Change is on the Horizon for the Agency Business Model
Over recent decades, the advertising industry began to shift from traditional labor-based compensation models to more modern approaches focusing on selling “value” rather than billable time. These compensation structures are referred to as “value-led” models. There are several, but the most important to discuss are:
“Value-based” compensation
“Performance-based” compensation
“Output-based” compensation
“Product-based” compensation
Each model has its complexities and challenges, and the agencies that adopt them must redesign the rest of their business model to make them work. They also need clients who are willing to accept them. Therefore, they are not for every firm. Traditional labor-based compensation remains dominant, and most agencies remain designed to sell billable time.
But this is about to change.
Read on to learn the following:
Why the dominant agency business models are not sustainable.
How AI is reducing campaign costs by tens of thousands of dollars, and other impacts on agencies.
Why a product-based model is the next evolution of output- and deliverable-based models.
How the four value-led models compare in terms of adoption, benefits, and challenges, with several agency examples.
What “productization” of agency services actually means, and what agencies should know before considering it.
Pivotal moments impacting dominant agency models.
Marketers and agencies are both seeking change
According to the ANA’s 2022 Trends in Agency Compensation report, 82% of marketers still used labor-based compensation with their agencies. However, roughly half of marketers expected to change their compensation approach. The ANA suggests this is due to changing business conditions or resource needs.
Agencies are also looking for a change. Some began trying value-led models around the 90s and 2000s driven by the ability to measure digital marketing, clients seeking greater accountability, access to new technology, and economic pressures leading to tight budgets.
Fast forward to the present day. Agencies that still sell billable hours are finding themselves with a business model that is no longer sustainable. The model is designed around the outdated labor theory of value, defining the economic value of a good or service by the total amount of labor required to produce it, or, in agency terms, billable time.
The result is as follows:
Sales and account growth conversations have focused on selling chunks of labor hours (AOR and project fees) rather than scoped bodies of work or deliverables.
Clients and agencies often have different expectations for how much work an agency can deliver within a specific chunk of hours.
Agencies must bill more hours to make more money, which conflicts with client interests.
Agencies must hire more people to bill more hours to scale revenue.
On the client side, some marketing procurement practices redefine "value" as “the best price,” incentivizing buyers to aggressively reduce agency fees.
The result is as follows:
Declining agency margins.
Agencies can’t offer competitive salaries to hire and retain experienced talent, so they can't easily scale.
Staff experience burnout from taking on extra work where new hires should.
Creativity and innovation suffer.
Michael Farmer, CEO of Farmer & Company and author, brilliantly deconstructs many factors leading to our current situation in his book, Madison Avenue Manslaughter.
But our industry is about to become even more complicated.
Did someone say AI?
In addition to current agency challenges, the rapid acceleration of AI’s capabilities is threatening the billable hour model. For example, some PR firms already use AI to support tasks like generating creative content, conducting research, writing pitches, memos, and summaries, distributing press releases, and completing administrative duties.
The Wall Street Journal reported how agencies are using ChatGPT, Midjourney, DALL-E 2, and other AI tools to create ad campaigns, draft responses to RFPs, and save time and money.
Fernando Machado, Chief Marketing Officer of NotCo, shared the following about a recent ad campaign that used AI-generated images:
“…the cost savings were enormous. Typically, I would spend $50,000 to $100,000 to create an image for an ad campaign, but in this case, it was like between $5,000 to $10,000,”
In a recent interview with Contagious Magazine, Farmer shared the following:
“AI has come into being, which is like pouring gasoline on [the trends discussed in] Madison Avenue Manslaughter.”
“Madison Avenue Manslaughter said, ‘You’ve got to charge for the work, not the people,’ and I think that this is going to accelerate the need for people to convert from labor-based fees, to product-based fees.”
Farmer's latest book, Madison Avenue Makeover, covers how the agency, Huge, transformed its business from selling hours to a product-based model. A recent Mi3 article shared how Huge’s CEO, Matt Baxter feels about the current dominant model of selling time and why they adopted productization:
“…a doomsday scenario thanks to a killer combo for agencies of AI-led automation and margin killing procurement teams. More automation means fewer people that can be charged out on an hourly basis – and those that remain will be razored by corporate cutters. Instead, building advisory-based products and templates – or “productisation” – is one alternative Baxter says every traditional agency will have to consider to beat the machines and the number crunchers. Some already are – Dentsu and parts of WPP included.”
In September of 2022, the venture capital firm, Sequoia Capital, published the below chart illustrating a timeline for how we might expect to see fundamental AI models progress and the possible applications. They guessed at the possibilities for 2025 and beyond, but from what we’ve seen in the past year, they may have underestimated some of the milestones.
