The digital agency model is broken

By | February 2, 2014

I used to own and operate a digital agency.

I did it because I thought it going to be lots of fun.

I don’t anymore.

Going in to this adventure I honestly thought that we would have fun building fantastic interactive products for amazing companies and that we would achieve that by building great partnerships with our clients.


I didn’t achieve this and that’s why I left. Looking back, I thought I’d share why and what digital agency managers should do to get out of the vicious circle in which they are trapped.

Consider the traditional agency model and the relationship between the agency and its client.

The pitch

1. When a client first approaches a digital agency, they provide a brief overview of the product or campaign they have in mind. They request that the agency come forth and provide their best idea on how to execute the project. “Give us your best idea, and maybe we’ll give you our business. If you do well on this first project together, there will be many more to follow.”

2. With their sights set on the long-term potential of the business, the agency allocates its best minds to build a fantastic proposal.

The cost of a single pitch for an agency? 10 to 100k, according to many sources.

3. As part of the pitch, the agency is requested to commit to a dollar figure as to how much it would cost to build product. As agency owner, you know that you’re competing with at least 2 or 3 other agencies, so you try to make your price appealing, focusing on the long term benefits of this client.

The problem with the pitch

Already, there are three critical issues at this juncture:

The obvious one is that you’re doing the most critical part of the work without any type of guarantee that you’ll be getting the business. It’s indeed a rather big problem, one which has been discussed at great length – but for me this is only part of the problem.

The second, equally obvious issue is that you’re being asked to provide a quote before you’re able to discuss the finer details of the project with your client, because the price you quote is a critical factor in how clients evaluate the possibility of working with you. That right there makes it impossible for you to even have a chance at estimating the scope of work accurately.

But the third and bigger problem IMHO is that developing the core concept of a product or campaign is not something that a digital agency should do on its own. It should be done in collaboration with the client. And until you officially get the deal, you’re not really collaborating with the client.

The honeymoon

4. An agency is selected, while 3 or 4 others take the hit.

Let’s pretend your agency is selected. Congratulations! You’re excited. Your client is excited. He showers you with compliments about the quality of your pitch, and the honeymoon phase begins.

“Let’s get together to brainstorm some more”, he says.

You get together. You bring champagne and croissants because it’s an exciting time – the beginning a new relationship. Again, your client emphasizes how much he loved your pitch. He just has a few ideas that can make it even better.

At this point, one of two things happens:

Situation #1: your client proposes some “minor changes”, completely oblivious to the fact that these “minor changes” have a huge impact on the required development efforts and the schedule you’ve prepared as part of your pitch.

Because you’re still in the honeymoon phase, you convince yourself that this is nothing but a minor bump in the road, and you’ll try to make it work in the calendar you’ve proposed initially. Let’s call this what it is: wishful thinking.

Your client is happy and you’re off to your office to begin the work.

Situation #2: your client proceeds to describe the campaign he’s wanted all along, which has very little to do with what you pitched.

Generally when this happens you start to feel a little uncomfortable in your chair. You think about the 10 – 100k worth of planning and strategy that just went down the drain and you’re mildly annoyed, to put it nicely.

But of course, you keep smiling and nodding, because in the back of your head you’re thinking “well, we’ve just acquired this client, let’s make him happy so that we can get more projects from them in the future.”

You go back to your office, and you invest a significant amount of time to rearrange your development calendar and shuffle the agency’s resources.

The problem with the honeymoon phase

There are two key problems with both of these situations:

The first, please excuse my bluntness – is that the agency doesn’t have the balls to say “hey, this isn’t what we bargained for” for fear that the client will turn to another one of the bidders who would most likely accept the mandate despite the changes.

With so many digital agencies competing for business in major cities, if you don’t accept the project and the conditions on the table, another agency will. So you think “I might as well, I have bills to pay after all.”

The second issue is that you’re not doing your client a favor. You’ve asked your best minds to come up with a great concept, building interactive products is your company’s core competency, and yet you allow your client to dictate the direction of the project.

Your client hired you because of your expertise in an area where he presumably isn’t an expert, nor intends to be since he’s outsourcing, and you’re letting him drive the development of the product. Neither one of you is doing the other a favor.

Development starts

5. Despite the initial hiccup, you begin the work with your team.

The team members who were involved in the pitch are slightly annoyed that the plan has changed, but you’re still able to generate a certain level of enthusiasm amongst your team.

You provide the initial wireframes and design draft to your client. He expresses that “it’s not quite what he had in mind.” No problem – you go back to the drawing board, make a few changes, and come back with another proposal.

Again, it’s not quite yet what he was looking for. He asks if he can talk directly with your designer. After some back and forth, you haven’t really gotten a firm sign off on the design, but time is running out, and you have to start developing.

