

WPP agency Essence has created a carbon calculator to look at emissions by media channel. Essence rank display and online video as relatively high emission channels, followed by magazines and social media. The lower emission channels include digital out of home, BVOD, Linear TV and static out of home. According to Essence traditional OOH is a sustainable channel with very low waste across both paper and vinyl channels. Cinema, digital audio and PPC are grouped low and newspapers and radio are the lowest ranking channels.

This may seem like uncomfortable reading, but these are issues we need to confront if we are serious about reducing advertising’s contribution to CO2 increases. If we know where we are generating CO2, we can either manage our emissions or consider appropriate offsets.

Some of the world’s largest media brands have committed to 100% renewable energy including Adobe, Apple, Facebook, Google, HP and Salesforce – thereby reducing the impact of some of the digital media channels and their delivery listed above. Microsoft has committed to being carbon negative by 2030 and has committed to offsetting all its historical emissions by 2050.

And what can you do as an individual marketing or media specialist?

Here are some surprisingly easy to implement recommendations to consider:

• Consider turning off auto play when you watch online videos
• When you leave your PC or lap top, turn it off rather than leaving it on standby
• Dim your monitor
• Limit “reply all” emails
• Consider doing more work on a tablet or smart phone if possible as these consume less energy than PCs and laptops.

You might think that as an individual your actions won’t make much difference to overall CO2 emissions, but you can make a difference. By acting in a sustainable way, your actions will influence others, and collectively more and more positive change will result.

Optimise your media plans to minimise CO2 emissions

It is possible to include CO2 emissions in MMM so that you can optimise both ROI and ROR (Return on resources i.e. minimise CO2 emissions). You can find out more here:

References

1. Forbes
2. Prime Data
3. MediaTel
4. IAB
5. Scope3
6. Fifty-five

The post Exploring the Carbon Cost of Advertising first appeared on Marketing IQ.

Effectiveness and efficiency are not the same but they are both critical in media strategy, planning and activation

Intro

Marketing effectiveness and campaign efficiency are intertwined terms across the open plan workspaces of both advertisers and agencies. But they mean very different things. Now is a good time to remind ourselves what these terms mean and to explore the differences between effectiveness and efficiency and how they apply in media investment.

Effectiveness is about doing the right thing. It is sometimes referred to as ‘goal orientation’- are we doing the right things to reach our goal?  At m/SIX we refer to these effectiveness options as “levers”.

Let’s look at some examples of effectiveness; if we want to build sales revenue by growing the market share of a brand we need to increase its market penetration. To increase penetration, we need to move our brand into the consideration and preference sets of more consumers and in order to achieve this goal we need to deliver increased reach. In this case the goal of increasing reach is our route to effectiveness.  Actual effectiveness is the degree to which our approach delivers proximity to the selected goals – increased market penetration through increased reach.

In another effectiveness example we may wish to increase revenues by repositioning our brand versus competitors. For example we may wish to position our brand as more environmentally friendly than other brands in the category. To do this we may need to change the way consumers view our brand and ask them to associate new meanings with it.  In order to do this we may need to change the memory structures associated with our brand which in turn may require the use of media channels capable of delivering that “change in memory structure” goal.

In a third example we might want to deliver revenue growth by increasing purchase frequency. To do this we might need to give consumers reasons to purchase more often by reframing the way they use the product. This would typically increase the number of usage  occasions that the product can contribute to. The decision to reframe the way the product is used, and our success in doing that is the measure of the campaign’s effectiveness.

Efficiency is about doing things right. Efficiency tends to be process or ways of working orientated. At m/SIX we refer to these efficiency options as “switches”.

Now let’s look at how the three examples above might benefit from increased efficiency.

In the case of increasing market penetration,  we would need to examine which channels are able to deliver reach most efficiently – typically, we might ask which channels can do this quickly, or which channels can do this in the most cost-efficient way – how much reach and attitudinal shift can be generated per pound or dollar invested. Another aspect of efficiency might be which creative assets we use, exactly when we use them, where we use them and the time and cost involved in producing them.

In the case of repositioning a brand, efficiency might be measured as the number of points of attitudinal shift per £pound or dollar invested. We know that some channels are more efficient at achieving this goal than others. We also know that certain ways of using those channels are more efficient – a moving image may be more efficient than a static image, a larger format ad may be more efficient than a smaller format ad. Higher frequency over a short time period may be more efficient than lower frequency – or vice versa.

