The final campaign report is a poor place to decide what success was supposed to mean. By then, a team has seen the results and can select the most flattering interpretation. DAGMAR addresses a more demanding task: describing the intended contribution of advertising before the outcome is known. Its historical importance lies as much in that discipline as in its familiar sequence of four stages.
Russell H. Colley’s Defining Advertising Goals for Measured Advertising Results appeared through the Association of National Advertisers in New York in 1961. Its title supplied the acronym DAGMAR. Smithsonian Libraries records the original publication as an illustrated book of 114 pages, while the Monash University marketing dictionary also identifies the term with Colley’s 1961 work (Colley, 1961; Monash Business School, n.d. a).
The central proposition was not that advertising had suddenly become completely measurable. It was that the task should be defined so its accomplishment could be assessed. A clear objective, an observed change and evidence of what caused that change remain different things. Preserving those distinctions makes the history more useful than presenting DAGMAR as either a universal formula or an obsolete set of initials.
A Practical Problem for Advertisers and Agencies
The Association of National Advertisers provided a setting concerned with the needs of organizations paying for advertising. Colley’s publication addressed a managerial question within that setting: how could an advertiser and its agency determine whether advertising had made the contribution expected of it? It was an approach to objectives and evaluation, rather than experimental proof of a universal psychological law.
Monash Business School defines the approach in terms of converting advertising objectives into specific, measurable goals (Monash Business School, n.d. b). The order matters. Selecting a sophisticated research technique does not settle what the organization wants advertising to accomplish. An investigation can be technically elaborate and still provide an answer to an ill-defined question.
Analytically, three responsibilities meet here. The budget holder needs a reason for spending. The agency needs an intelligible assignment. The researcher needs an assessment criterion that will not be rewritten after the findings arrive. An explicit objective can connect these responsibilities without removing all differences of interest between the participants.
This helps explain why the concept extends beyond a diagram. Four stages are easy to memorize. Agreeing on a particular task is harder because it reveals expectations that might otherwise remain vague. It also exposes a possible disagreement over responsibility: a campaign can be expected to explain an unfamiliar service without being able, on its own, to ensure that the service is affordable, available or suitable for every potential buyer.
Communication Objectives and Commercial Objectives
A company may seek additional sales while assigning an individual campaign the more immediate task of increasing awareness. Those objectives can be connected, but they are not interchangeable. A higher awareness score is not already an additional sale. Equally, unchanged sales do not establish that no communication effect occurred.
In his criticism of hierarchy models, William M. Weilbacher notes that Colley himself recognized advertising as one influence among several contributing to sales. Product and availability also matter (Weilbacher, 2001). Communication objectives therefore specify advertising’s intended contribution; they do not require ignoring the company’s commercial purpose.
An analytical distinction follows. A business objective describes a desired economic outcome. A communication objective identifies a change among particular recipients that is expected to support that outcome. Whether the proposed connection exists, how strong it is and whether it is sufficient are questions requiring evidence. Stating the connection does not prove it.
This guards against two simplistic judgments. Advertising cannot reasonably be treated as the sole cause of every short-term movement in sales. Communication measures also cannot be treated as permanent evidence of commercial success regardless of their relevance. A useful objective makes the proposed relationship visible and allows it to be questioned. If people understand an offer better but still find it unsuitable, the organization needs to learn from that difference rather than rename understanding as a completed commercial achievement.
What Makes an Objective Assessable?
George E. and Michael A. Belch explain DAGMAR through a concrete communication task, a specified audience, a starting benchmark, a desired change and a period for accomplishment. They also describe Colley’s checklist of 52 advertising tasks as an aid to establishing objectives, rather than 52 compulsory stages (Belch and Belch, 2003, pp. 206–208).
An entirely invented example illustrates the distinction. An industrial maintenance provider might want 45 per cent of a defined group of production managers to name its service without prompting within four months, compared with a starting level of 25 per cent. These numbers are illustrative; they are not results from a historical campaign described by Colley.
