How to push back on your agency without killing the relationship

Action Box: The New Economics (and Philosophy) of Value in the AI Era

Learn how businesses can describe and price their value in the AI era. Join us for a free AI Guild session on Wednesday, September 2nd, at 10 am EDT. Register now. (from MECLABS AI, MarketingSherpa’s parent company). 

Step #1: Set the agency up for success 

Before we point a finger let’s look in the mirror, shall we? If your advertising or marketing agency is presenting campaigns and strategies that are off it could be because they don’t have the proper context about your organization. 

This is where every agency-brand relationship should begin. Most agencies will be adept at having an onboarding process to get this information. But there are a few things you can do on your side as well. 

Invest time, not just money, into the agency 

When looking to hire an agency, you consider their financial cost and how it fits in your marketing budget. But don’t overlook another cost beyond the money – your (and your team’s) time. 

When you sign an agency, you are likely hiring them for their expertise…but also their hands, so to speak. Your team has limited bandwidth and the agency can help you get things done. So you’re thinking it will save you and your team time. 

And while a quality agency should execute on your brand’s behalf (once trust is established), they will still require some of your time to get them set up properly as well as also your ongoing time to make sure they are well aware of key changes in your company. 

I’ve seen brand-agency relationships where the agency was working in a vacuum outside the company. This could be because the key decision maker who originally signed them left the company or took on a new role. Or it could simply be because the internal team had shifted focus. Either way, the agency’s ideas will be off the mark if they don’t understand the full and current context of what’s going on in your company – strategy shifts, budget moves, upcoming launches, etc. 

Arm them with your company’s (and products’) value proposition and history 

Sometimes this happens because a company hasn’t internally nailed down the value proposition. If you aren’t internally clear and aligned, how can you expect an external company to be clear? 

You also want to share a history of what marketing your brand has done before – what worked, what didn’t, why, how do you know it didn’t work, etc. 

If you have a test library, this can be especially helpful. It will save the agency from going down a dead end you’ve already tried…or allow them to build on the customer wisdom you’ve already accumulated. 

Define success 

What are you expecting from the agency’s campaigns and strategies and how will you measure it? For example, growing awareness will necessitate a different strategy than improving conversion. Getting more new total customers is not the same as focusing on revenue per customer. 

Help them navigate your company (and partners, clients, etc.) 

If they’re writing case studies, do you need to make introductions to key customers, or perhaps to account execs who own the customer relationship? If part of their role is not just external media but updating pages on your website, do they have access or a direct relationship with an internal person who can make the changes? 

I hate to use the word internal politics, so let’s say at least make sure they know the lay of the land. The who’s who in the zoo. A RACI matrix (to understand who is responsible, accountable, should be consulted, and informed). 

And, perhaps most importantly, make sure they’re able to have the right meetings with the right people. When they are first learning about your company or pitching ideas, it can help to have small meetings. 

For example, when I discussed corporate communication and marketing innovation with Jasmin Guthmann on How I Made It In Marketing, she told me a story of when all the brand marketers and agency teams flew into Frankfurt to finalize the creative and kickoff a campaign. Her boss got upset about a certain feature, and from there “it all went to hell within seconds.” 

The meeting (and campaign) eventually got back on track when the global marketing officer stepped in. “Everything came to a grinding halt, which is when Ian Forrest stepped in and calmly but very firmly reminded everyone of their roles and responsibilities and the reason we had gotten together in the first place.  So he called everyone to order and to the higher cause made a decision right there for them, weighing the input everyone had been giving. And that really saved not just the day, but I think the whole project,” Guthmann said. 

Guthmann told me she learned two lessons from that experience. One is that the socialization of ideas should happen in advance, not at a major launch meeting like this.  

Advocate for them 

The other lesson Guthmann learned was how to de-escalate a highly emotional situation. 

And kudos to Forrest for stepping in and doing that. In this situation, other execs might have simply thrown the agency under the bus (even if they had nothing to do with the issue). 

When you have small meetings before the big meeting, you (or your team) may have encouraged them to make specific changes to the campaign or strategy they’re proposing. If someone in the bigger meeting or someone higher-ranking than you in the organization doesn’t like those ideas, step in and own them. 

