Marketing effectiveness · Incrementality · Budget decisions
A channel can receive credit for sales it did not cause.
Arcus helps leaders connect marketing investment with customer behaviour, incremental contribution and defensible resource decisions. A useful review distinguishes reporting attribution from causal impact, and separates lower purchasing costs from better market outcomes.
The decision before the program
01 / What would customers have done without the spending?
02 / Does the additional contribution justify the full cost?
03 / Which evidence would support scaling, redesigning or stopping?
Begin with the decision the report must support
An illustrative campaign reports strong revenue from existing customers who clicked an advertisement before buying. Management proposes increasing the budget. Yet those customers may already have intended to purchase, and the report may credit the ad for demand generated elsewhere. The dashboard can be accurate under its attribution rules while remaining insufficient for the investment decision.
Arcus’s central thesis is that marketing effectiveness should be assessed against the behaviour and contribution the investment changes. Reporting which channel appeared in a customer journey is useful for operations; it does not by itself establish what would happen if that channel received more or less money.
Keep three questions separate. Are we purchasing efficiently? Are we influencing customer behaviour? Is that influence worth more than its cost? A supplier saving can improve efficiency without proving stronger demand. A rise in purchase intention can support a positioning decision without establishing realized sales.
Four explanations for apparently strong marketing results
Use these competing explanations to decide what further evidence is needed.
01 / Incremental demand
The activity may bring forward a decision, reach a new customer or change a purchase choice. Examine which behaviour changed, for whom and over what period. Distinguish genuinely additional demand from a shift in timing.
02 / Selection and existing intent
People likely to buy may also be more likely to see, seek or click the marketing. A high conversion rate can reflect the audience selected rather than a large effect of the activity.
03 / Credit moves between channels
A paid interaction may replace an organic visit or receive credit for demand created by other activity. Review total outcomes and channel substitution, not only the credited revenue within one reporting system.
04 / Other changes drive the result
Pricing, distribution, seasonality, competitor actions or product availability may explain part of the change. A before-and-after comparison needs a credible account of these influences before attributing improvement to marketing.
What large-scale field evidence contributes
Blake, Nosko and Tadelis’s research on paid-search effectiveness at eBay, published in Econometrica in 2015 uses field experiments to distinguish causal effects from conventional estimates. The study found that effects differed by customer usage and that conventional measurement could substantially overstate returns. Its historical eBay context does not establish that branded search or another channel is ineffective for every business.
The management implication is to investigate the counterfactual: what customers would do without the activity. Arcus’s interpretation is that evidence should be matched to the decision and its stakes. A small business with few transactions may not be able to resolve a modest effect quickly; an inconclusive test should not be relabelled as proof of no effect.
Experiments also require design judgement. Audience overlap, geographic spillovers, changing offers and an inappropriate observation period can undermine interpretation. Where a credible test is impractical, combine available evidence with explicit uncertainty and a proportionate commitment rather than presenting observational attribution as certainty.
The Arcus spend-to-contribution review
This original management diagnostic structures a budget decision. It is not a guaranteed attribution model or a substitute for appropriate statistical design.
01 / Decision
Specify the spend change under consideration, customer group and intended behaviour. Separate a decision about acquiring new customers from one about retaining existing customers or improving brand understanding.
02 / Counterfactual
Explain what would probably happen without the proposed activity. Identify a feasible comparison, experiment or other evidence and its limitations. Record alternative explanations and possible substitution between channels.
03 / Contribution
Translate additional outcomes into contribution after variable delivery costs, returns, discounts and relevant support costs. Include campaign, production, agency and measurement costs as appropriate. Avoid substituting gross revenue for value.
04 / Commitment
Agree what result would support increasing, maintaining, redesigning or stopping spend. Consider uncertainty, capacity and reversibility. Evaluate the next spending increment rather than assuming the historical average return continues unchanged.
Worked example: attributed return versus incremental value
Suppose a fictional campaign costs $20,000 and receives credit for $100,000 in sales. A dashboard reports a five-times revenue-to-ad-spend ratio. At a 40% contribution margin before campaign cost, the credited sales represent $40,000 in contribution—but only if all those sales are additional, which attribution alone does not establish.
