How to estimate design work
You can tell easily how much a car costs just by looking at the price tag. But what is the cost of producing that car? Or, how much did it cost to design that specific car, so it can withstand competition for at least one product lifecycle, be reliable, meet safety and brand standards, customer demands, and perhaps even become a design icon at the end of its production cycle? You perhaps agree, that foreseeing the cost of design – and its Return Of Investment — is anything but easy. Design is an iterative process and you don’t know at the beginning of a new project how many design iterations you need to go through, justify, negotiate and test until you arrive at an acceptable solution, especially when it’s someone else you are designing for or with. In this article, we explore some of the pitfalls of cost estimations for digital product- and service design. We start with a summary of my learnings from working for almost 22 years in the digital, creative space, on more than 200 projects, in partnership with 25 creative agencies, and in-house design departments.
You are being tasked to estimate the cost of something that does not yet exist, but need to know when it will unveil itself, where it will appear, how it is going to look and what it can do for, and will do to, the people who will eventually use it. In short – you are forecasting lightbulb moments.
Key Takeaways:
- Design is not a sprint. Good design is expensive, it involves many steps and layers, it is time-consuming, it relies on many decisions and variables you often cannot foresee. This is especially so when you design for and with other people, delve into new unknown markets, and know little about the product category, its applications, uptake, (mis)usage, and the mood your stakeholders are in.
- In reality, you are not estimating a specific deliverable. You are estimating a process; the path you must take to arrive at an agreed, desirable, and functional solution. This path is not linear; it’s explorative, at times chaotic, highly iterative, selective, experimental, and corrective.
- Designing for yourself is different than designing for someone else. That’s why identical design tasks can result in extremely varying timeframes and costs.
- Clients and agencies often don’t understand the difference between the output and the outcome of a design project. It is your job as a designer to make that difference clear in the way you approach tasks and design challenges. “Oh, I didn’t know we need to progressively onboard new users to our mobile app”, and so on.
- Clients rarely understand the full complexity of a design challenge that needs to get solved and how it might be tied to deeper underlying business problems, nor who best can help them solve these issues. For example – say, a companies website underperforms not because of an outdated UI design, but because the editorial team does not have a proper pipeline set up, no time given to write compelling content, or the skill set needed to use the CMS effectively.
- Software design is especially laborious to estimate: There are many ways to infer what users want or should do. There are many ways to choose a design and many ways to fail in predicting consequences for users. Not to mention the tech requirements you need to meet.
- Most digital products and services are tied to larger ecosystems and networks and changes made to one of these nodes affect many others.
- Just because your future client or teammates gave you a picture of the desired end-product, doesn’t mean you get there faster, or it’s THE THING that is going to be successful in the market months or years down the road. “Just copy Apple and we are all good.” 😵
- Being able to envision the end-product helps to choose the path to get there and estimate for more accurately. Oftentimes, the client does not need a custom design, cannot wait for it, and would be better off using an off-the-shelf standardized ‘design template’ as a starting point.
- Siloed teams estimate less accurately than collaborative, integrated design teams.
- We need a new qualitative and quantitative approach for predicting and estimating complex design tasks.
- Teams who estimate design efforts more accurately prove that they understood the clients' problems better.
Design complexities
I know you are not a lawyer but just imagine for a brief moment that you are drafting a contract that outlines the plan for how you will design, build, deliver, assemble and support a multi-family living home for 10 families. The writing of the first draft will not take long since you can rely on templates from your colleagues who have completed the same task many times before. You are finished within a day and then pass it on to all 10 parties to get sign-off. Here is what happens next.
The first family comes back and signs it right away. They love your plans. The second family likes most of your suggestions but would like to adjust one paragraph. The third family asks you to adjust two additional paragraphs and is curious to hear what made you decide to go down a certain route when it comes to building materials and contracting workers for the foundation; they want to see alternative suggestions. Then the fourth, the fifth, the sixth, and so on ... family comes back with ideas, suggestions, and concerns. The amendments you made for the second family forced you to renegotiate with all the other families, plus you needed to get another approval from the city, that the changes your clients want to see align with laws and regulations.