The statement immediately requires more detail. Which production managers qualify? Which region and kinds of factory are included? What counts as a correct answer? Is the same question asked under comparable circumstances? Without those decisions, the numerical target has an appearance of precision but an uncertain meaning.
The target itself also needs justification. A specific number can be unrealistic, too easy or irrelevant to the business problem. Adding a date does not make it reasonable. The point is to combine an unambiguous definition with a defensible expectation of progress. Otherwise, an organization can comply with the form of measurable planning while avoiding its substance: deciding what change would actually matter and why that change is plausible within the available time and resources.
Awareness: Known by Whom, and in What Sense?
The commonly taught DAGMAR sequence consists of Awareness, Comprehension, Conviction and Action, often abbreviated ACCA. Rishi Chakravarty and Nripendra Narayan Sarma identify this sequence with Colley’s 1961 contribution in their historical review of hierarchy-of-effects models (Chakravarty and Sarma, 2022). Awareness concerns knowledge of an offering’s existence, rather than a complete understanding of it.
A contemporary application must specify the form of awareness being examined. Recognizing a name when shown it and producing the name without a prompt are different tasks. Their results should not be silently combined under a single general claim. The wording of the objective determines which task is appropriate.
The relevant population matters just as much. A specialist service can be widely recognized among people who will never need it and poorly known among those who might use it. Conversely, a niche supplier does not necessarily need mass public recognition to accomplish its intended task. A percentage is informative only in relation to the group whose responses have been counted.
As an analytical matter, awareness opens possibilities rather than guaranteeing their realization. Someone can know a brand and have no use for it. Someone can recognize a name while misunderstanding its activity. A report should therefore explain what an awareness finding establishes and which further conclusions remain unsupported. That restraint makes the measure more useful, because it preserves the distinction between a problem of recognition and a problem elsewhere in the offer or communication.
Comprehension: Recognition Without Understanding
Comprehension directs attention to what recipients understand. Chakravarty and Sarma describe the move from awareness to understanding the communicated message as part of the ACCA sequence (Chakravarty and Sarma, 2022). The distinction is especially helpful where an offering is unfamiliar or difficult to explain.
Consider an imaginary software supplier whose name is recognized but associated with the wrong application. Increasing name recognition alone would not necessarily correct that misunderstanding. The communication task might instead be to establish what the software does and which problem it addresses. A test of remembered names would provide limited evidence about that particular task.
The question used in the research should not supply the understanding it is supposed to measure. Explaining the benefit at length and then asking whether respondents agree is different from investigating what they already know. Similarly, choosing the right answer from a very simple list need not show a usable understanding of a complex service.
These are methodological implications rather than descriptions of a particular study conducted in 1961. They show why selecting a stage label is only a beginning. A term needs to be translated into an observable task, and that task must correspond to the claimed change. Without that connection, the result may look like evidence of comprehension while reflecting recognition of wording, an educated guess or information introduced by the measurement process itself.
Conviction Is Not Simply Liking an Advertisement
In the familiar ACCA account, conviction concerns a favourable disposition towards the offer and the prospect of action. Chakravarty and Sarma discuss the movement from understanding towards persuasion and purchase (Chakravarty and Sarma, 2022). Applying the distinction requires care about the different positive responses people can give.
An advertisement can be enjoyable without its claim being credible. A claim can be credible without being relevant to the respondent. An offer can be appropriate while a purchase remains impossible because there is no current need or budget. A general statement of approval does not adequately distinguish those situations.
For the hypothetical maintenance service, the researcher might examine confidence in the provider’s ability, perceived suitability or a stated willingness to request information. The measures could be related, but they should not be exchanged after the results are known. A team should not report whichever one improved as if it were the objective originally agreed.
