Marketing IQ https://www.marketingiq.co.uk Tue, 31 Mar 2026 10:35:24 +0000 en-GB hourly 1 https://wordpress.org/?v=6.9.4 Using Bayesian Priors in MMM: Do Better‑Looking Results Mean You Have a Better Model? https://www.marketingiq.co.uk/using-bayesian-priors-in-mmm-do-better-looking-results-mean-you-have-a-better-model/ Fri, 20 Feb 2026 13:07:34 +0000 https://www.marketingiq.co.uk/?p=5374 There is an active and polarising debate about whether Bayesian priors should be used in MMM. Broadly speaking there is a progressive group of academics and […]

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

]]> There is an active and polarising debate about whether Bayesian priors should be used in MMM. Broadly speaking there is a progressive group of academics and practitioners who argue that the use of Bayesian prior values, effectively applying guideline results bounds to your model, makes models more flexible and insightful. And on the other side of the debate are the statistical purists who argue the classic approach is more reliable. The argument against using Bayesian priors is that they “fix” model outputs to meet the expectations of marketing teams, thereby undermining the very foundations of MMM:  unbiased and independent full funnel attribution. [1,2,3,4].

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.

]]> How reliable is your marketing mix model (MMM)? https://www.marketingiq.co.uk/how-reliable-is-your-marketing-mix-model-mmm/ Wed, 28 Jan 2026 21:21:49 +0000 https://www.marketingiq.co.uk/?p=5355 Can you trust your MMM? A marketing Mix Model (MMM) can produce challenging results, especially if you have been relying on last touch( LT) attribution.  When […]

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

]]> Can you trust your MMM?

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.
  1. 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%.
  1. 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
  1. 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.

]]> UK Media Adspend and Revenue Forecasts 2026 https://www.marketingiq.co.uk/uk-media-adspend-and-revenue-forecasts-2026/ Thu, 13 Nov 2025 19:25:49 +0000 https://www.marketingiq.co.uk/?p=5274 In this article, we ask how the UK media spend landscape might look as we move through 2025 and into 2026. We use publicly available AA […]

The post UK Media Adspend and Revenue Forecasts 2026 first appeared on Marketing IQ.

]]> In this article, we ask how the UK media spend landscape might look as we move through 2025 and into 2026.
We use publicly available AA WARC adspend data as a reliable source for baseline spend levels, and we use our own modelling to identify and extend underlying trends to forecast 2026 spends by channel.

 

Key UK Adspend forecasts by channel – key take-outs for 2026:

  • Cinema will enjoy moderate growth, and we forecast adspend / revenue of £254m in 2026
  • DM will continue a significant downward trajectory, and we forecast adspend / revenue of £874m in 2026
  • Magazines will continue to decline quite strongly, and we forecast adspend / revenue of £414m in 2026
  • Newspapers will also continue to decline strongly, and we forecast adspend / revenue of £656m in 2026
  • Online classified has not recovered since 2022 and is likely to continue to decline and we forecast adspend / revenue of £984m in 2026
  • Online display including social will continue to grow and on a strong trajectory and we forecast adspend / revenue breaking through £20bn in 2026
  • OOH including DOOH will continue to grow but growth rates are reducing slightly year on year, we forecast adspend / revenue of £1.5bn in 2026
  • Radio, including online radio will be stable YoY and we forecast adspend / revenue of £752m in 2026
  • Regional news will continue a strong decline, and we forecast adspend / revenue of £394m in 2026
  • Search will continue strong growth alongside online display and social and we forecast search will also break through the £20bn level in 2026
  • TV including BVOD will continue to recover vs 2023, but the overall pattern does suggest continued decline and we forecast adspend / revenue of £5.4bn in 2026

 

Let’s examine the AA WARC estimates to look at the how media spends have grown across the period 2021 to 2024 as actuals (Chart 1)

We see that total Adspend has grown significantly since 2021. Of course that was a COVID year, but recovery has been strong in terms of total spend.

UK Media Adspend by Channel (2021 to 2024 with 2025 estimate)

Totals spend in the channels shown has grown from £33.6bn to £43.8bn. Within this, Search has grown from £11.6bn to £16.9bn and Online Display has grown from £10.8bn to £16.7bn. TV and BVOD has remained relatively flat declining only slightly from £5.5bn to £5.3bn. OOH and DOOH has grown from £901m to £1.4bn.  Radio is also flat at £720m to £738m. Cinema, which was hit heavily by COVID restrictions made a strong recovery in both audiences and revenue and climbed from £103m in 2021 to £212m – more than doubling revenue.  Fallers include Online Classified £1,053m to £1,017m, Direct Mail falling from £1,082m to £964m, Newspapers £844m down to £727m, and magazines down from £556m to £469m.

