Happy Sunday, Everyone!
I hope you’re all enjoying the first weekend in August – it’s difficult to believe that summer is winding down and Q4 is less than 60 days away, but alas, that’s where we find ourselves.
Since asking about topics to cover, seven readers have responded with questions about brand-agency engagements – what to look for in an agency, red flags to avoid, etc. In an effort to be both responsive and efficient, I’ve tried to answer each reader's question, while also providing a broader framework for how I think about agency relationships.
To start big picture: I’ve been on just about every side of the brand-agency engagement: I’ve worked for real estate firms and managed parts of that relationship; I currently run an agency; I am one of the managing partners at a venture fund that invests in early (pre-seed to series A) companies, some with agency relationships; I’ve helped many of our agency clients conduct reviews of their agencies, and I’ve worked with PE and VC funds while they’ve evaluated agency proposals. Long story short: I’ve seen a lot of these relationships, from many angles.
Let me start with my central premises regarding agency relationships:
1. There’s good and bad
Not all agencies are good. Not all agencies are bad. While this issue will focus on the bad (so you can hopefully avoid it), I want to be clear that many agencies I’ve encountered and many of the agency people I know are genuinely good, caring, diligent and smart. There are also 100% people and agencies I advise clients to avoid. Every tree produces some bad apples, and some apple trees are rotten to the core. It doesn’t follow from those two observations that all apple trees are bad or destined to be rotten – if that were the case, we wouldn’t have apple pie (and I’m quite fond of apple pie).
2. They often make financial sense
For brands in specific situations, it is often cost-advantageous to use an agency. At their core, agencies are selling four things:
a. Expertise - agencies (should) have teams of skilled, trained people that clients can “rent” at a substantial discount to what hiring that team internally would cost. Expertise should be the baseline expectation when hiring an agency.
b. Experience + Data - virtually every agency works across multiple clients; that gives the agency a broader perspective + more data (more ad accounts, more analytics accounts, more tests, etc.). The ability to synthesize this data, identify patterns + operationalize those patterns for clients is worth more than talent alone, because it saves time.
c. An Ecosystem - by virtue of working with many clients, agencies naturally develop an ecosystem: SaaS partnerships, workflows, integrations, etc. that can be leveraged by each of their clients. This is a massive value add to a brand – especially when doing so avoids the pitfalls + costs of poor selection.
d. Acceleration - Ultimately, any agency worth its fee enables clients to go farther, faster via a combination of superior talent (“expertise”), superior data and the combination of experience, vetted partnerships, training + existing assets (“ecosystem”).
Those four things each have massive financial benefits to clients – especially clients that don’t have the budget to spend $1M+ per year on a marketing team (which is barely enough these days to cover a Director of Marketing, 2x Creatives, a media buyer + a marketing coordinator).
3. Agencies, if used strategically, augment teams + brands
Finally, and (sometimes most critically) agencies can augment in-house teams – providing new, fresh ideas/perspectives and enabling brands to gain access to world-class marketing talent that is otherwise unavailable to them, while retaining the efficiencies of in-house talent for other functions.
4. De-Risk Hiring
Hiring an agency (or freelancer) doesn’t just bring efficiency, talent and acceleration; it also reduces risk. Most brands hiring for marketing talent don’t know what they’re looking for (or what to avoid); this isn’t their fault - it’s par for the course anyone non-technical plays when hiring for a technical role. The simple reality is that most marketing interviews are conducted in sentences & soundbites, while all marketing initiatives take place in novels. It’s deceptively easy to sound like you know what you’re doing over the course of 1, 2 or 3 interviews; it’s much more difficult to actually execute a successful series of marketing initiatives over the course of 3, 6, 12, 18, 24+ months. I’ve shared multiple times how a single bad marketing hire can cost a brand $1M+ - and I stand by that. I’ve seen it happen first-hand. Agencies, by contrast, are much easier (and much cheaper) to replace.
5. Focus on the Core + Win-Win-Win Outcomes
I’ve always been a proponent of brands focusing on their “core” and outsourcing the rest. Agencies (along with law firms, accounting firms, etc.) enable just about any brand to do that - and it makes sense. The brand gets specialized expertise that would otherwise be inaccessible, the agency is able to charge a premium, and the people who work for the agency get to work for a broader, more diverse group of clients than would be possible if they went in-house – accelerating their learning + career development. When these relationships are done right, everyone involved wins.