The analytical benefit is diagnostic. If understanding improves while perceived suitability remains unchanged, the message may be clear but the proposal unattractive. Repeating the same explanation more often would not automatically resolve the difficulty. Evaluation should help distinguish possible decisions, including whether to change the message, reconsider the offer or revise the initial assumptions about demand. A favourable score is useful when it supports that reasoning, not merely because it makes the campaign look successful.
Action and the Meaning of a Counted Event
Action occupies the final position in the classical ACCA sequence, and the historical review discusses it in relation to purchase (Chakravarty and Sarma, 2022). Modern applications often use the underlying idea for other observable actions. Such adaptations should be acknowledged as contemporary choices rather than attributed retrospectively to Colley.
A click, a registration, an enquiry and a completed purchase are distinct events. Calling them all actions does not give them equivalent value. A specialist information campaign might reasonably seek a qualified enquiry, whereas a campaign for an immediately available product may set a target closer to an order. The measure should follow the purpose.
An observable action can also be a poor indicator of a useful outcome. A large volume of unsuitable enquiries can increase the workload of a specialist supplier without improving its position. Defining the desired action therefore includes considering its quality, not merely its ease of recording. Otherwise, formal target achievement and practical disappointment can coexist.
This is an analytical extension of the objective-setting principle. DAGMAR does not establish that any particular digital event is always the right indicator. It encourages a more specific question: what should happen, among whom, and why should that occurrence count as progress? The answer must be settled in relation to the actual task, not selected because a platform happens to display a convenient number.
Several Models in the Same Year
The October 1961 issue of the Journal of Marketing carried a separate contribution by Lavidge and Steiner, entitled A Model for Predictive Measurements of Advertising Effectiveness. The publisher’s abstract begins with what advertising is supposed to do and how its functions should inform measurement choices (Lavidge and Steiner, 1961).
Their familiar sequence distinguishes Awareness, Knowledge, Liking, Preference, Conviction and Purchase. It is a six-stage model, rather than Colley’s four-stage ACCA account. Chakravarty and Sarma treat both as distinct contributions within the same period (Chakravarty and Sarma, 2022). Sharing a publication year does not establish that the models were identical or emerged from a common research commission.
AIDA also needs to remain distinguishable. Attention, Interest, Desire and Action form a different sequence. These frameworks belong to a longer tradition of organizing responses into stages, but a shared family resemblance does not permit their names and authors to be exchanged without explanation.
The historical question is consequently more demanding than asking which diagram has more boxes. Additional stages do not automatically provide greater explanatory power, and fewer stages do not automatically produce a better practical tool. A model must be considered in relation to its purpose, the observability of its concepts and the evidence supporting the proposed connections. Clear attribution is part of that assessment because it prevents later summaries from merging genuinely different contributions.
Advertising Measurement Before 1961
Stefan Schwarzkopf examines measurement systems developed by consumer psychologists and market researchers during the 1920s. His study of the Starch Advertisement Readership Service also considers advertising artists’ resistance to evaluation through attention, memory and related indicators (Schwarzkopf, 2015).
This earlier history changes the meaning of Colley’s milestone. He did not invent observing attention or comparing reactions. His contribution concerned a particular relationship between defining an objective and judging its accomplishment. That is different from developing every measurement technique that later users might employ.
The tension between creativity and measurement therefore cannot simply be described as a reaction to DAGMAR. Numerical comparisons could help advertisers evaluate commissioned work. Creators could regard those comparisons as an inadequate account of what their work achieved. Both positions concerned substantive responsibilities, although they did not necessarily produce agreement about the appropriate criteria.
A historical account should consequently avoid dividing the past into a completely unmeasured era followed by a scientific age beginning abruptly in 1961. Research procedures, commercial interests and theories developed together and sometimes conflicted. DAGMAR became important within that environment. Its significance is easier to identify when it is placed among existing practices rather than imagined as a solitary solution introduced into a world that had never attempted to investigate advertising responses.