I’ve had this happen to me when working for an ad agency – made changes based on a marketing manager’s insistence in an earlier meeting only to fall on my sword when the key decision maker didn’t like them in a later meeting (and not a peep came from the marketing manager who actually ordered the changes). 

If this is the relationship you have with your agency, they will do everything they can to avoid the meeting before the meeting and only want to speak to the key decision makers…ultimately hurting all the campaign and strategy ideas. 

Step #2: Understand their point of view 

OK, now that we’re sure you’ve set the agency up for success, let’s jump into the moment when you get the agency pitch…and you don’t agree with it. 

I’ll tell you an upside from having a little experience under my belt in my career – I’ve mellowed out. 

Earlier in my career, I would come in too hot too quickly.  

I’ve learned to start by asking, instead of telling. If you’re pitched a campaign you don’t think is right, here are some questions you might ask: 

• How does your campaign serve the customer? 
• What specific customer insight led you to this recommendation? 
• What customer evidence did you use – calls, interviews, reviews, support tickets, search terms, win/loss notes? 
• What is the single core assumption this strategy is built on? 
• Why did you make [specific creative choice]? 
• What tradeoffs are we making with this campaign or strategy? Why are these the right tradeoffs to make? 
• What would make you pause this strategy within two weeks? 
• What’s the closest analogous campaign you’ve run, and what were the results? 
• Where did it fail in the past, and what did you change to fix it? 
• What part of this is tailored specifically to our category and buyer? 
• What are we doing in this campaign or with this strategy that our competitors could not do? 

Also realize, if you’re having multiple meetings as described in Step #1, you rarely have to make a kneejerk reaction when you’re first pitched the idea. That gives you time to work on some of the following steps…after clarifying and getting a better understanding of why your agency is pitching you that specific idea. 

Step #3: Get feedback from your network 

We all have our blind spots and our personal peccadillos. And we’re limited by our personal experience. 

So I like bouncing – bouncing ideas off of other professionals I respect and seeing how those ideas land. And then understanding why they had that reaction. 

If you work in a large organization, you may have a robust internal network you can bounce these ideas off of. 

But if you work alone at home, it helps to be part of a bigger virtual community (and even if you do have a large internal network, the external virtual community can help you escape your company’s internal groupthink…especially helpful if you’re the lone person pushing back). 

This is a perfect time to mention the free-to-join AI Guild, brought to you by MECLABS (MarketingSherpa's parent company).

Think of a community as your outside lab. You can garner feedback based on their previous experience. Or what’s working right now for your peers. 

The agency has the benefit of working across several brands. You need that context and insight as well, and can get it from a community. 

Step #4: Leverage your own customer wisdom 

A core skill of the marketer or entrepreneur is customer intimacy. Whatever your other tasks or official role, make sure you stay close to the customer and understand what drives them. 

If you’ve built up that skill and wisdom over time, now is your chance to use it. The agency may have a perfectly sensible campaign, they just don’t understand a specific nuance of your customer. 

So instead of just shooting down the campaign entirely, share what you’ve learned, why it gives you concern, and give them an opportunity to pivot and react. Here’s an example of understanding those customer nuances. 

Quick Example: Asking if the campaign matches the emotional state of the buyer 

“When agency advice feels off, ask whether the campaign matches the emotional state of the buyer,” advised Tricia Watts, real estate investor, MaxNet Homes. 

BEFORE 

The original agency direction was focused on urgency-first messaging, with lines like: 

• ‘Sell your house fast for cash.’ 
• ‘Get a cash offer today.’ 
• ‘Close in as little as 7 days.’ 

But that approach did not sit well with her understanding of the ideal customer. “Many distressed sellers are not ready to be pushed. They’re trying to understand their options,” Watts said. 

She knew this because she had been listening to her ideal customers. And they’d been telling her things like: ‘I don’t even know where to start,’ ‘I’m behind, but I don’t want to be taken advantage of,’ and ‘I need to know what happens if I cannot make the next payment.’ 

AFTER 

Instead of leaning into urgency, the team decided to lean into trust. They launched a campaign with plain-language educational content around foreclosure timelines, probate sales, repair-heavy properties, and how cash offers work. They used lines like: 

• ‘Understand your options before foreclosure.’ 
• ‘Sell without repairs, cleaning, or showings.’ 
• ‘Choose a closing timeline that works for your situation.’ 