But this isn’t an article on why agencies can make sense; it’s an article about where things go wrong (and how to avoid it) - so let’s get into that. However, I find any conversation about agencies and their flaws necessarily takes a “all agencies suck” detour, and it’s my sincere hope the above helps to balance what follows. On the whole, I believe in agencies (and not just because I run one) - I advise the companies I’ve invested in to bring on agencies consistently.
The six points below are all taken from my actual experiences over the past 4 years (with several of these ongoing situations I’m working with clients/brands to navigate):
Issue #1: Contracting + Engagement Structure
Every agency agreement should have a simple, no-fault termination provision with a reasonable “out” for the brand. I’ve encountered two of the most predatory agency contracts I’ve seen in 15+ years in the past three months: both with no termination provision (so the brand was forced to stay with the agency for the full Term of the agreement), along with penalties that prohibited the brand from disassociating with the Agency.
To be perfectly clear: that’s bullshit. If any contract (agency, law firm, CPA, SaaS, whatever) does not include a reasonable termination fee, don’t sign it. What constitutes “reasonable” is largely a function of the engagement, but I tend to think, for an agency/freelancer engagement, 30-120 days is reasonable (again, depending on the work being done and the resources committed).
If an agency is acting as an outsourced marketing department, fulfilling 3-5+ roles (marketing lead, email/SMS, creative, media buying, etc.), then I’d expect a longer termination period. The agency is taking on more risk (allocating that much talent to a single client that would need to be reallocated in the event of termination), and a smooth transition back to the brand is likely to take significant time. On the flip side, if the agency is just doing one function - media buying, PR, creative, etc., then a shorter termination period is likely fine.
There are two reasons for this, both eminently practical: (1) in the event a contract is terminated, there’s a need to transition the work being done to a new partner (or back to an in-house person/team) and (2) the agency needs sufficient time to plan/re-allocate resources.
It’s for these reasons that I’m also against any “same-day” or “no notice” termination periods.
From the brand perspective, when I see a no-notice termination period, I always encourage the leadership to ask how the agency is able to afford that: if an agency is properly staffing each client account, then why is the agency willing to have an unknown period of time (a week, a month, a quarter) where their people are now not optimally booked? If the answer is that those resources can be shifted to other accounts, why aren’t they now? Are they under-serving other accounts, or overworking current employees? Neither one is good.
Some other things to look out for in contracts:
On Future Work: every agency agreement should be clearly scoped, with no guarantees for future work. One agreement I’m helping a brand navigate included a provision that all future work belonged to the agency at their then-current pricing (yes, you read that right: any future marketing, web development, advertising or communications work the brand needed, for any current or future assets, had to go to the agency). That’s categorically insane. Every agency agreement should be clearly scoped and include no future guarantees.
On Guarantees: no agency can guarantee performance. I can’t promise what your ROAS will be, or your CPA, or your CTR or CVR. If I was able to predict those things, I’d take my talents to Wall St and I’d own an island (or two). If you see “guaranteed 5.0 ROAS” - run.
On Agile + Sprints: I hate agile for most agency work - it’s a fancy way to screw clients out of money. If you’re building a website, you don’t need sprints – most of what you’re building isn’t going to evolve mid-project. Any decent agency should be able to clearly scope a website build or a branding process or a market research project without needing to resort to “sprints”. If you’re getting into app development or something more complex, sprints become more reasonable as the entire product can shift quite suddenly + rapidly.
Quick horror story: I was brought in mid-project for an enterprise website redesign. The agency involved had already been paid nearly $1M and promised (via many fancy decks) a shiny new website. However, after 10 sprints, the site wasn’t functional – and the agency came back to the brand requesting another $600k+ for the additional sprints necessary to finish the job and deliver what was initially promised.
On Compensation Structures: I know there are strong feelings throughout marketing-land about this, and while I have my thoughts, the bottom line is this: just about any compensation structure can work, provided that everyone is clear about it. I tend to prefer structures that protect both parties and accurately reflect the level of effort required. For instance, I don’t love the percentage of spend model because the level of effort/skill doesn’t always scale in the same way as spend. Likewise, performance-based compensation sounds great, right up until either (a) the agency crushes it, becomes too expensive + gets fired, (b) the agency does all kinds of weird things to take credit for non-incremental sales/leads and/or (c) a breakdown later in the value chain (i.e. sales team) results in conflicts everywhere.