Earlier and International Examples: Lux Across the Atlantic
Schwarzkopf’s study of Lever/Unilever and J. Walter Thompson examines surveys in the 1920s and 1930s that helped reposition the soap brand Lux. It emphasizes early qualitative as well as quantitative consumer research in Britain and the United States (Schwarzkopf, 2009).
The case documents an earlier connection between investigation and commercial decisions. It does not establish an early use of DAGMAR. Making that claim would require comparing the actual objectives, procedures and vocabulary of those participants with Colley’s later approach. A preceding practice is not automatically a direct conceptual ancestor.
The introduction to The Routledge Companion to Marketing History similarly presents market research as having multiple geographical and social-scientific roots, including British survey and sampling traditions (Tadajewski and Jones, 2016, pp. 3–4). The broader setting challenges a simple account in which research knowledge moved in only one direction from an American point of invention.
Recognizing those histories does not obscure the American publication context of Colley’s work. It keeps the scale of the claim appropriate. Advertisers, agencies and researchers developed relationships in several settings, and comparing those relationships can explain more than awarding a series of worldwide firsts. The useful historical distinction is between documenting a particular contribution and making that contribution stand for the entire emergence of a field.
An Older Argument About Advertising’s Value
In a 2007 CHARM contribution, Yumiko Toda reconstructs Paul Cherington’s writings of 1913 and 1928. Her account describes his defence of advertising through information, goodwill and possible economic benefits, situating his work within a broader discussion of advertising’s role (Toda, 2007).
That debate asks a different question from an individual DAGMAR objective. Assessing the economic or social value of advertising as an activity is not the same as determining whether a particular campaign achieved a specified communication result. Their difference makes the comparison useful rather than irrelevant.
A campaign could reach a planned awareness score without resolving every social criticism of advertising. Conversely, a general argument that advertising can inform consumers does not demonstrate that every advertisement is well designed or worthwhile. Claims made at one level should not be treated as proof at another.
DAGMAR consequently belongs within a long history of justification, management and investigation. It addresses an important portion of that history explicitly. Its boundaries become clearer when the reader distinguishes the evaluation of a business decision, the explanation of a psychological response and the assessment of an entire commercial institution. All can be legitimate subjects, but they need different evidence. A campaign’s measurable objective is a beginning for a particular inquiry, not a complete answer to every question about advertising.
Research During the Creative Revolution
Patricia M. Kinneer’s historical study of advertising between 1958 and 1972 examines tensions among creative work, research and managerial demands for evidence. It places DAGMAR in an environment where establishing a clear advertising–sales relationship remained contested (Kinneer, 2001, pp. 12–13).
The setting prevents an excessively smooth account of adoption. A publication could call for explicit objectives without practitioners already agreeing on their selection or authority. Agencies and clients still had to decide which investigations were useful, how much to spend and what to do when results conflicted with expectations.
An analytical implication also complicates the apparent opposition between creativity and research. A clear assignment can help creative work by making the intended task intelligible. A poorly selected criterion can favour an execution that produces an easily tested response while serving the underlying purpose less well. The difficulty is then the suitability of the criterion, rather than the mere existence of measurement.
The productive question is therefore which information would improve a particular decision. DAGMAR can prepare that question by making the task explicit. It does not automatically identify the strongest idea or guarantee a satisfactory way of observing every effect that idea may have. Evaluation and creativity have to be connected through the problem being solved, not simply placed on opposite sides of a historical argument.
A Baseline Is More Than an Administrative Detail
The starting measurement in a contemporary application deserves attention before the campaign begins. Its timing determines what “before” means. If publicity or an early release has already reached some recipients, the initial result may no longer represent a wholly unexposed situation. That does not make the measurement useless, but it changes the comparison that can reasonably be described.
The population and instrument also need continuity. An initial survey among established customers followed by one among prospective customers would not describe the same group’s aggregate situation. Changing the wording or introducing a prompt can similarly change the task. An apparent improvement might partly reflect the measurement procedure rather than the intended communication.