Your developers ask you a few questions about certain aspects of the project that you hadn’t yet completely defined. You answer as best you can.

Typically this is the moment where your developers start using complex programming jargon that you don’t understand. “Oh well this is going to require some complex server-side rending with a MySQL cloud infrastructure on a 6-gig SSD machine.”

Fortunately you have a magical go-to response ready for this kind of situation: “Whatever – just get the job done, we have to present a first version in two weeks.”

Your developer looks annoyed, he knows full well that’s not going to happen, but since you seem unwilling to continue the discussion, he goes back to his desk.

6. Comes the time of the first delivery.

Two days before, you start putting an enormous amount of pressure on your team. The first delivery is critical – you have to make a good impression.

The day before, you ask if you can see the product for the first time. Your developer answers you something along the lines of: “I haven’t quite yet finished the metadata structure of the database table for the XML feed, so I’m going to need a little bit more time.”

Not wanting to sound ignorant, you say: “oh, right, the XML thingy, right. Ok. Just get it done as fast as you can.”

Then the designer tells you that the client has again asked for the modifications to the design, so he can’t quite yet send over the files to the developers, but no problem, he’ll finish it today.

Later that day, you stop by your team members’ desks, and they let you know that they’re going to need a little bit more time.

Night comes. Still no product.

The next morning, you bring coffee and donuts for the team. Your intention is to cheer them up and also to let them know you need their collaboration today more than ever.

Around noon, you’re finally able to see something that ressembles what you’re supposed to deliver. Oh no – wait, not quite. Your developer just needs to tweak a few more things.

3pm comes. Your client asks how it’s coming along. You playful respond: “it’s coming along, it might be a little late, but it’s coming” with a smiley face.

7pm. You finally get to see the product for the first time. And… it’s not quite there yet. Many parts are missing, it’s filled with bugs, and not really in a presentable state.

You keep the team as long as you can. Finally you write to your client and say: “I’m sorry, it’s not going to happen today, I don’t want to deliver something that’s not quite ready, we’re going to need a bit more time.”

A few days later, feeling the heat, you deliver something that ressembles what you promised, but that even you are unhappy about.

7. Finally your client has something tangible that he can see.

He shows it around the office. His team is excited to see it. They come up a list of 63 things that you should do to improve the product.

That’s when the relationship starts to get tricky.

You go through the list of 63 items. Some of them require dismantling components that have already been built only to rebuild them differently. Other ideas you know will be time-consuming, but make tremendous amount of sense – you just can’t say no, because in the end, you’re company’s name is also on the product, and you want to build something good.

All of the sudden you’re facing a situation where you’d need 2 additional months to get it done. But you have a contract with your client that states that you’ll deliver for a specific date, and they have a lot riding on this. They’ve already bought media placement, maybe even TV ads.

At this point, agencies generally do two things:

One, they put a tremendous amount of pressure on the team to deliver a beautifully creative product in a ridiculously short timeframe, and two, they throw more people onto the project, hoping that 9 pregnant ladies can make a baby in 1 month.

As agency owner, that’s when you look at your burn rate and you think “this client better be worth it in the long run, because we’re not making a lot of money on this project.”

Development continues

8. Somehow your team miraculously pulls through and delivers a second version of the product that integrates a lot of the changes requested after the first delivery.

Time is running out and there’s still a lot to do. Over time, the key members of your team are needed to help on pitches for other potential clients. Suddenly, you have to deliver the project with a group of employees who weren’t involved in the initial planning of the project. Knowledge is lost. Mistakes are made. Productivity diminishes.

Your client comes back with more feedback from everyone in his office – another 63 change requests.

The relationship isn’t so cordial anymore. You’ve burned through your entire profit margin, and by then you just want to get it done. You’re exasperated by the new change requests.

At this point you have two choices:

Choice #1: You go back to your client and you say: “we can’t make these changes without more money.” He replies: “well we don’t have any more money, because you said you’d do this for [x amount.]” The relationship deteriorates.

Choice #2: You start cutting corners to get this out the door as fast as possible. Every additional minute spent on this project, you’re losing money.

You give up on the many great ideas that you had to improve to the product. From now on, any task that isn’t critical to the delivery of a project is out of the question.


9. Deadline is a few days away.

You send another version of the product to your client. “It’s almost there,” you tell him. It seems like there’s finally light at the end of the tunnel.

He responds with more change requests. You flat out refuse them all: “we just can’t do it if you want us to deliver on time.”

Your client is disappointed. He didn’t get what he wanted. You’re frustrated because it feels your client doesn’t recognize all of the hard work you’re putting in.

By then, you thought the project would be long over. The company’s other projects require your staff, and it’s making every other project late. Tension between all parties increases.