A timeline for how we might expect to see fundamental AI models progress and the possible applications.
As an example of how quickly generative AI has been advancing, Johan Leandersson, Creative Director at Supernatural, created the two images below exactly one year apart using the same prompt in midjourney: “Buzz Lightyear in the style of Gregory Crewdson.”
These two images were created exactly one year apart (2022-2023) using the same prompt in midjourney: “Buzz Lightyear in the style of Gregory Crewdson.”
“Things are moving fast,” to quote Supernatural’s Co-founder, Paul Caiozzo.
Fast, indeed. The recent article, AI is going to eliminate way more jobs than anyone realizes points out that “McKinsey estimated that robust large language models such as GPT-4 would be developed by 2027. But they are already here. And seemingly overnight, OpenAI's generative AI was integrated into Microsoft products, and in the span of a few months, corporate giants including Amazon, AT&T, Salesforce, and Cisco have rushed to incorporate enterprise-grade AI tools. McKinsey's latest report predicted that around 2030 and 2060, half of today's work tasks would be automated. Their best guess as to when this will happen — 2045 — is almost a decade earlier than previously estimated. Things are changing fast. And as adoption picks up, so will the downstream effects of the technology.”
Why product-based fees?
As Farmer suggested in his Contagious interview, there is an accelerating need for agencies to convert from labor-based fees to product-based fees. But why "product-based" over other models such as "value-," "performance-," or "output-based"? We can answer this question by understanding how agencies and marketers have adopted these other options and their benefits or challenges.
The rise, fall, and RISE of value-led models
“Value-led” is a term I’ve borrowed from my friend and colleague, Tim Williams, founding partner of Ignition Consulting Group. For the past two decades, Tim has worked with agencies around the globe to transform their approach to pricing using value-led pricing structures. I’ve listed the four that I believe have been the most popular, but there are other approaches I haven’t listed.
Let’s explore the adoption rates, benefits, and challenges of these four pricing structures:
“Value-based” compensation
“Performance-based” compensation
“Output-based” compensation
“Product-based” compensation
1. Value-based
What "value-based" means: Agency fees are negotiated based on an estimated value of the work or outcomes the firm will deliver.
An example: In 2014, AdAge reported that Coca-Cola revamped its compensation system to pay agencies more for work in newer, groundbreaking areas. Coca-Cola defined “value” based on such things as how strategically important the work is to the brand, how strategically important the brand is to the company, and "industry dynamics," defined as, "Is this agency uniquely qualified to deliver this work?
Sarah Armstrong, Coca-Cola's, director of worldwide agency operations, explained that "To be unique in this industry is to be able to charge a premium for what you're offering. If you're not unique, we're not saying we're going to go move to another agency. …What we're saying is the market will set what we need to invest in this work."
Adoption: The ANA reported that in 2022, 11% of marketers used value-based compensation with their agencies, down from 20% in 2016. Lindsay Slaby, founder of the marketing strategy consultancy Sunday Dinner, recently shared feedback from her marketing clients that provides insight into one reason adoption could be falling:
“‘Value-Based Pricing’ is a dangerous term with CFOs,” she writes. There are different ideas between agencies and marketers about what value-based pricing means. Agencies often consider value-based pricing to mean “pricing is based on the perceived value of what I am delivering and what people are willing to pay.”
This definition is not uncommon. Hubspot’s article, Everything You Need to Know about Value-Based Pricing shares more on this approach. But it may not be right for every scenario or play well with marketers if not carefully executed. Slaby advises:
“Taking this to a CFO without a breakdown and strong metrics - saying this is simply the agency value does not go over well.”
CFOs and CMOs define value-based pricing as a price based on “the value that our work can drive for the client. You base your project price on a percent of the value you bring to your client’s business.”
But this is challenging to do, as she explains:
“Unless you are very deeply embedded, it will be hard to get shared KPIs and monetary value of what they are hiring you to create. The upside, yes. Roadblocks, many.”
The comments on the post are also worth a read, here.
Performance-based
What performance-based means: Agency compensation is based on client success metrics, often used with labor-based fees or media buying commissions.
An example: In 2015, Airbnb's CMO, Jonathan Mildenhall, shared that they pay their agencies on a “per night” basis, where the agencies take a cut every time a customer books accommodation through their site. He was concerned that the traditional agency compensation model was “squeezing the life out of creativity,” as marketing budgets were increasingly restricted.