These are methodological implications of a comparison, not additional historical instructions attributed to Colley. They show how the objective-setting principle becomes concrete. A baseline is useful because it gives later findings a reference point, but the reference point must be defined and documented.
It is equally important to distinguish percentage points from relative change. Moving from 25 per cent to 40 per cent is an increase of 15 percentage points, or 60 per cent relative to the initial value. Both descriptions concern the same arithmetic. Reporting one as the other changes the apparent size of the result and weakens the clarity that measurable objectives are meant to provide.
Before and After Does Not Establish Cause
Take the invented figures of 25 per cent before a campaign and 40 per cent afterwards. The observation establishes a difference between two measurements. It does not yet identify what proportion of that difference the campaign caused. News coverage, competitors, changes in the offer or changes in the sample could also matter.
The causal question requires a comparison with what would have happened without the intervention. That alternative outcome cannot be directly observed for the same people at the same moment. A research design therefore attempts to construct a credible comparison. Which design is suitable depends on the question and the conditions available.
This is distinct from adding another stage to ACCA. It concerns the evidence for an effect. DAGMAR specifies a desired change; the research design determines how convincingly its cause can be assessed. Defining a good objective helps prepare an investigation but does not replace the comparison needed for a causal claim.
The distinction should survive in reporting. An organization can truthfully say that its planned communication level was reached while qualifying the attribution of the change. Conversely, an estimated positive advertising contribution can coexist with failure to reach an ambitious target. Those are different results, not contradictory descriptions. Keeping them separate supports learning about both the campaign and the quality of the initial objective.
Digital Data and the Persistence of the Measurement Problem
Randall A. Lewis and Justin M. Rao examined 25 large field experiments involving digital advertising. Their 2015 Quarterly Journal of Economics article describes the difficulty of estimating advertising returns precisely and the selection problems affecting observational approaches (Lewis and Rao, 2015).
The finding does not require dismissing digital measurement. It requires distinguishing recorded events from the estimated additional contribution of an intervention. A system may count purchases following contact with an advertisement. That count does not automatically identify which purchases would have occurred without the contact.
Nor do additional decimal places settle the issue. Precision of presentation and strength of inference are different properties. Reporting should make clear whether the number concerns contacts, attributed events or an experimentally estimated incremental effect. A carefully limited claim can be more informative than a more impressive number whose meaning remains ambiguous.
DAGMAR’s continuing relevance lies in demanding a clear question before analysis. It does not lie in crediting Colley with tools developed decades later. New data sources expand opportunities for investigation, while the choice of what is being measured and the evidence for the conclusion remain substantive responsibilities. More observation can improve an answer, but it cannot by itself decide which question an advertiser ought to ask.
An Unchanged Total Does Not Establish an Unchanged Audience
Another methodological distinction concerns aggregate scores and individual paths. In an invented example, 40 of 100 people know a brand at the first measurement and 40 do at the second. The total does not establish that the same people retained awareness. Some may have forgotten while others learned the name.
Separate samples describe distributions at different times. Repeated observation of the same people can reveal individual changes but introduces its own participation and comparability requirements. The appropriate choice depends on the question. Neither design should be described as providing information it did not collect.
This matters for interpreting a hierarchy: percentages assigned to different stages do not themselves demonstrate that individuals moved through those stages in the order drawn. A claim about that path requires its own evidence.
The Sequence Remains a Proposition to Examine
Weilbacher’s 2001 critique challenges hierarchy models as general descriptions of how advertising works, particularly their account of interacting brand impressions and influences (Weilbacher, 2001). That argument should be treated as a scholarly position, not converted into a claim that measuring any communication response is pointless.
Chakravarty and Sarma also show that the model literature contains alternative sequences and added dimensions. Their review includes Ehrenberg’s connection of awareness, trial and reinforcement and later accounts allowing different response orders (Chakravarty and Sarma, 2022).