The bottom line is that there’s no perfect engagement structure. My golden rule for evaluating any comp structure is whether or not both parties are happy with it when things are good or bad. If things go to complete shit, what’s your obligation? If things go fantastically well, what are you committed to paying? If you’re unhappy with the answer to either, don’t sign.
Issue #2: Ownership of Assets
I’m going to cut to the chase on this one: under no circumstances should an agency have ownership over any paid-for asset produced, created or worked on for a client. The same contract I referenced above also stipulated that the agency owned:
- The ad accounts
- Website Analytics + pixels
- Website itself
- All materials created
This is absolute garbage. It’s what sham agencies do. And it gives every reputable agency heartburn, because the shops that do this indirectly harm the entire agency ecosystem. It is never ethical or acceptable for an agency to claim that they own your ad account, or the content of your website, or your website itself.
If any agency makes any claim over any of these assets at any time, run. I will only caveat this with a payment condition (which is similar to what you’d get anywhere else: if you haven’t paid the agency for the materials, they don’t belong to you yet). I think that’s pretty fair to both sides.
Issue #3: The Team // Learning on the Brand’s Dime
To be blunt on this one, I have a particular loathing for situations where an agency staffs a client account with junior-level talent. Is it good (and even advisable) to add junior staffers to accounts in order to help them gain experience? Absolutely. But every account should be led and managed by someone with real experience and expertise. Full stop.
One of the core values of an agency is the caliber, talent and experience/expertise of its people. If an agency isn’t putting people on your account who are better than what can be hired on the open market, then one of the central rationales for hiring the agency goes out the window.
I can't count the number of times I’ve seen an agency pitch a client on a specific team, only to switch them out for junior people. It’s not acceptable. Every one of our clients has my cell number and/or email. We have three senior officers, and one of them oversees and checks in on each account – along with each account being led by a director-level (or above) team member (often more than one). We hire junior people infrequently, and when we do, we spend an inordinate amount of time training and mentoring them.
If an agency isn’t willing to commit to putting a particular team on your account, or switches out the team you were pitched with another team absent good cause (this including: members of that team left, the work your account required materially diverged from what you initially requested, etc.), that’s a big red flag.
Issue #4: Testing Everything
This one is controversial, but hear me out: any agency that says “test everything” or “every account/business is different, so we should test everything to find out what works in your niche” is either incompetent or dishonest.
One of the central tenets for hiring an agency is that you’re hiring the expertise, experience and insights gleaned from dozens, hundreds or even thousands of client engagements. You’re getting the accumulated knowledge of hundreds (or thousands, or tens of thousands) of ad accounts and campaigns.
Why in the actual **** would you hire people with this kind of knowledge, experience and expertise, then tell them to start from square zero?
Any agency should have a set of better practices (not “best” practices) that they start from – that’s the benefit of accumulated knowledge. After all, you’re bringing in an agency to accelerate your growth, not to spend time testing things they should already have a decent perspective on. Will those “better practices” always work? No. But that’s why you’ve hired smart, talented and experienced people: to identify the edge cases where these practices don’t hold and intervene before you blow through a ton of resources.
Finally, testing is expensive: it takes resources (people to design the test), time (to execute it) and money (ad dollars, creative fees, etc.). You want a partner that’s going to help you test smarter, not test harder.
My go-to question: what’s your default setup for accounts like ours? Why? What would you recommend we test, and how should we do it?
Any good agency has a clear POV on this. You may or may not agree with it – but at least that starts the conversation in a productive direction. If the agency gives you the “we test everything” answer to this question, that’s a blood red flag. As for testing methodology, I’ve shared my thoughts here and here.
Issue #5: Training, Evolution + Adaptation
One of the more insidious issues I’ve found in agencies (though a lot of reviews, audits, interviews and experiences) is what I’ve come to call “appealing rigidity” – which is almost always characterized by a refusal to deviate from antiquated principles/approaches.
This is particularly common in vertical-specific agencies, characterized by some (or all) of the following:
- An outdated, cookie cutter approach to every client account
- Proprietary “keyword sets” or “data models” or “optimization methodologies”
- Claims about “knowing how to X” or “Insider knowledge about Y”
- A small set of templates used for websites, printed materials, etc.