48 hours before the final deadline, it sounds like you’re going to pull through, but you’ll need every minute until then. Developers are invited to sleep in the office for the next few days. Lunch and dinner will be provided if they accept to eat at their desk and continue working while they eat.

That nice QA testing period that you had planned goes out the window. “Let’s just pray that it works, our developers don’t make bugs anyway.”

And here it is – the biggest mistake a digital agency can make

The bitterness of poor quality remains long after the sweetness of low price is forgotten.

The bitterness of poor quality remains long after the sweetness of low price is forgotten.

On the final day, 15 minutes after the official deadline has passed, you unveil the curtain on the final product.

The relationship with your client has deteriorated, but on this joyful occasion, it remains courteous. Your team enjoys a few beers, talks about how that kind of project from hell will never happen again, and everyone goes home for some well-deserved rest.

10. The next day, your client calls you in a panic. There’s a problem with the site in Internet Explorer 8.

“Who still uses Internet Explorer 8?” you think.

Your client’s CEO, that’s who. Him and a number of high-profile companies with very strict IT policies.

You search all over the office for a computer with Internet Explorer 8. You finally find one in the back room. You dust it off and give it to your developer, who is supposed to be working on another project that’s already late because of this first one.

Your developer fixes the problem. He deploys the fix. Problem solved. You can finally attend to the other burning fires in your office.

11. For the next two months, your client calls you daily with new problems. Broken links, emails not being sent, weird layout issues specific to iPad 3 with iOS 7.0.1, and with every call you resent him more and more.

You know it’s partially your fault, but you’re not willing to accept it. “It’s just a crazy client” you tell yourself.

Finally after two months of daily bug fixing, you cut the cord. You’ve lost a huge amount of money, your team cringes at the sole mention of this project, and every other project is late because of this one. You inform your client that you will no longer support the product and he should look for another agency.

He’s angry. He feels that you’re letting him down after dropping a malfunctioning product on his lap, and he vows never to work with you again.

Everybody loses.

I know that this isn’t everybody’s reality.

But sadly, it is one that is all too common.

I’ll be honest, I think it’s difficult to do something about it. That’s why I personally moved on to something else.

The fact is that money is the primary motivator in this kind of relationship. But it shouldn’t be. QUALITY should be the primary motivator for both parties.

Because quality is the only thing that both parties can fundamentally agree on.

Recently there’s been a movement from industry leaders who have spoken against the current model.

The team at Dynamo, with whom I’ve had the pleasure of working a few times, has publicly announced that they are moving to a retainer-based model to avoid the shenanigans of the generally accepted model.

I think that it’s really great, and I think that it takes some serious balls. One might argue that Dynamo has been around for many years and their realizations to date are what makes this possible – their clients know first-hand that Dynamo will deliver quality, therefore the required trust is in place.

I still think it might be difficult to achieve since most clients like to know beforehand how much a project will cost. But it would be really great if Dynamo can accomplish what they’ve set out to do with this change.

But what about younger agencies who are still working towards building their regular clientele?

Freshbooks, the Toronto-based accounting software company, as published a fantastic eBook called Breaking the Time Barrier in which they advocate a value-based approach to quoting project instead of an hourly-based approach.

What they propose is that instead of pitching in the traditional sense, digital agencies should work with their potential clients to determine the probable return on investment that the the project will yield.

With that value in hand, the agency can then quote a price that is a reasonable percentage of that return on investment – say, 20% – instead of attempting to determine the number of hours required to build a product that everyone knows will change along the way.

Embrace change to focus on quality

With money out of the equation, you can focus on your common goal: building a quality product. That’s the only way both parties will benefit from the relationship.

There are two aspects that for me are crucial if you want to build a successful agency:

One, hire nothing but the best. Don’t hire two junior developers thinking that they’ll get more work done than a single senior developer. If your core value proposition is web development, hire the best web developers that you can find. Your clients will thank you, and they’ll come back to you.

Two, don’t approach projects as single, fixed-term assignments. Don’t just sell the development of a product. Instead, approach web and software development as a long-term experiment.

I find that the lean six sigma approach: Define, Measure, Analyze, Improve and Control really applies here.

DMAIC Lean Six Sigma

Define, Measure, Analyze, Improve and Control

Launching an interactive product for a client should only be the beginning of the journey. When you’re planning a development project, plan ahead for numerous iterations after the first one. Include quality assurance and usability testing. Analytics. A/B testing. SEO. Everything a good website or mobile app needs.

One-off development contracts are like one night stands: you both feel awkward when the deed is done.

But if you are both already committed to making it work in the long run, then there’s room for the relationship to bloom and the potential to produce beautiful offspring.

What do you think?

Send us your thoughts on Twitter: @crowdsourcingqa