An analytical implication follows. A planning scheme can provide an intelligible order for questions without accurately describing every purchase process. Someone can try an offering before forming a strong favourable conviction. Experience of use can then alter attitudes. A diagram should not exclude such possibilities simply because its arrows suggest another order.
A historically careful application separates objective-setting discipline from belief in a universal staircase. It is possible to investigate understanding or awareness without asserting that every buyer moves through all stages at the same speed or in the same direction. That distinction preserves the practical value of asking explicit questions while leaving behavioural propositions open to examination. The model then supports inquiry rather than dictating in advance what every result must mean.
Later Editions and Retrospective Attribution
A revised second DAGMAR edition appeared under Solomon Dutka’s name in 1995. Library records explicitly identify it as a revision of Colley’s original 1961 work (Dutka, 1995). It belongs to the concept’s subsequent history, but should not automatically be treated as an unchanged reproduction of the first edition.
This affects more than dates. When a later textbook adopts an example or formulation, the relevant edition should be established before the material is assigned to the original author and year. Publication, revision and later application are separate historical events. A familiar name on the cover does not remove the need to distinguish them.
Schwarzkopf’s study of advertising historiography warns that broad interpretive assumptions, including modernization narratives, can shape historical accounts without being made explicit (Schwarzkopf, 2011). Applied here, the warning concerns an apparently inevitable march from vague advertising to perfect measurement. The record is more complicated than that progression suggests.
Colley’s contribution does not need the larger claim. It made an exacting demand visible: an advertising task should be described so that its accomplishment cannot simply be redefined after the outcome. How organizations interpreted and implemented that demand remains a further question, rather than something settled by the existence of the publication alone.
What Remains of DAGMAR
DAGMAR is a milestone because it connects the definition of an advertising objective with the evaluation of a result. Awareness, Comprehension, Conviction and Action illustrate different possible tasks. They should not obscure the larger requirement to decide beforehand which change matters, among which people and by what evidence its occurrence will be recognized.
The earlier history limits an exaggerated invention claim. Advertising psychology, consumer investigation and arguments about advertising’s value were established before 1961. Later research limits an exaggerated guarantee: a measured change is not automatically an incremental financial return, and an ordered diagram does not supply a complete explanation of behaviour.
Between those limits, Colley’s contribution remains substantial. Explicit objectives expose assumptions to examination. They help distinguish the work completed from the response observed and both from the causal contribution established. That is a durable basis for learning from advertising, even when the answer is less certain than a campaign report might initially suggest.
References
Belch, G. E. and Belch, M. A. (2003) Advertising and Promotion: An Integrated Marketing Communications Perspective. 6th edn. Boston: McGraw-Hill/Irwin. Used: pp. 206–208 of the electronic version carrying a 2003 copyright notice; library records also date the print edition to copyright 2004. https://hatemshiba.com/media/58ca7f11e8106.pdf.
Chakravarty, R. and Sarma, N. N. (2022) ‘Evolutionary framework of hierarchy of effects models: exploring relevance in the shifting of customer path’, Vilakshan – XIMB Journal of Management, 19(1), pp. 59–68. First online 2021. https://doi.org/10.1108/XJM-09-2020-0151. Open Access, CC BY 4.0: https://creativecommons.org/licenses/by/4.0/. Used for original summaries and interpretations, not unchanged reproduction.
Colley, R. H. (1961) Defining Advertising Goals for Measured Advertising Results. New York: Association of National Advertisers. Smithsonian Libraries bibliographic record: https://collections.si.edu/search/detail/edanmdm:siris_sil_383233. The first edition was not available for full inspection; the substantive account uses the scholarly sources identified in the text.
Dutka, S. (1995) DAGMAR: Defining Advertising Goals for Measured Advertising Results. 2nd edn. Lincolnwood, IL: NTC Business Books; New York: Association of National Advertisers. Used: bibliographic evidence concerning edition history, https://library.kaist.ac.kr/search/ctlgSearch/posesn/view.do?bibctrlno=428621&se=b0&ty=B.