Percentage YoY changes are shown in chart 2 below, with channels ordered by largest spend on the left and smaller on the right (Chart 2)

We see that as well as dominating levels of adspend, Search and Online Display (including Social) dominate growth trends. Between 2023 and 2024, these channels enjoyed YoY growth of 13% and 15% respectively.  The 2025 estimates suggest this growth might be calming slightly to 8% and 9% respectively.

UK Year-on-Year Growth in Media Ad Spend by Channel

In Chart 3, we see estimated 2025 (full year) media spend across these 11 channels.

We see Search, Online Display and Social are accounting for more than 70% of spend. TV and BVOD sit at 11% and the remaining 8 channels share the balancing 12.4%, with OOH and DOOH the largest of this group at 3%.

2025 Estimated Full Year UK Media Spend and Share by Channel

Total spend vs growth in UK media 2025 vs 2024 (Chart 4)

Chart 4 plots 2025 estimates spend and YoY growth from 2024. We see Online Display (including Social) and Search continuing to plough ahead in the top right with high growth and the highest revenues. Unfortunately, in the opposite quadrant, we see a lot of traditional print media in YoY decline; Magazines, Regional Newspaper, Direct Mail and Newspapers.

UK 2025 Est Media Spend vs 2024 YoY Growth

2025 UK Adspend full year estimates (Chart 5)

In Chart 5 below we see growth rate changes between 2021 and the 2025 estimate from AA WARC.   We see that Online Search, OOH including DOOH, Online Display (including Social), Radio and TV and VOD are in growth over the period whilst there are five consistent decliners: DM, Magazines, Newspapers and regional news all decline consistently. Online classified has declined consistently since 2022.

UK Media Ad Spend Trends by Channel (2021–2024 with 2025 estimate)

UK media adspend forecast for 2026

Chart 6 uses a modelling technique that extrapolates the evolving (non-linear) growth curves into 2026.  We see growth in 2026 for Cinema, Online Display and Social, OOH and DOOH, Radio, Search and TV and VOD, although TT/VOD is slightly down from its 2021 position. Downward trends and forecasted to continue for DM, magazines, newspapers and regional news.

UK Media Ad Spend Trends 2026 Forecast by Channel

Implications for marketing and media planning

The cost of media is generally a reflection of supply and demand – that applies across almost all biddable and non-biddable channels.  If audiences (supply) grow, media owner revenues can grow at a flat CPM. This is good news for marketers and their media agency partners (providing demand) because media owners are less likely to raise unit costs (CPMs).

In our analysis of the AA WARC data, we see that some media channels are in high growth; Cinema continues its post-Covid recovery, Online Display (including Social) and Search are powering ahead with growth of more that 8%, slightly down from last year. OOH and DOOH look positive, as does Radio.   This growth means that advertisers are likely to see more stable costs in terms of audience CPMs overall (some sub channels and audience sub groups might differ) – simply because media owners can maintain growth within significantly increasing unit costs.

But we also see that some channels are facing revenue declines; Online classified, DM, Newspapers, Magazines and Regional News. Clearly, this is a major concern for both the media owners and advertisers. If the businesses in these channels have high fixed costs, then they have to increase unit costs to increase maintain revenue. This has both short and long term implications; in the short term, media becomes more expensive and potentially performs less well. This in turn makes these channels less likely to be used in the medium term which further exacerbates the revenue decline problem.

 

Please note this caution: The 2025 and 2026 data are forecasts which identify and extrapolate the underlying growth trends for each channel. They do not account for other important factors such as the economy, the international situation, changes in consumer demand, the health of firms using marketing spends and the budget shifts they may apply in response to changing market scenarios. 

 

The post UK Media Adspend and Revenue Forecasts 2026 first appeared on Marketing IQ.

]]> Techniques to evaluate marketing uplift experiments https://www.marketingiq.co.uk/techniques-to-evaluate-marketing-uplift-experiments/ Thu, 07 Aug 2025 15:55:56 +0000 https://www.marketingiq.co.uk/?p=5173 Experiments are an important way to validate marketing effectiveness measurement.  This post will take you through some approaches to evaluating marketing uplift experiments. Let’s assume you […]

The post Techniques to evaluate marketing uplift experiments first appeared on Marketing IQ.

]]> Experiments are an important way to validate marketing effectiveness measurement.  This post will take you through some approaches to evaluating marketing uplift experiments.