They also changed the landing page flow. Instead of pushing the form immediately, the team added more explanation around the process, what sellers could expect, common situations they help with, and why a cash sale may or may not be the right fit. 

RESULTS 

Watts claims that lead quality improved because conversations started with more context. 

Now, she uses these insights about her ideal customer’s psychology to inform future marketing campaigns. “If the copy would make a distressed homeowner feel pressured before they feel informed, it’s not ready to run,” she concluded. 

Step #5: Pull a reality check from your own (or other vendor’s) data 

I’m not that guy. I’m not the consultant that can stand in from of a room and make bold projections about 10x scaling. I’d rather base a projection on 1PD (first-party data) instead of WAG (wild guess, and I’ll let you figure out for yourself what the letter in the middle stands for). 

So when it comes to a projection, I want to see what we did before. What was the conversion rate? What was the revenue growth rate? 

Now, if we’re projecting to improve on that – why? Let’s get specific. Why do we think we’ll do X amount better? 

This doesn’t only apply to your in-house data, and it doesn’t only regard growth projections. Yes, your agency shared their opinion based on their experience. But you likely work with other vendors – platforms that applicable data. Here’s an example. 

Quick Example: Referencing send data from previous domains to ramp slowly to 1.2 million sends per day 

AGENCY PROPOSAL 

Last quarter, NC Kumar was working with a new brand to set up its sending domain. The growth agency that was also working with the brand told him to ramp quickly – 50,000 sends on day one, with the full send by the second week. 

 “Their pitch: warmup is a relic, ISPs stopped enforcing it in 2023, here are 14 slides of charts,” said NC Kumar, founder, sendmsg.io. 

DATA REALITY CHECK 

“I asked our deliverability lead to take the agency's curve and replay it against postfix logs from five prior domains we'd warmed. Four out of five tripped a Gmail block on day two or three,” he said. Specifically, when hourly sending crossed roughly 3,000 to 5,000 from a domain with no history. 

“The recovery time on those four wasn't quick either – averaged 38 days of throttled sending before Gmail started letting them through normally again,” Kumar said. 

RESULTS 

They decided to ramp up slowly: 

• 200 sends on day one 
• 50,000 per day by day 14 
• 200,000 per day on day 30 
• 1.2 million per day on day 90 

The bounce rate stayed under 0.4%, with complaints around 0.02%. 

“The lesson I keep coming back to: your MTA (Mail Transfer Agent) logs and Postmaster Tools are operators talking to you in real time. An agency's playbook is statistics about what worked for other people last year. When those two disagree, the logs win,” Kumar concluded. 

Step #6: Determine how ‘off’ the campaign is 

Sometimes, you just can’t abide the agency’s campaign or strategy idea. In that case, the best option is to kill it (using the information and data you’ve collected so far as evidence) and get a page one re-write. 

Step #7: Identify the smallest test that validates the most critical assumption 

But if you’re working with an agency you trust, it’s very rare for the idea to be entirely wrong. In that case, start with a small test: 

• Determine what critical assumption the strategy or campaign rests upon (which you probably garnered from your questioning in an earlier step) 
• Hone in on your concern around this assumption, and communicate it using the data and information you gathered in previous steps 
• Develop a hypothesis for the smallest possible way (in cost, risk, time, etc.) you can test that assumption. 

This balances two equal but opposing forces. Yes, you should trust your gut. What we call a ‘gut’ is based on your specific experience and understanding of what your customers need. It’s a subconscious pattern recognition. 

However, very few of us have the perfect ‘golden gut.’ I’ll readily admit, there were decisions I’ve seen launched that I was vehemently against and thought would fail miserably…but actually worked pretty good. 

Of course (clears throat, brushes off lapel) there have also been ideas I tried hard to kill, couldn’t, and had a sad but upfront view of seeing them go down in flames. 

So get past the gut. Get past the politics. And get some data with a test. 

Related resources 

Marketing Campaigns: Lose the brand ego and lean into humility (podcast episode #130) 

Strategy: Don’t think of your customers as a ‘target’ to acquire (podcast episode #65) 

Interact with MECLABS’ AI Guild community