- Common, specific-length articles for “SEO”
- Pre-built email/SMS workflows with just the brand + colors changed out
- A penchant for “growth hacking” or “exploiting loopholes” in Google/Meta/whatever
These agencies do the same things in 2024 that they did in 2017 – despite the macro environment and underlying technology being radically different – then have the gall to claim that it’s to the brand’s (their client’s) benefit (spoiler alert: it isn’t).
While you want an agency with a clear PoV, that POV must be tempered with a willingness to adapt and evolve. The easiest way to identify this: ask the agency how their “better practices” have changed over the last 24-48 months. They should be able to clearly articulate at least 4-5 changes, as well as the data/insights/platform changes that gave rise to them.
The same goes for their people: ask about training for their team members. How does their team stay on top of trends, whether it’s in design, digital, email, SMS, brand, whatever? What resources do they have in place to support the ongoing development of their team? What industry events (in-person or virtual) do they attend?
In this day and age, there’s zero excuse for an agency to not attend a 100% virtual, 100% free event like SMX Next – and if they don’t, why not? The recordings are available on-demand (so you don't need to take time from standard working hours), and a member of their team could easily attend + summarize talks to maximize efficiency. If the agency doesn’t have a clear structure for how they help their people grow + learn, odds are that they aren’t – which means your account could be staffed with people who aren’t as current as they could be. That’s not immediately disqualifying in all marketing disciplines (for instance, does your PR person need to attend PRSA ICON? Probably not. But they should probably talk about how tools like SparkToro inform their outreach strategy).
Automation & Adaptation: Finally, it’s always worthwhile to ask (especially of late) how the agency is using generative AI / ML in their workflows – what have they changed since these technologies have come onto the scene? Where do they categorically refuse to use them? What do they see as the strengths + weaknesses of these technologies? How do they ensure that anything automated is QA’d?
Issue #6: The Work
While I’ll do a separate issue on actual ad account red flags / issues / questions, I want to close with the work itself. Your agency should have a clear POV on how they operate – when/how they make changes, their creative process, etc.
In my mind, this comes down to four primary buckets:
- Data - how does the agency approach data? What role(s) do they see it playing in the management of your account? What specific actions do they take with it? Big picture: data is the single-most-impactful optimization lever for any ad account – and if an agency isn’t doing things like offline conversion import, that’s a clear yellow (arguably red) flag.
- Accountability - while agencies should never guarantee results, they should be able to produce defensible forecasts and budget models that provide a likely range of outcomes based on reasonable assumptions. Forecasts create accountability for all parties - especially when they are used to guide testing + optimization.
- Metrics + KPIs - I’m on record (at many, many events) saying that: (1) contribution margin + contribution dollars should be north-star metrics for most brands and (2) every metric can (and will) be abused if it is not paired with 2 other “balancing” metrics (balance rates like ROAS with absolutes like total contribution dollars + net new customers acquired).
- Approach - finally, there’s how the rubber hits the road: the cadence of communication, the rate at which updates are pushed into the account, etc. I’m an unabashed fan of batched changes + updates (because it avoids the day trading in ad accounts issue I’ve seen far too often) – but you should have this conversation up-front with any partner. Ask about their philosophy + communication style. Ask who does the work, how it’s QA’d and why. Ask how many creatives they recommend launching, and how they go about creating and iterating on those assets over time.
Any quality agency has clear, specific, readily-intelligible answers and examples to each of these buckets. They can provide examples of budget models, along with how they were used to adjust a client’s strategy. They can provide examples of reports, dashboards + metrics, showing why certain decisions were made or how they balanced a client’s desire for efficiency with the need to grow the business. They can defend why they take an eyes-on, hands-off approach to management of ad accounts, or why they believe in launching X new creatives for an account of your size.
If the agency doesn’t have an answer (or it’s a generic, pedantic or jargon-filled bunch of gibberish), consider it a red flag.
As I stated at the beginning, I’m not pro- or anti-agency by default - I think good agencies can accelerate growth and help brands reach ever-greater heights. I also know that there are legitimate scumbag agencies out there that prey on clients. I hope this issue helps those good agencies gain more clients, and allows brands to avoid getting burned by the bad ones.
Cheers,
Sam
Loving The Digital Download?
Share this Newsletter with a friend by visiting my public feed.
Follow Me on my Socials