Let’s assume you run a test TV / VOD campaign over 4 weeks in March 2025. How can you measure the sales revenue uplift it created? In this post, we’ll look at three ways to evaluate your marketing experiments:

  1. Year on Year measurement
  2. Causal impact studies
  3. Difference in Difference (DiD) regression
Option 1 – Year on Year uplift measurement
  • Year on year measurement is a very simple and relative clean way to make empirical judgments about marketing and media campaign performance.
  • We compare the sales pattern over time this year to the sales pattern over time last year.
  • We chunk the data over time into three phases, pre-campaign (4-6 weeks before the campaign), in-campaign (4 weeks) and post-campaign (4-6 weeks after the campaign) –  this latter stage is important as it captures post campaign effects.
  • Using an analysis tool like R or Python we can produce the following Year on year uplift outputs:
YoY Uplift Test

YoY Uplift Test by week showing change by week and campaign in grey

 

YoY Uplift Test

YoY Uplift result in pre-campaign, in-campaign and post-campaign periods

YoY Uplift Test table showing results detail

Option 2 – Causal Impact uplift measurement
  • The principle behind this technique is the measurement of an intervention, where the intervention could be our new TV / VOD campaign.
  • This technique estimates the levels of sales that would have been generated without the intervention and then estimates the weekly (pointwise) and cumulative (build) of sales after the intervention.
  • Casual Impact is very useful as it doesn’t need YoY measurement, so it’s especially useful in launch situations where historical data is limited.
  • Using an analysis tool like R or Python we can produce the following post intervention sales uplift outputs:

Causal Impact Test Example

Causal Impact Test Example showing estimate of underlying sales without intervention and observed sales (top facet), with observed weekly sales in the middle facet and the cumulative incremental sales build over time in the bottom facet.

Option 3 – Difference Regression (DiD)
  • Difference in Difference uses a regression approach to measure the difference in the changes between pre- and post-campaign periods during each year for 2024 and 2025.
  • The DiD estimate subtracts the changes observed in 2024 from those observed in 2025 to calculate the campaign uplift.
  • DiD automatically controls for underlying seasonality and year-on-year trends in the data because it is comparing changes within two different years.
Difference in Difference Uplift Measurement

Difference in Difference Uplift Measurement outputs showing campaign uplift as 224 sales per week at 5% sig.

Which option should we use?

Any of these options will give you a good measure of your campaign uplift, but once you have set up your tests, run them and collected and formatted your data, all three are relatively straightforward to run in a code environment. My recommendation  – do all three.

The post Techniques to evaluate marketing uplift experiments first appeared on Marketing IQ.

]]> What is incrementality in marketing – extracting trend, seasonality and brand equity https://www.marketingiq.co.uk/what-is-incrementality-in-marketing-extracting-trend-seasonality-and-brand-equity/ Wed, 11 Dec 2024 12:16:29 +0000 https://www.marketingiq.co.uk/?p=5010 Marketing incrementality is sales revenue that is over and above that which might be expected with no marketing activity. Establishing incrementality is critical if you want […]

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

]]> Marketing incrementality is sales revenue that is over and above that which might be expected with no marketing activity.

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.

Bike sales weekly sales data 2017 to 2022

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.

Sales trend component

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.

Sales seasonality component

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.

Sales random component

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.

]]> Brand fame; what it is, what it delivers and how to build it. https://www.marketingiq.co.uk/brand-fame-what-it-is-what-it-delivers-and-how-to-build-it/ Tue, 19 Mar 2024 17:41:23 +0000 https://www.marketingiq.co.uk/?p=4101     Let’s take a closer look at why brand fame is such an important part of building brands In this POV originally written for mSix […]

The post Brand fame; what it is, what it delivers and how to build it. first appeared on Marketing IQ.

]]>  

 

Let’s take a closer look at why brand fame is such an important part of building brands

In this POV originally written for mSix & Partners, I explore brand fame; what it is, what it delivers and how to build it. We are assisted by insight from some of advertising’s leading thinkers; Paul Feldwick, Sir John Hegarty, Byron Sharp and Orlando Wood. Simon looks at how fame contributes to brand success and identifies the media channels that are best positioned to build it.

Think of a brand and write its name here…………………………………………………………

Chances are you wrote one of the following brand names: Apple, Microsoft, Amazon, Google, Samsung, Toyota, Coca-Cola, Mercedes, Disney, Nike, McDonald’s, Tesla, or BMW.

These are some of the world’s most famous brands – let’s unpack the meaning of fame in a bit more detail. The word fame is rooted in the Latin ‘fama’ meaning fame, hearsay, kudos, renown, repute, and rumour. You will see ‘fama’ present in modern words like “familiar,” “familiarise” and “famous.” In short, having fame means being in ‘the state of being known or talked about by many people, especially on account of notable achievements [1].’ You will see that all the brands above, tick all the boxes in the Latin phrase book.