You get the idea in which direction we are heading here – designing something for yourself is not like designing something for someone else. It affects your design process in general; your approach, the tools and methodologies, timeframes and timelines, and essentially estimation and pricing.
For instance: What steps would you go through if FIAT CHRYSLER all of a sudden knocks on your door and asks you to come up with a new name for their new car company? You certainly would not write down your three best name choices on a piece of napkin and present it back the next day and charge for it. You would be surprised, I had senior colleagues who think that’s the approach you should estimate for. Good design is not a sprint.
On the hunt for a wedding band, you have two options; get a run of the mill ring from PANDORA, or head over to a goldsmith who can make you a bespoke piece of jewellery. Decide what type of design agency you want to be, and how you want to approach and charge for the value you provide.
Design agencies are all getting it wrong
You’ve probably seen this scenario play out at least once in your agency life:
A new Request For Proposal (RFP) arrives at your office's door and after a brief initial assessment, the leadership- and sales team decides to respond (red flag 1). The issuer, a large bank, is asking for vendors to help redesign their website.
Several processes get kicked-off and members of your division are being tasked to respond swiftly with a compelling proposal deck and a price tag that’s within the given or ‘internally eyeballed’ budget range (red flag 2). The issuer – the bank – is giving you the chance to uncover missing details of the RFP during a life Q&A session with some of their key stakeholders and project owners (red flag 3).
You feel prepared to start estimating the design work. The team comes together and looks at all the information they have access to (red flag 4), conducts a quick analysis of their current website (red flag 5), and perhaps looks at past projects to pull-in some comparables (red flag 6). All relevant team members contribute to the response deck and the strategy team drives the overall response approach (red flag 7). The product manager, who recently joined copies over from an old deck, a fancy design methodology – adding the flair of human-centered design and innovation trendiness (red flag 8). The client director or sales lead asks everybody to complete the early itemized estimation sheet and to come up with an alternative, and a more lean approach to be able to provide a price range that helps the issuer decide more quickly.
The whole process took just under 5 working days and everyone is happy with the results. Before it goes out, the leadership team approves a 10% discount to express the excitement of the whole agency and to open up the possibility to start working on this redesign asap (red flag 9).
All this not only sounds familiar, it actually is. At creative agencies, it is common to have a sales pipeline in place that streamlines and formalizes the creation of first responses, Q&A sessions, initial proposals, pitch decks, and related sales material. This process is mostly a reflection of the agency’s business model. But what if this way of operating is outdated, and does not work on complex digital design and business problems anymore?
Dismantling hidden estimation traps
In this section, we will try to better understand what makes this approach of estimating design as illustrated above so prone to errors and dismantle some of the hidden traps.
Trap 1: Sales- or executive team decide which RFP’s to respond to
While we all love to keep our jobs and are grateful for anyone that knocks at the door to request our help and expertise, there is still a little bit of reflection needed to respond with confidence and honesty. Some say you cannot polish a turd, and occasionally it’s just better to walk away from a fundamentally flawed and poorly defined project brief.
Most of the time though, the gate-keepers of this stage do not reach out to the people who later need to execute the requests of a future client and oftentimes fail to accurately interpret what the client is actually asking for (or really needs at this stage). And so, your execs agree to take on the new design challenge, despite all the traps, vague descriptions, and holes baked into the issuers RFP.
Trap 2: Clients already know the solution and how much it will cost to get there
I acknowledge the fact that there are senior people and true experts on the client side who are fully capable of understanding their needs, design complexity, and cost. I have worked not only on the agency side but also on the client-side for many, many years. Truth is, departments get a certain budget they can spend and are allocating money towards certain initiatives – e.g. let’s redo our website – which is not a straightforward process.
Departments often have to work with a given bucket which you can then see reflected in the RFP that lands on a vendor’s table. Obviously, clients fish for contractors with big, but more often with smaller worms. It is really hard to spot this upfront. And just because, someone suggested making it look like Apple, the process to get there isn’t more clear or straightforward as it was before.
Trap 3: First client call
Talking to potential new clients is an exciting opportunity and you can build up trust and show off design maturity, expertise in a field or market, strategic thinking, and your teams' motivation to take on this job. These early chit-chats are often kept light and vendors are afraid of asking difficult questions that unveil deeper, underlying business problems. Who wants to hear that it’s not really your website, but rather the fact that your department fails to manage and to update it frequently and appropriately. You even suck at using Google Analytics tbh. Also, sales and client management people rarely know what causes a design project budget to explode and what questions could surface these triggers.