Adoption: Performance-based models had peak adoption in 2013 at 61%, but usage of the model has since declined to 41% in 2022, its lowest rate since 2003. The ANA suggests this may be due to the complexity and effort required to structure an effective performance agreement. The model tends to be more common with larger advertisers. Smaller advertisers (defined by ad spend under $200MM) may be unable to invest the time to develop the incentive agreement, may not have the proper measurement tools, or their budgets may be too small to allow for enough upside.
Smaller marketers also want predictable budgeting and to know they will pay a fixed amount.
Output-based (or deliverable-based)
Output-based means: An agency uses fixed pricing for specific outputs from their work or deliverables they create.
An example: The London-based firm, Hard Numbers, is a performance-driven marketing and communications consultancy. They deliver media relations campaigns that drive pipeline, sales, and company growth. They price work based on the type and number of deliverables, such as the media/speaking opportunities, leads, and meetings they will generate.
From their site:
Hard Numbers was set-up to be different from the hundreds of me-too agencies with exactly the same business model that have been set-up in the last ten years. No member of our full-time team has ever or will ever fill out a time sheet.
We do not sell time. We sell results.
Adoption: The deliverable or output-based model has been increasing among larger advertisers. The ANA shared that 53% of the largest advertisers employ this method, up 5% from 2016.
Many agencies now work on an output or deliverable basis, but they still price based on an estimated number of hours to complete the work, a blended hourly rate, and an added percentage as a buffer and margin. The approach is called cost-plus pricing (or “markup”).
The primary benefits are:
Alignment on value: Agencies and marketers negotiate a price for deliverables, not for hours.
Margin potential: If an agency can quickly complete a project, it can retain more margin.
Opportunity for automation: Faster delivery is in everyone’s interest. If an agency can automate aspects of its work, it can become more valuable to its clients and more competitive in the market.
The challenges of an output-based model include the following:
Constantly reinventing the wheel. Each new assignment has the potential to be very different and needs to be scoped, estimated, negotiated, scheduled, staffed, and delivered within budget. There is a heavy staff load to manage custom engagements.
Risk to the agency. The agency takes on more risk. If the agency takes longer than planned to deliver the project or requires more resources, the agency covers the costs. Accurate scoping is critical to the agency’s profitability. And each time the agency reinvents the wheel, it must address any new 'unknowns' which introduce more risk to its margin.
Emphasis on efficiency. The quicker the agency can complete a project, the more margin it will retain. This could potentially cause agency staff to focus more on time and efficiency than the quality of work or whether clients achieve their outcomes.
Limited flexibility. If the agency needs to deviate from the scope of the agreed-upon output, it may need to renegotiate the agreement, which can impact the agency/client relationship.
Productized
Productized: An agency uses fixed pricing for standardized (productized) packages of service and deliverables.
An example: XenoPsi Ventures owns an agency that works with direct-to-consumer e-commerce brands. They share a philosophy that brand growth falls into three phases, so XenoPsi packaged different combinations of the agency’s services into “products” each named to reflect the outcome it will deliver to the brand in each phase. The products are Launch, Traction, and Scale.
MichaelAaron Flicker, founder and President of XenoPsi Ventures, explains how the model gives his firm control over its business, opens new revenue streams, and builds client trust.
“One of the ways my team and I flipped the script on this race to the bottom was to “productize” what we do. This crystallized the vast array of processes that go into the art of creating, producing, and distributing creative ideas into tangible, explainable, and demonstrable products that deliver proven outcomes.
The benefits of transforming our services into products, besides eliminating haggling over compensation and poisoning what should be collaborative cooperation, will be realized by both sides.
The brand can have confidence they are buying a “center of excellence” that will deliver value because the agency has successfully applied it multiple times before.
From the agency’s perspective, revenues will grow from selling a product repeatedly rather than reinventing the wheel for each new assignment. And because of that, unlike under the traditional ad agency model, agencies would not have to hire the heavy staff load required to accommodate executing the work for new project-based assignments. This lets agencies better control their costs and do better work while guaranteeing clients an outcome at a fixed price.”
There are challenges with this model, too. To explain them, I first need to provide a more thorough definition of “product” and “productization.”
Productizing agency services explained
What does “productizing” agency services mean?
“Productization” is the art of packaging, pricing, and promoting intellectual capital as a product.
What do we mean by “product?”
A “product” is a solution to a problem or fulfillment of a need for a specific audience delivered in a reliable, repeatable, and scalable manner.
Let’s break down this definition further:
“A solution to a problem or fulfillment of a need”
The specific problem or need should be well-aligned with the agency’s core competencies—the reasons clients hire the firm.“A specific audience”
The audience should be a type of industry category, brand, audience group, or marketing role the agency knows best.“Delivered in a reliable, repeatable, and scalable manner”
The agency’s delivery method should be proven and standardized, such as a repeatable framework and process, a baseline for scope and pricing options, known roles and responsibilities that will be assigned, and a predictable timeline.