Within the market context, fame is slightly more nuanced. Brands are famous because they have intrinsic appeal, they communicate with mass audiences, they demonstrate distinctiveness, and benefit from wide social diffusion. These are, according to account planning’s elder statesman Paul Feldwick, the key components of brand fame [2].

The Marketing Society and ITV [3] described the main elements of brand fame as connection, standout, talkability, familiarity and universal meaning, with universal meaning and familiarity being the most important components of the set. Connected to these elements is consumers’ need to be seen to be making endorsed choices; both during and especially after purchase. Mass appeal equates to mass endorsement.

In this POV we will think more about brand fame;  why it’s important and how it is achieved, and how, over the last decade we may have lost sight of the best ways to build brand fame.

Why is fame important?

Fame is important for three reasons. First,  high fame means high mental availability and we know from the work of Byron Sharp [4] that high mental availability confers commercial benefits. Feldwick summarises this as being “thought of more often, by more people, and therefore chosen more often by more people.” Second, fame can disturb consumers’ economic rationality. This is one of the main, and sometimes overlooked, functions of ‘brand’ advertising. If consumers are prepared to pay a 10% price premium for a brand because, in Feldwick’s words, reasons of intrinsic appeal, mass audience communication, distinctiveness and social diffusion, then the brand will generate more scale and revenue. And thirdly, if the brand can use its fame to sell more volume through higher purchase frequency at that slight price premium, then the brand’s revenue will be even greater.

How do we build brand fame?

Now we know the value of fame, we can explore how it is built. Because the ‘whole’ of brand fame is composed of intrinsic appeal, mass audience usage, distinctiveness, and social diffusion ‘parts,’ it follows that our strategies to build brand fame should be strategies to build those component parts.

Whilst intrinsic appeal is driven by product utility, mass audience appeal and usage, distinctiveness and social diffusion can be assisted by marketing communications like advertising, media, and PR activity.

And here’s where the quest for fame becomes more challenging. If we want brand fame, we need to seek connections that are not necessarily logical but which appeal to right brain. These connections are not driven by rationality, but by emotional appeal. System 1’s Orlando Wood summarises this in his book “Lemon” he says the right brain is guided by implicit connections, empathy, novelty and metaphor. Contrast this with the left brain, dominated by explicit facts and logic, “cause and effect, literal, factual” [5].

How do we apply this to media strategy?

Rather ironically, we have media insight from one of the industry’s best known creative practitioners, As recently as March this year, Sir John Hegarty made an impassioned plea in the BBC’s CEO Secrets series, arguing for more use of broadcast advertising, “if you are constantly wanting to expand your brand, make it more famous and add value to it – only broadcast does that” [6] . In addition to Hegarty’s comments, we have strong clues from more of advertising’s most respected and prolific thinkers. Paul Feldwick talks about mass audiences whilst Byron Sharp extolls us to maximise mental availability.

Most importantly, Orlando Wood requires that we stimulate emotions – which points again towards the channels that can do that, TV, AV, VOD, Cinema, and online video. The fact that the moving image elicits an emotional response is long-proven by both academics and empirical experience. This response has been researched extensively by Uri Hasson and his team at NYU who have coined the phrase “Neurocinematics”. The NYU team found that film can elicit a powerful neuro response, provided that the film itself is structured in certain ways [7]. Although the work was originally based on movie responses, the learnings on how to tell a story to maximise emotional ‘System1 response’ are clear throughout the paper.

These findings about using film to build emotional connections are also corroborated when we build models to analyse marketing and media effectiveness. When we analyse different forms of media activity, we find that these moving image channels are often delivering some of the strongest results.

In conclusion:

Fame builds brands and emotional connection builds fame, and the moving image builds emotional connection. So, if you want to generate powerful emotional engagement with your brand, use moving image media channels. If you want to read more about the connections between emotional engagement, cinema, brand development and the impact on short-term performance, in an applied media planning context, then the DCM Cinema Effectiveness Roadmap [8] is a good place to start.

References

  1. Oxford Languages  / Dictionary definition of Fame
  2. Why does the pedlar sing? Paul Feldwick, 2021
  3. How much is fame worth to the bottom line? Market Leader, 2005
  4. How Brands Grow Byron Sharp, 2010#
  5. Lemon, Orlando Wood, System1, 2019
  6. Has Social Media killed the famous ad?” BBC News 14 March 2023
  7. Neurocinematics: The Neuroscience of Film Uri Hasson, Ohad Landesman, Barbara Knappmeyer, Ignacio Vallines, Nava Rubin, and David J. Heeger, NYU, 2008
  8. Cinema Effectiveness Roadmap, Digital Cinema Media, 2023

The post Brand fame; what it is, what it delivers and how to build it. first appeared on Marketing IQ.