In the case of increased purchase frequency, the most efficient route might be how an agency and marketing team can remind consumers with prompts or triggers to change their behaviour – this is usually signals-based targeting; it could also be a carefully planned search campaign to target recipe searches for example. Or it may be a signals-based media and creative optimisation to target active meal planners; if we know that a consumer is going to shop online, we need to deliver our prompts and triggers in the right way and at exactly the right planning moments.

Efficiency and effectiveness is not a binary choice between one approach or the other – we have to deliver both, but in the right measures

Now we have explored these two concepts, we need to emphasise that one without the other amounts to suboptimal marketing and media investment.  Making effective strategic decisions without efficient delivery is likely to be slower and more expensive than it needs to be. Delivering campaigns efficiently, does not necessarily deliver the best goal delivery – ie effectiveness outcomes.

At m/SIX we manage both effectiveness and efficiency;  the levers and the switches. We have teams of strategic planners who are able to focus on making the right goal choices to maximise marketing and media effectiveness. We have teams of audience planners who look for the audiences most likely to deliver our goal and the channels and targeting criteria that will deliver those audiences in the most efficient way. And we have teams of display, search, social and CRO ad CX specialists who help us ensure that the strategy is delivered efficiently.

But whilst the choice is not binary, the balance between maximizing effectiveness and efficiency has to be carefully considered – our strategists and analysts work on optimsing this balance so you can be assured that your budgets are being invested in ways that will maximize your overall business outcomes.

WE OFFER MEDIA ATTRIBUTION MODELLING TO OUR CLIENTS: FIND OUT MORE HERE

The post Maximising Media Effectiveness and Efficiency first appeared on Marketing IQ.

With MMM models built, we are able to run scenarios and forecast outcomes from different types of media activity. This enables advertisers to optimise their media investments to maximise sales returns from media or marketing budgets invested.

Why is it called Marketing Mix Modelling?

MMM derives its name from the traditional ‘marketing mix’. The name recognises that all the variables in the marketing mix have a role in generating sales. So, let’s remind ourselves of the traditional marketing mix – the so called 4 Ps – Product, Price, Place and Promotion.

Product: In MMM, we might attach some attributes to the product – fast, smooth, light, powerful, low CO2 etc. These might be significant drivers within the category.

Price: Price is always a key determinant in consumer behaviour. As a general rule, products priced competitively sell more, and products priced less competitively relative to a category or competitors sell less. If we are to avoid confusing price effects with media effects, we naturally have to isolate the effect of price and take it out of the effectiveness equation.

Place: i.e. Distribution variables, whether they’re online (like delivery times) or offline (like store opening times) distribution variables usually have an effect on sales. If more stores are open, more product tends to be sold. If delivery times are long, versus competitors, less product tends to be sold. If we are to avoid confusing distribution effects with media effects, we also have to isolate the effect of distribution and take it out of the effectiveness equation.

Promotion: This variable is the promotional or advertising media spend variable. It can include advertising in media channels like TV, online display, search, social, OOH, cinema and print media.

What does an MMM actually look like?

This is a question many clients want to ask. MMM is often presented as a complex black box, when in fact it is simply an equation that captures the impact of the different elements of the marketing mix outlined above.  A Marketing Mix Model equation looks like this:

Sales = base (the levels of sales with no marketing) + (product attribute * a coefficient) + (price * a coefficient) + (distribution* a coefficient) + (promotion * a coefficient) + an error term.

Think of the coefficient is a statistically determined response rate which captures the rate at which sales are generated from changes in investments or activities in each of the Ps.

In statistics, the above might be written like this:

y = B0 + B1xX1 + B2xX2 + B3xX3 + error

How is MMM used to optimise paid media investments?

When we know the rate at which each “P” in our model generates sales, we can set values for each of the Ps excluding media advertising and then adjust the media advertising investment levels to see how those changes impact sales. This is a powerful tool in the marketer’s portfolio because it permits strong cases for media investment to be made. And that’s essential if to you want to maintain or secure increased budget from your CFO.

WE OFFER MARKETING AND MEDIA MIX MODELLING (MMM) TO OUR CLIENTS: FIND OUT MORE HERE

The post What is Marketing Mix Modelling (MMM)? first appeared on Marketing IQ.

These non-linear effects are normally grouped into two areas:

1. the delayed effect of advertising and
2.  diminishing returns in advertising.

Let’s look at each one in turn:

1 – Adstock or Carryover

The first type of non-linear effect we see in media investment is the carryover effect. When we advertise, we know that the effects are not always seen immediately. This is because advertising, well good advertising, gets remembered and the memory effect on consumer behaviour may be felt some time after the ad is seen.