Kinneer, P. M. (2001) ‘Fact or Friction: The Research Battle Behind Advertising’s Creative Revolution, 1958–1972’, paper presented at the AEJMC annual convention, History Division, Washington, DC, 5–8 August. Used: individual paper pp. 12–13, PDF pages 44–45 of the collected proceedings. https://files.eric.ed.gov/fulltext/ED459487.pdf.
Lavidge, R. J. and Steiner, G. A. (1961) ‘A Model for Predictive Measurements of Advertising Effectiveness’, Journal of Marketing, 25(6), pp. 59–62. https://doi.org/10.1177/002224296102500611. Used: publisher abstract and bibliographic information; sequence also checked through Chakravarty and Sarma (2022).
Lewis, R. A. and Rao, J. M. (2015) ‘The Unfavorable Economics of Measuring the Returns to Advertising’, The Quarterly Journal of Economics, 130(4), pp. 1941–1973. https://doi.org/10.1093/qje/qjv023. Used: publisher abstract.
Monash Business School (n.d. a) ‘DAGMAR’, Marketing Dictionary. https://www.monash.edu/business/marketing/marketing-dictionary/d/dagmar.
Monash Business School (n.d. b) ‘DAGMAR Approach’, Marketing Dictionary. https://www.monash.edu/business/marketing/marketing-dictionary/d/dagmar-approach.
Schwarzkopf, S. (2009) ‘Discovering the Consumer: Market Research, Product Innovation, and the Creation of Brand Loyalty in Britain and the United States in the Interwar Years’, Journal of Macromarketing, 29(1), pp. 8–20. https://doi.org/10.1177/0276146708327615. Used: publisher abstract; first online 2008.
Schwarzkopf, S. (2011) ‘The Subsiding Sizzle of Advertising History: Methodological and Theoretical Challenges in the Post Advertising Age’, Journal of Historical Research in Marketing, 3(4), pp. 528–548. https://doi.org/10.1108/17557501111183653. Used: abstract in Copenhagen Business School research portal, https://research.cbs.dk/en/publications/the-subsiding-sizzle-of-advertising-history-methodological-and-th/.
Schwarzkopf, S. (2015) ‘Value Measurement Systems, Professional Narratives and the (Un)Making of Market Regimes in Twentieth-Century American Advertising’, paper presented at the American Sociological Association annual meeting, Chicago. https://doi.org/10.13140/RG.2.1.4585.0960. Used: abstract in Copenhagen Business School research portal, https://research.cbs.dk/en/publications/value-measurement-systems-professional-narratives-and-the-unmakin.
Tadajewski, M. and Jones, D. G. B. (2016) ‘The history of marketing practice’, in Jones, D. G. B. and Tadajewski, M. (eds.) The Routledge Companion to Marketing History. Abingdon: Routledge, pp. 1–22. Used: published preview, particularly pp. 3–4. https://api.pageplace.de/preview/DT0400.9781134688685_A26553252/preview-9781134688685_A26553252.pdf.
Toda, Y. (2007) ‘A True Pioneer of Research in Advertising: The Contribution of Paul Cherington’s Works to Early Advertising Thoughts’, Proceedings of the Conference on Historical Analysis and Research in Marketing, 13, p. 294, extended abstract. https://ojs.library.carleton.ca/index.php/pcharm/article/view/1590/1434.
Weilbacher, W. M. (2001) ‘Point of View: Does Advertising Cause a “Hierarchy of Effects”?’, Journal of Advertising Research, 41(6), pp. 19–26. https://doi.org/10.2501/JAR-41-6-19-26. Consulted text: https://www.superbrand.net/Point%20of%20view_CauseHierachyOfEffects.pdf.
All online sources last checked on 22 October 2026.