]]> Why digital media attribution could be compromising your media ROI https://www.marketingiq.co.uk/why-digital-media-attribution-could-be-compromising-your-media-investments/ Tue, 17 Oct 2023 08:25:09 +0000 https://www.marketingiq.co.uk/?p=4001 You’ve probably heard the expression ‘The devil is in the detail‘. It tells us that focusing on detail is the way to solve problems.  In many […]

The post Why digital media attribution could be compromising your media ROI first appeared on Marketing IQ.

]]> You’ve probably heard the expression ‘The devil is in the detail‘. It tells us that focusing on detail is the way to solve problems.  In many ways, this expression is true, but in this post I’d like to argue that placing too much focus on the digital detail can mean marketers and their agencies miss the bigger picture and it is, in fact, the big picture that drives your commercial sales and success.

How marketing and media metrics have changed

Prior to around 2005, the main metrics marketers used were of three types:

  1. The media metrics that monitored the delivery of their campaigns – GRPs, reach, frequency etc.
  2. The attitudinal metrics that measured how these campaigns had changed attitudes towards their brands – e.g. brand consideration, preference and purchase intent.
  3. And of course, commercial metrics that captured the impact of marketing investments: unit sales, value, volume, purchase frequency and market share.

Since 2005, the digital media industry and particularly its giants, Google, Facebook Microsoft have produced huge amounts of microscopic detail covering almost every digital movement made by millions of online consumers. Through the cookie, we are able to see exactly where consumers have been, what they’ve looked at, what they’re interested in, where they have engaged, what they have registered for and what they have bought. And modern marketers inhabit this world of tracking, measuring, analysing and reporting the microscopic detail produced by digital media owners and their platforms.

Real time micro-measurement has become the main source of campaign performance insight for a generation of marketers. It is relied up by marketers and their agency partners across the industry and across the globe.  Micro-performance data is used to set budget and optimise campaign on the presumption that it is accurate and correct. But what if it isn’t accurate and it’s not correct?

Some senior marketers in leading brands have questioned real-time digital measurement data. Here are two examples:

‘This real-time ROI can mean brands get tempted into ploughing investment heavily into digital – but, actually, he noted, that can result in short-termism that doesn’t ultimately grow the brand or sales, and can give “misleading” results – Simon Peel, Global Media Director, Adidas.

‘Digital attribution doesn’t take into account the [full consumer journey], [like] the fact [that consumers have] been influenced by a TV ad, or that their mum recommended this product to them. While it’s brilliant that we’re getting more accurate with digital measurement, there are so many more factors that influence why and what the customer does’- Rosie Hanley, Head of Marketing, eBay

This problem may be even worse that it looks when we consider the opportunity cost of doing the wrong thing

There is good evidence that managing and optimising this digital performance detail compromises your overall media ROI and even worse, too much focus on this detail can harm a brand’s commercial health and have a major opportunity cost. Here are four very strong large-scale case study examples that have provided support for this point:

Case study 1 – Airbnb
  • In 2020 AirBnB cut $50 million of performance media investment. The result: it made no difference to their overall business performance.
  • During an earnings call in February 2023, Airbnb CEO Brian Chesky said that AirBnB now sees the role of marketing as evolving from buying customers to educating markets and has shifted its marketing priorities accordingly.
  • Airbnb CFO, Dave Stevenson added that this strategic change in marketing had proven to be incredibly effective during the period 2020 to 2022. He added “Our brand marketing is delivering excellent results overall with a strong rate of return, and it’s been so successful that we’re actually expanding it to more countries”.
  • Great news. But consider for a moment the resource costs required to deliver the digital planning, activation, tracking, measurement and reporting that $50 million of performance marketing spend would require.
Case study 2 – Adidas
  • AirBnB are not alone. Around the same time, Adidas undertook a similar shift. The result: they concluded that they had too much focus on short term ROI and this had led them to over invest in performance marketing at the expense of brand building.
  • What’s interesting about the Adidas case is that they had previously assumed only performance activity drove e-commerce sales (ie total reliance on the digital ecosystem), but further analysis showed the brand development activity was actually driving 65% of sales across wholesale retail and e-commerce.
  • At that time Adidas’ marketing investment was split 77% into performance and only 23% into brand.  The reason for this misalignment was an overfocus on short term digital performance metrics. Simon Peel, the global head of media at Adidas, called out some specific metrics as being responsible: Google last click, Google custom, Adobe and Facebook, and within these platforms, too much of an emphasis on short term, real time measurement.
  • This cycle was only broken when Google AdWords went down in Latin America and search was halted. During this time, Adidas did not see a dip in traffic or revenue from search marketing activity.
Case study 3 – ASOS
  • The third case study is ASOS, who also made a similar set of discoveries. Across the 2020-22 period more than 80% of the ASOS marketing investment had been put into performance marketing. 
  • According to ASOS new CEO, Jose Antonio Ramos Calamonte, insufficient levels of brand investment was a contributory factor to a slowdown in customer acquisitions. Calamonte observed that historically ASOS had under invested in marketing relative to its peers (aka Share of Voice), and that marketing spend had not been “effectively prioritised”, or “managed effectively” to ensure a return on investment.
  • As in the case of Adidas, it was a halting of spend, in this case brand spend, that led to the change in marketing investment thinking; after pausing a broad reach [brand] campaign in the US, ASOS saw customer acquisition and visits growth slow.
Case study 4 – eBay
  • In 2015 eBay was spending 90% of its budget on performance using hyper-targeted product to audience techniques. By 2017 revenues had fallen to pre-2010 levels at $7.4bn.
  • By 2022 it had switched back to full funnel marketing and a focus on the experience of using the eBay brand. Revenues grew to $9.8bn.
  • In a 2022 earnings call CEO Jamie Iannone said the shift away from “just lower funnel optimisation has worked out really well for us”.
  • These four brand case studies are further supported by multiple additional studies. In March 2022, Kantar chimed into the debate saying, “There is inalienable evidence that unbalanced brands won’t win in the long term. Multiple Kantar studies reveal that if marketing mix allocation consistently favours performance marketing, baseline sales will steadily weaken”.
Case Study 5 – Uber
  • In 2018, Sundar Swaminathan, an analyst at Uber was reviewing data and suspecting that Meta was not driving incremental returns in Uber new driver acqusition.
  • As a result of his recommendations, Uber ran a dark test turning off Meta acquisition activity for new riders in a test region.
  • The test ran for three months.
  • The results of the test showed that there was no incremental gain from Facebook activity.
  • Uber turned off this activity permanently across the US and Canada and saved $35m.
Is there any robust experimental research evidence to further support this view?

Yes. A brilliant and comprehensive large scale, field experiment designed to measure the true effectiveness of brand and generic keyword search terms was undertaken by eBay and the university of Chicago in the US in 2013.

These were not small scale tests but large scale experiments. One stopped bidding on a 30% sample of eBay’s US traffic across a 60 day period.

This study sought to understand whether search marketing really has any genuine incremental uplift effect on consumer purchase behaviour. Here is what the eBay experiments found:

The brand, keyword, advertising experiments found that halting brand terms resulted in no detectable drop in traffic and sales.

Search engine marketing did have a significant effect on new registrations and those consumers with a low purchase frequency <2, but this was not sufficient to offset inefficient results across higher frequency eBay users.

The generic keyword experiments showed that search engine marketing had a very small and insignificant effect on sales.

Conclusion and actionable insight

These case studies make clear that an overemphasis on the detail of performance marketing does not add value to the business and risks a significant opportunity cost through misplaced marketing budget investment.

This is not just about the unhelpful “brand” and “performance” categories and nor is it about digital versus traditional mainstream high reach media. The problem is around why and how much marketing budget we deploy across all channels. It’s about the objectives we set, the strategies we develop, the plans we implement, and the way we measure and optimise.

In terms of actionable insight, simple Occam’s Razor maths tells us that in the case of Adidas, if 23% of budget was driving 65% of sales then 35% of budget could deliver 100% of sales. And, more importantly, shifting more budget into brand would grow sales substantially. In this case, 50% of budget could potentially grow sales by 150%. That’s a 50% increase in sales for 50% of the current budget.

More broadly, we must ask, how much more shareholder value would have been created if the $50 million spent by Airbnb would have been generated if this money had been focussed on growing market penetration, purchase, frequency, and overall market share?

If you are working in a category where the majority of spend is over committed to performance marketing, you have a significant opportunity to build share whilst your competitors over optimise activity that is probably not contributing to business growth.

And meanwhile, over at Google

The company posted annual revenues of $182bn in 2020, $257bn in 2021 and $280bn in 2022.

Just imagine the increases in market penetration, purchase frequency and market share that marketers would have generated if just a fraction of that revenue had been invested in building and strengthening in high reach media.

The post Why digital media attribution could be compromising your media ROI first appeared on Marketing IQ.

]]> What are the 4Ps of the Marketing Mix https://www.marketingiq.co.uk/what-are-the-4ps-of-the-marketing-mix/ Sat, 13 May 2023 21:04:02 +0000 https://www.marketingiq.co.uk/?p=3901 The 4Ps are one of the key concepts that underpin marketing strategy and tactics. The Ps stand for Product, Price, Place and Promotion. They were conceptualised […]

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

]]> The 4Ps are one of the key concepts that underpin marketing strategy and tactics. The Ps stand for Product, Price, Place and Promotion. They were conceptualised by the distinguished US marketing and research academic, E Jerome McCarthy.