If a credible comparison instead estimates $30,000 in incremental sales, the contribution is $12,000. After the $20,000 campaign cost, the immediate result is negative $8,000 before any other relevant costs. Future repeat purchases could change the conclusion, but require evidence about retention, margins and timing rather than an assumed lifetime-value multiplier.
At that 40% margin, the campaign needs $50,000 in additional sales to cover its $20,000 cost, before other effects. These illustrative figures show the decision boundary. Compare a lower-cost design, a different audience or an alternative use of budget; do not simply optimize the credited ratio.
Compare practical improvement routes
The response should follow the constraint identified by the evidence.
Improve purchasing efficiency
Review supplier scope, rates, procurement and duplication. Can reduce cost without changing the customer proposition. Protect quality and capability; a cheaper input is not evidence of stronger incremental demand.
Improve customer and offer fit
Research the benefit barrier, segment, proposition or buying process. Can address weak conversion that media changes cannot solve. Requires evidence of actual customer needs and operational ability to deliver the promise.
Reallocate and test spending
Compare audiences, channels and spending levels with a credible learning plan. Requires sufficient data, suitable comparisons and time. Budget changes can affect other channels, so examine total business outcomes.
Strengthen measurement and governance
Reconcile definitions, costs, customer records and decision ownership. Can make choices more explainable. Additional reporting is useful only if it changes a decision; avoid building an elaborate dashboard around untested attribution assumptions.
Assess the whole economics and operating effect
Include the cost of creating, delivering and measuring activity, not just media. Check whether additional demand can be fulfilled at the assumed margin and service level. Capacity shortages, discounting or returns can turn a promising acquisition result into a poor business outcome.
Distinguish short-term response from longer-term brand or customer effects. Use measures and review periods suited to the objective, with clear limits on what can be inferred. Brand awareness, purchase intention, leads, orders and profit are different outcomes; they should not be presented interchangeably.
Agree data access, customer privacy arrangements and responsibilities before collecting or linking information. Assign an owner for the budget decision and a reviewer for material assumptions. A practical measurement plan should specify the action that follows a result and what happens when uncertainty remains.
How Arcus supports marketing effectiveness
An engagement can focus on a particular investment or a broader review of marketing allocation and capability.
01 / Diagnose the evidence
A review of objectives, reporting definitions, spend, customer behaviour and available comparisons. Findings distinguish purchasing efficiency, attributed activity, incremental effects and unresolved data gaps.
02 / Investigate the mechanism
Customer research, benefit-barrier analysis and a proportionate measurement or testing plan. The work establishes which alternative explanations matter for the decision.
03 / Compare budget options
Contribution and cost scenarios, supplier or program review and recommendations for allocation, redesign or further testing. The decision record makes assumptions and uncertainty explicit.
04 / Implement and learn
An action roadmap, ownership, reporting requirements and review milestones. Capability support helps the team use the evidence in subsequent decisions rather than commission a one-off report.
Historical examples from Arcus’s published work
Arcus’s earlier page reports a client saving over $2 million following a vendor audit and competitive bidding, and a 40% first-year saving associated with a VSM recommendation. These are retained as historical reported examples of purchasing and supply-management work, not current benchmarks, independently revalidated results or promised savings.
The earlier page also describes stopping a sponsorship program and reallocating investment, and a food-brand positioning assignment reporting a 40% increase in purchase intention among a younger group. Purchase intention is not realized revenue, and the published summary does not establish a causal revenue effect. These examples illustrate different decision questions rather than a common guaranteed return.
The published historical interview with Anthony Lacavera of Globalive provides additional leadership context. Attribution reflects the role identified in the original interview.
Discuss the spending decision you need to make
Tell Arcus which investment is under review, what the current reporting shows and why the decision is uncertain. We can scope a focused effectiveness review, customer research or broader marketing strategy assignment. Please begin with the business question rather than sending identifiable customer records.
Related services include market research and capability building. View experience through the central consultant directory.