Some end of my internship reflections on the good and less good about creative agencies and how they work.
Here is an LLM version of the recording of my thoughts — it is not exactly the same as the recording, but it captures the essence rather nicely.
here are some of my reflections about my internship at Ogilvy. First of all, I was incredibly grateful that the people at Ogilvy were so open to give me such. gift of spending so much time with them, to explain to me what is going on and how they work and so on. I deeply appreciate that generosity and I don't take it for granted.
My first observation is that the people who seem to work in advertising, or maybe it's just Ogilvy, are very young, very attractive and dress in a very interesting way. In fact, taking a coffee break from time to time and just watching people around was fantastic. I also think that they thrive on creativity and that they very nicely challenge each other to be more creative. It doesn't feel like a competition of who is more creative. It feels more like a team sport of trying to get people to be more extreme, think more creatively, and have more and better ideas.
The collaboration seems very interesting. People seem to have lots of time for discussions. They seem to truly enjoy brainstorming. I was in multiple sessions. There was never a sense that the hour or the two hours are ending and we have to come up with a solution. In fact, it felt more like a process of getting marinated in the ideas and that the marination was for its own sake. In a few cases, people went back to ideas that were floated before. In any case, it seems like they valued the creativity, they enjoyed the creativity, and they were not rushing themselves to come up with a creative process. I'm sure that there are many other aspects of the work in which they rushed themselves very much, but in the creative process, they didn't seem to rush themselves very much.
My next point is that the whole process of the agency is very different than the process that I do when I go to companies and try to get them to adopt some behavioral change. When I usually come to companies, I usually start the day in a workshop, and the workshops look like this. I spend 45 minutes discussing a general framework for behavioral economics with a few main points, and then I turn to the audience and I say, how would you apply these specific points to the current things that you're currently doing? In other words, I'm asking them to think about how they would look differently in what they're doing if they took the behavioral economics perspective.
People come with all kinds of ideas and they start to understand the value of behavioral economics within their framework, within their daily business. Then I say, okay, now let's move to another topic. Let's talk about the psychology of money or motivation or loyalty or trust or a topic. And then I give the lecture and then I say, and now how would you apply that principle? So we just learned some things about trust. We just learned some things about loyalty. Now that you understand trust and loyalty a little bit better,
What would you do with it? And then we move throughout the day, and usually, we end the day by saying, okay, up to now, everything we did was to say, how would you do what you're doing currently? Slightly better with these principles from behavioral economics, but now let's do the last session and let's think about how would you approach your business from scratch? If you have all of these principles in mind. The idea, by the way, is to create some tangible ideas, but also, of course, mostly a hypothesis that needs much more work before they are applied.
Okay, so I have my process in mind in general, and of course, when I was exposed to the agency model, I looked at that in contrast to my model. The agency process is very different because the creative, people in the agency, when they approach a problem, they approach it as a tabula rasa. They say, okay, we don't understand this problem with X. Let us listen to customers. Let us get data from the company that this is their project. Let's understand what they understand, and what they don't understand, and let's do some primary research and so on.
I really value their approach because they come to a new situation, they don't come with preconceived notions, they don't come with specific ideas, and they're open to interpreting it. And I'm sure that the right model is to do probably both. But I started wondering why don't they do it all, my model? Why don't they take everything they've done so far on trust? And why don't they take everything they've done so far on pricing or reputation management and so on? And why don't they start with something that they know already?
It struck me that there are some good reasons, of course, to come tabula rasa and not be involved and so on. But it struck me that one of the reasons for this is the hourly model. The hourly model basically says there are things you get paid for and there are things you don't get paid for. And when you get paid for, you get paid by the hour, not by the quality of the idea or not by the impact.
In this particular case, what it means is that if a company spent time deeply understanding a topic, like trust or loyalty, and then they went to a client with that, under the hourly model, they couldn't charge them for it. This work could serve lots of clients, but if they start with, here is what we know already, they wouldn't be able to charge a client for it. And in fact, the whole work process might take shorter and actually give them less money.
So if you think about the hourly model, it means that they don't get paid on research that they came up with early, before the engagement. In my case, I'm an academic. This is not part of the feature. I'm there about creating learning. But an agency, if they put 30% of their effort into creating this generalized understanding of these important topics, nobody would pay for it. And not only that nobody would pay for it, once they pitch it to a client and
The client says yes, and it cuts the duration of the project by, let's say, 30%, the agency would make 30% less money. So the incentives are actually quite against creating generalized knowledge and create against spending effort to understand the topic in a general sense, not in a specific sense, but in a general sense.
Then it also made it clear to me that this hourly model has other downsides. For example, it makes companies, agencies, less incentivized to spend time on proposals. Nobody pays for proposals. So eventually proposals are probably not as good as they could have been.
Now the companies who are asking for a proposal probably think it's a great deal. They say, oh, you know, I'm asking for five companies to all give us proposals and they all give us proposals and it's all for free. Yes, but are these the best possible ideas? Do we just pretend that we got five proposals for free or is it in fact the case that we got five not so great proposals that are all for free but are not really giving us even an indication of what these people could do if they spend real time on this?
I'm guessing that the assumption is that if we got the free proposal, it's a really good indication that the correlation between the quality of the free proposal and the final paid work would have been a correlation of one. It's a predictive element that the free proposal is predictive of the real quality, but I don't think it's necessarily the case. So it could be that companies who are using this strategy of getting free proposals and using those to judge, to predict the quality of the real effort when it's getting paid for are making a mistake.
By the way, in nothing the correlation is one, but it could be the correlation is not even that high and that the correlation between the free proposal and the real work is not that high. To do this study would be complex, but I think it's worthwhile thinking.
And then I kept on thinking about other elements of the hourly model and it became clear that it also reduces collaboration. The goal of having a research lab in my case with many different people with many different skills is to get lots of people to be involved and to work together. The goal of an agency is to get lots of people with different creativity to work together, but if the model is a billable hour model, it is less likely that people would help each other.
In fact, the structure is to say, hey, I want 15 minutes of your time. Let's say I came to you and I say I want 15 minutes of your time and you say, oh, that's billable. And now you also have an incentive to say, well, this will not take me 15 minutes, it will take me three hours. The moment you move for a billable hour model, you get people to be less incentivized to help each other and you have lots of things that are wrong.
Finally, I would say the following. Creativity is very magical. Creativity is very magical. And I understand why clients want control and billable hours seems to give clients control.
Trol. And I understand why agencies would want to portray to people a process and work and what we call operational transparency. What we call operational transparency is the idea that we want to show people effort. We want to show people effort because that increases willingness to pay. So I understand all of those things.
I understand why we might want the reasons for billable hours and control. I understand why the agencies want to show a process and value per money and so on. But I think that these things might also hurt creativity. And it will be useful to think about maybe a hybrid model where maybe we don't throw away completely the hourly model, but we have some of it that is not in that model. And maybe we have a more controlled process for creativity, but not everything is controlled creativity.
Anyway, I had an amazing time. I learned a lot. And I'm looking forward to visiting again.