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.

]]> Why we must learn to build brands in digital media https://www.marketingiq.co.uk/why-we-must-learn-to-build-brands-in-digital-media/ Sat, 15 Apr 2023 08:44:47 +0000 https://www.marketingiq.co.uk/?p=3883 In this article, written for the mSix’s website, I highlight how digital media has passed the tipping point and why marketers must evolve beyond the simplistic […]

The post Why we must learn to build brands in digital media first appeared on Marketing IQ.

]]>

In this article, written for the mSix’s website, I highlight how digital media has passed the tipping point and why marketers must evolve beyond the simplistic ‘long and short duopoly’ and use digital media to build brand fame as well as driving performance.

The time is right to challenge and develop the original works of Binet and Field.  ‘Marketing in the Era of Accountability’ [1] and ‘The Long and the Short of It’ [2] are now more than a decade old. This work reflected the mesia landscape as it as then and put TV at the heart of brand building, but since then the media consumption landscape has changed – and it has changed dramatically.

In 2013, and taking the US as an example, TV, radio, and print dominated with 53% of media consumption minutes. Within this, TV delivered 4:30 minutes per day (38%). At the same time, digital channels accounted for the 4:50 mins (40%).   But by 2022, US TV, radio, and print consumption had fallen to 34% and US mobile and desktop had grown to around 62% of consumption (although total media consumption minutes have increased by around 15%, likely due to mobile ubiquity) [3]. Most significantly, US mobile consumption has almost doubled from around two hours per day to over four. Patterns for the UK are broadly similar, with 2:26 mins on TV in 2012 failing to 1:42 in 2022 and 24:00 mins per day in Social growing to 1:17 mins by 2022 [4].

Against these tectonic shifts, Binet and Field’s 2013 argument that TV is the prerequisite brand-building channel must be challenged and, as a consequence, we need to ask, how do we build brands in digital media?

Traditionally, TV has been the place where ‘System 1’ messaging has been delivered through  emotional connections, fame and reach.  System 1 messaging is a core ingredient of fame building. This type of communication aims for high mental availability and fast effortless decision making through ubiquity – which is essentially, fame.  At the core of this thinking is reach – reaching as many of your potential audience as possible. Put in a slightly different way by the late Jeremy Bullmore, “if you want to be as famous as BMW, it’s no use being known only by the tiny percentage of the population who can afford to buy your car today”.

Until recently, digital channels have been used primarily to target these “tiny percentages” with System 2 “buy now” messaging. But this approach severely undervalues the reach potential of digital channels.  According to IPA Touchpoints the post lockdown high reach digital media channels are social media and functional internet (commercially funded websites which are not for media, social media or communication e.g. search, shopping, researching). These channels can deliver 70% to 80% weekly all-adult reach, putting them on a par with commercial TV and way ahead of commercial radio, magazines, newsbrands and cinema.

Of course all this means moving your guardrail KPIs from performance to brand metrics like attitudinal shifts, recall, preference, purchase intent, and incremental brand growth. Measurement and monitoring techniques need to shift uplift experiments, MMM and tracking studies.

The shifts in focus that are required to build brands in digital are summarised in the checklist below:

  1. Targeting: Move from tight signals targeting to brand audience reach
  2. Mental message processing: Move from System 2 (aim to close the sale) to System 1 (aim to change instincts)
  3. Messaging: Move from rational to emotional engagement
  4. Mental availability: Move from Low to High
  5. Optimisation: Move from short term CPA to longer term attitudinal metrics like consideration and purchase intent
  6. Evaluation metrics: Move from performance metrics to a set of agreed attitudinal metrics
  7. Evaluation cadence: Move from short term to medium term

By changing the way we plan, activate, measure and benchmark digital channels we unlock their ever-expanding potential to build and reinforce brand attributes and, in doing so, prove it’s now  time to move beyond the long and short of it.

References

  1. Binet and Field, ‘Marketing in the Era of Accountability’ IPA 2007
  2. Binet and Field, ‘The Long and the Short of it: Balancing Short and Long-Term Marketing Strategies’, IPA 2013
  3. ‘Average time spend with media in the US’, eMarketer April 2016 and April 2022
  4. IPA Touchpoints 2023
  5. IPA ‘Making Sense: the Commercial Media Landscape’ 2022

The post Why we must learn to build brands in digital media first appeared on Marketing IQ.