So, for example if an advertiser buys 100 GRPs in a week, the full effects of that investment are not confined to that week. What happens in practice is that the effect of that advertising tends to “carryover” into the next week and the week after that. How do we know this? We know because when we build models (e.g. MMM) to quantify response to media investment, they tend to be much more accurate when we carry-over the effect of advertising into the following weeks.  We call this carryover effect Adstock.

You might ask “how much do we carry over?” The answer to this is found by testing different Adstock carry-over levels and analysing how they correlate with sales response over time.  The most commonly used analogy here is borrowed from nuclear physics (don’t worry it’s not as complicated as it sounds). In nuclear physics radioactive substances have a half-life, that’s the time it takes for their radioactivity to decay by exactly half.  Marketers borrow this thinking and use half-life decay rates to model lagged advertising effects. We refer to the length of time required for advertising Adstock to fall by half as the ‘half-life’.

So, as an example, if in week 1, 100 GRPs create 1000 sales, a one week half-life might see that effect carry-over to 500 sales in the second week, 250 sales in the third week and 125 sales in the fourth week.    Any model that counts only the 1000 sales in the first week underestimates the lagged ROI of those first 100 GRPs. That’s because over the four weeks those 100 GRPs delivered 1,875 sales (1000+500+250+125) rather than the 1,000 sales originally reported. We can see that by considering Adstock, the ROI of the first week’s 100 GRPs almost doubles.

2  – Diminishing Returns

The second type of non-linear effect we see in media investment is diminishing returns. The law of diminishing returns states that as more of something is bought, the less utility is gained from it.   A frequently quoted example is agriculture – as more resources are invested into an acre of land, the yield of corn does not increase proportionally.  A more day to day example I like to use is buying coffee. The first coffee of the day is wonderful and hugely satisfying. The second is less satisfying and by the time I venture to more than three cups I’m not getting much satisfaction at all. These are both examples of diminishing returns and the same patterns can be seen in media investment.

Let’s assume we are investing in media to drive web traffic. If we buy 100 GRPs in a week we might see 100,000 visits.  But if we invest in an additional 100 GRPs in the same week we might see these incremental GRPs deliver only 50k visits. And if we invest in a further 100 GRPs in the same week we might only see 25k additional visits generated. We can see the visits we are generating fall by half for every 100 incremental GRPs we buy. This is a diminishing return and it applies to all channels from TV to PPC.

What’s the cost of this diminishing return? Given that 100 GRPs might cost £350k we can see how  taking the spend over a certain level in a specific time frame starts to reduce ROI significantly.   Whereas the first 100 GRPs generated 100,000 visits, 300 GRPs only generated 175,000 visits (100k+50k+25k). Our CPV has increased four times from £3.50 on the first 100 GRPs to £14.00 on the third 100 GRPs. When we apply these examples to large scale media budgets, we can see how diminishing returns can have a dramatic effect on media effectiveness. In the worst case scenarios budgets are set at levels so high that they risk producing no additional sales response at all.

What causes diminishing returns?  Diminishing returns are usually caused by market size constraints. If a brand has a consideration pool size of 5m consumers, with ten percent actively in market in a week or  a month, over-spending excessively against this group will not change purchase behaviour sufficiently enough to match your increased spend, you will simply spend more, sales will not grow at a proportionate rate and your media ROI will fall.

What does this mean for media planning and investment?

The challenge for media planners is to arrange media investment to leverage the carry-over effects produced by Adstock whilst reducing the impact of diminishing returns.

The main implications for media planning are around setting budget weights and phasing to leverage these two effects to maximise media effectiveness. Budget weights have to be contained within acceptable diminishing return limits while Adstock carry-overs can be used to fill gaps in a pulsing media strategy.

Further reading

Broadbent, S. (1979) “One Way TV Advertisements Work”, Journal of the Market Research Society Vol. 23 no.3

Joy Joseph, 2006, “Understanding Advertising Adstock Transformations” (independent)

Fry, T.R.L., Broadbent, S. and Dixon, J.M. (2000), “Estimating Advertising Half-life and the Data Interval Bias”, Journal of Targeting, Measurement & Analysis in Marketing, 8, 314-334

WE OFFER MARKETING MIX MODELLING TO OUR CLIENTS: FIND OUT MORE HERE

The post Adstock and Diminishing Returns: non-linear advertising effects first appeared on Marketing IQ.