Think of the word “product” as a container for the value an agency produces from its expertise, experience, partners, proven process, etc. A product can be priced and sold or packaged with multiple products to create something larger such as a “solution” or “program,” or pick any other container word.
Agencies have an opportunity to apply their creativity to how they package and promote their value. This is another tool to further differentiate themselves.
Differentiating products
Products can fall into three categories:
Service-based. An example is a strategy or discovery package, such as Hyperact’s set of offerings, or a grouped set of deliverables, such as Wray Ward’s marketing solutions.
Tech-based. An example is a service using proprietary technology, such as a custom-built AI tool, like Supernatural’s "The Machine,” or an ad-tech tool, such as Empower’s ClearTrade.
Subscription-based. An example is the recurring delivery or access to a service-based or tech-enabled offering, such as Current Resident’s Perfect Machine for unlimited post-production editing.
“Product-based” is the evolution of “output-based”
As we heard earlier from XenoPsi Ventures President, MichaelAaron Flicker, agencies can experience many benefits from a product-based model. I’ve captured them and more here.
Benefits of a productized model
Financial stability and predictability:
Sales conversations can focus on the value of your program, products, or outputs, not the hours that go into them.
Revenue can scale from product pricing rather than hiring more staff.
Forecasting, budgeting, and financial planning become more predictable.
Connecting programs and product options make it easier to upsell and increase the lifetime value of accounts.
Reduced costs and faster delivery:
Repeatable offerings with standardized workflows results in clearer roles, predictable resourcing, and reduced costs for project management, estimating, time-tracking, and reporting.
Agencies can adopt what I call “creative product team” structures for greater innovation, work quality, and faster delivery.
Stronger client relationships and satisfaction:
Clients are clear about what they are buying (the problem it solves and the scope to do it), eliminating misaligned expectations.
Agency/client interests become aligned—the model incentivizes value delivery and speed, not hours worked.
Clients experience a simple, transparent, predictable, and lower-risk client experience.
Competitive advantage and differentiation:
Packaging intellectual capital as a product communicates a clear and confident perspective about the impact an agency’s work delivers.
Productized offerings make it difficult for procurement practitioners to compare an agency against competitors.
Challenges shifting to a productized model
As I mentioned at the beginning of this post, when adopting a productized model, agencies must be prepared to redesign the rest of their business model to make them work. Other challenges agencies will need to prepare for include:
Needing a new way to manage the agency without billable time: Without a business model based on billable utilization, agencies must learn how to manage the agency and their engagements with different KPIs.
Addressing the perception that productization restricts creativity: Some may initially think standardizing and productizing services will diminish a firm’s ability to deliver custom thinking or creativity. But this is not true. Agencies can design their products to allow for more or less flexibility. They are a “container” or proven framework for solving business problems, and the walls of those containers provide protection for a team to do their best work.
Introducing change without disrupting the business: Agencies must figure out how to shift into the model with minimum disruption to their existing business. There’s not a one-size-fits-all approach to do this, but there are a common set of factors to consider to determine the best path forward. Some of these factors are:
Whether the agency already has a strong strategic positioning to inform the product strategy
The number of existing core services and size of client engagements
The size of the firm and the number of people available to lead or support the transformation
The existing team structures and organizational agility
Learning how to sell products to procurement: As an agency grows and begins working with larger brands, the firm will likely experience procurement pressure to switch back to billable hours “just for this one client.” It’s a slippery slope.
What does this all mean?
The truth is that agencies have always produced products. But because the dominant agency business model was developed around the labor theory of value, agencies have monetized the wrong thing—the hours to create them rather than the products themselves.
In the book, It Doesn’t Have To Be Crazy at Work, Jason Fried and David Heinemeier Hansson, founders of 37Signals and makers of Basecamp software, share how they view their entire company as a product:
“…when you think of the company as a product, you ask different questions: Do people who work here know how to use the company? Is it simple? Complex? Is it obvious how it works? What’s fast about it? What’s slow about it? Are there bugs? What’s broken that we can fix quickly and what’s going to take a long time?”
Just as 37Signals improves their software through iteration, they iterate on their company too. Our industry has experienced iteration as well, but significant changes need to happen at a much more rapid pace. MichealAaron Flicker nailed it when we said, “If we don’t change the conversation from haggling over costs to building synergistic partnerships that provide value, we may not have much of an industry in the next decade that is worth saving.”