]]> Towards Attention Metrics by Media Channel https://www.marketingiq.co.uk/towards-attention-metrics-by-media-channel/ Thu, 05 Jan 2023 17:28:39 +0000 https://www.marketingiq.co.uk/?p=3821 Research into attention to advertising is going to drive a rethink in how we view media channels Attention metrics concern the degree of attention consumers give […]

The post Towards Attention Metrics by Media Channel first appeared on Marketing IQ.

]]> Research into attention to advertising is going to drive a rethink in how we view media channels

Attention metrics concern the degree of attention consumers give to different advertising media channels.  This is important if we accept that higher attention is likely to lead to higher cognition and memory retention. The value of attention isn’t universally agreed and the way attention converts to memory is a complex area [1]. If you take the view that advertising is consumed on a subliminal level, then the case for attention is possibly diminished. But if you believe that when information receives less attention, memory encoding decreases [2] then obviously attention is important. Either way, attention metrics are now firmly in the advertising and media planning Zeitgeist [3, 4] and there is no doubt that these metrics are going to be a big feature of media research, strategy and planning into 2023 and beyond. 

So, how do we address the issue of attention by media channel?

Lumen Research [5] have undertaken a study to estimate average dwell time as a proxy for attention to a range of channels.  Lumen’s findings are outlined below:

The formula being used by Lumen is % view x av. eyes-on dwell time x 1000 impressions. But in my view this formula could benefit from consideration of size of screen, distance from screen and clutter.

Let’s develop this further by estimating screen size and viewing distance ratio for each channel.

I have estimated screen sizes in feet so for a example, an average cinema screen is about 65 feet, an average TV screen about 3.5 feet (42″, diagonal) and the average PC or tablet size 1.2 feet (15″ diagonal). Mobile phone sizes are estimated at 3″ across. I have also added a metric for distance. Distances are also in feet with average distances from cinema seat at 50 feet, TV and BVOD at 10 feet, PC or Tablet at 1.25 feet and Mobile devices at 1 foot from the viewer. I have not included OOH as distances and screen sizes vary significantly – think tube cross track 6 sheet vs roadside 48 sheet. I have also excluded radio as size and distance metrics are not relevant. To create the size / distance ratio, size is divided by distance.

Adding clutter metrics to attention metrics

To add clutter metrics, I have used a scale of 1-5 where 1 is low clutter and 5 is high. Cinema, TV, VOD and BVOD advertising messages tend to be delivered sequentially so there is no surrounding clutter from a visual perspective. PC and mobile display advertising is often delivered in parallel and so tends to attract higher clutter. Mobile can have higher clutter and in some cases we see multiple ads, underlays and overlays being observed in the same content feeds – this can be seen in the example below [6], with two ads running simultaneously in a recipe page:

 

Combining these, we derive the the following size/ distance ratios and derive a clutter-weighted ratio by multiplying the size distance ratio by the clutter weighting metric (Ratio incl Clutter).

Media channel attention metric ratios

The Ratio including the clutter weighting can be graphed as below:

Cinema, given its dominant delivery scores highest, followed by full screen incline VOD, TV and BVOD within the range 1.30 to 0.35. Display PC activity ranges between 0.24 and 0.19. Mobile channels with their very small screen size and high clutter score in the 0.08 to 0.06 range. PC-based VOD scores highly when the screen size and closeness to screen are combined with the low clutter of an inline delivery.

Comments and adjustment suggestions welcome.

 

  1. Interactions between attention and memory, Marvin M Chun and Nicholas B Turk-Browne, Science Direct 2007 https://ntblab.yale.edu/wp-content/uploads/2015/01/Chun_CONB_2007.pdf
  2. Memory and Attention Long, Kuhl, and Chun in Stevens’ Handbook of Experimental Psychology and Cognitive Neuroscience (pp.1-37)
  3. What are attention metrics and why are they crucial for digital advertising? Mateusz Jędrocha, The Drum, August 25, 2022 https://www.thedrum.com/profile/rtb-house/news/what-are-attention-metrics-and-why-are-they-crucial-for-digital-advertising
  4. No Longer a Novelty, Attention Metrics are Now Fully Ingrained in Agencies’ Planning and Measurement, Tim Cross VideoWeek, 21 July, 2022https://videoweek.com/2022/07/21/no-longer-a-novelty-attention-metrics-are-now-fully-ingrained-in-agencies-planning-and-measurement/
  5. Lumen Research, “Media Buying” https://lumen-research.com/media-buying/
  6. Bread in four easy steps, BBC Good Food, https://www.bbcgoodfood.com/recipes/bread-four-easy-steps, retrieved 12 June 2024

 

The post Towards Attention Metrics by Media Channel first appeared on Marketing IQ.

]]>