Trap 4: Getting access to information, documents, and people
This is getting better and better and clients often do give you access to quite a lot of material you can and should look at before drawing any haphazard conclusions about their overarching goals, where they stand as organizations (skills, processes, products, etc.), their design maturity level, how well they understand the new digital markets, their own products and the people who use it.
There is often a tremendous gap between what the company thinks customers want, and what they actually do. Especially out there, in the wild internet. To cover that consumer insight gap, you often have to walk the extra mile, to bring everything up to date and up to standard, especially when you promised to deliver a human-centered design approach. Is that work reflected in your response or follow up SOW? Probably not.
Don’t try to commoditize design. Design is not a linear process and certainly not a process that’s repeatable. You cannot arrive at unique design solutions, by being repetitive; otherwise, you might as well just sell off-the-shelf finished products, rather than design as a service and a value creation exercise.
Trap 5: Assessing the current state of the product
Your client asked you to redesign their logo. You know exactly the dimensions and playing field of this design project. But how about an enterprise software or website? How do you easily and quickly assess the scope of such a complex product or digital service? What is attached to it, what features are an integral part of it and which ones aren’t? Is this content still part of the website redesign or will it be scrapped? The client wasn’t even aware that this section of the app or website is still accessible and online. So on and so forth…
At this stage, you can confidently assume that you were only able to look at 70% or less of the ‘product’ or features you need to redesign at a later stage. Are you factoring this in? If not, you are losing another 30% of design estimation accuracy right here.
Trap 6: Benchmarking against past projects
It’s good practice to look at past projects that had a similar scope to understand your teams’ velocity and get a breakdown of costs, tasks, and timeframes. The one-piece you cannot extract from your project archive, but slowed down or speed things up the most last time around is stakeholders. How responsive were they, how quickly did they agree on group decisions, how much did they contribute to finding the right design solutions? This is all the good stuff nobody remembers or tracks, but which can make the design process easy or hard, pleasant or unpleasant.
Trap 7: Strategy drives the RFP response
Strategy can be powerful. Product and services that are built with and around insight-driven strategies are more successful than the ones without clear guidance and structure. The risk here is that strategic frameworks take over the design. More so, product strategy gets tangled up with business, brand, marketing-, communication-, digital-, sales-, product-, platform-, service-, UX-, CX-, you-name-it-strategy. The list goes on and on and nobody knows at this stage what should and what should not be part of the design project at hand; what leads to the most desirable, viable, and feasible design solution that is economical to built and manage for the years to come.
At this early phase, you don’t even know what product features are tied to internal and external business workflows, partner networks, value, and supply chains, respectively the wider how and why. So what always ends up happening is that any strategy that gets added at this stage needs to be vague and generic. Instead, you should focus on the approach, the people, and skills you need, and outline the activities that will move you towards the first design milestone and then the next. Estimate your teams' velocity, not concrete deliverables, and factor in increasingly volatile and dynamic markets, that force you to change directions at any time.
Trap 8: New or junior members add inappropriate design activities
All designers and design agencies face competition and the internet has certainly shifted the market dynamics where now software quality and production speed reshaped most design methodologies, tools, tasks, and outcomes. Hello Design Thinking – where have you been all these years? Be cautious and understand what design steps need to be taken, what design methodologies need to be applied and charged for, before handing over your proposal to a new potential client. There is no need to kill the velocity of your design team with unnecessary steps when design solutions can be reached in a more lean and elegant way. There is a place for design thinking workshops, as long as the outcome does not end up on the corridor wall on the way to the washrooms, like so many persona posters, experience maps, service blueprints, and customer value maps often do. Change management, customer development, and product design should be estimated and sold separately.
Demonstrate innovation and design thinking not by doing more or less of the same things. Do things differently.
Trap 9: Adding unjustified discounts
I know I can expect a discount when I buy 10 of the same T-shirts, a soon to expire yogurt, or a cheap piece of tech that needs to leave the warehouse at BestBuy on Boxing Day, but how can we really justify a price reduction on a complex project on which six people work full time over a period of, say, four to six months? The more important question you have to answer here is probably; where and how do you compensate for that loss of income? After all, you run a people business and someone, not something has to pay for that. Because of that discount you offered, the work the client asks you to do is not being reduced, but you are certainly increasing the risk of project failure.
When discounting your work, agree on different, more flexible terms to give your ‘discounted team’ more freedom to complete the same amount of work with the same design quality. If you are in for a bidding war, you don’t win with discounts, you win with the value your team can bring to the table. Add value – don’t diminish it.
Trap 10 (add-on): The list of assumptions is too long
It seems too common that vendors who are in a tender process add assumptions to the proposal that tries to better frame the scope of the project and avoid dealing with too many surprises further down the road. This list just shows that you need to do your homework better and uncover what is truly needed more accurately; that is, what’s the situation, what are the main tasks, what actions to take to solve these tasks best, and what realistic results can we expect to achieve within a given context and timelines. Every item scratched from this list of assumptions will add precision to your design estimates. Do the homework upfront so you don’t pay the duties later on or pass these down to your teams. It kills the morale and motivation of your colleagues and that of clients, too. You certainly don’t want to go back and renegotiate a higher price tag if something could have been foreseen just by doing a bit more research, analysis, or client workshopping upfront. That said, perhaps we need to rethink the way we estimate complex design projects in general. The next section offers some high-level thoughts on this.
A new approach for design estimation
Approach A: Peer-Estimation
People have already written tons and tons and tons of articles and research papers on predicting the cost for design, so I am always puzzled why most design agencies who work in the digital space are not adopting these superior qualitative and quantitative approaches. The answer is most likely that we don’t have the time for it –and– we can always change things on the fly. It’s the internet, right?
Well, what if you see cost estimation as part of the planning process; documenting your and your clients' path to success. Now, that sounds like a true investment everyone can buy-into, and ideally, this planning takes place together with the client, not in separate rooms. How can you make this happen, given the status quo of the industry? The answer to that leads us to the second approach I’d like to discuss with you.
Approach B: Simulation
Two separate design teams can arrive at the same design solution, taking different paths, working with varying time constraints, tools, methodologies, or skill sets. You can imagine ending up with a huge cost range, for perhaps an identical product and project outcome.
As I stated before, design describes a process and the best way to understand and predict the costs of such a process is to simulate it with all the variables that influence your teams and your clients' performance along the way. After all, your client is heavily responsible for your performance throughout this design process that can take years for certain product categories. And when I say ‘simulate’ – it’s not just adding roles, hours, and tasks in your go-to estimation sheet or overly simplified Gantt chart. Break the mold and take the time to convince your future client and offer a free full-day problem-solution definition workshop. It will certainly pay off further down the road.
So for software and application UX design, you generally want to plan and simulate with these main variables outlined below (in no specific order):
- Product complexity and feature set – E.g.: A website page template can have little functionality and a lot, very complex feature sets built-in with multiple, dynamic views and states.
- Product dependency on e.g. content, tech, etc. – E.g.: You are working on a website redesign that has outdated content that also does not match modern content consumption patterns. Your new design relies heavily on someone else work and output to levels up the standard of this content.
- Your clients' design maturity level and design expertise – E.g.: Every time you present a new piece of design you need to go back to the basics and explain the rationale behind it.
- Your teams’ design knowledge and experience – E.g.: You need to use a new tool or work with inexperienced colleagues, but still need to deliver the same quality and outcome.
- Design collaboration & workflows – E.g.: The design process involves many teams from different departments on both sides, but all of them have different mindsets, approaches, and perhaps are at different stages. This holds you back and costs you extra time as well as adding internal meetings, meetings, and more meetings.
- Information accessibility and insight – E.g.: Your client or team members did not know, you (designers) could have benefited from a piece of information upfront.
- Access and knowledge about end-users and markets – E.g.: You were left in the dark who will actually most use and benefit from the product you designed. This could have helped you to better prioritize your work and the features you spent most of your time on.
- Your experience level – E.g.: You felt overly confident in your own abilities and also wanted to be polite and help your team meet the requested budgets the issuer outlined in their brief. Note – inexperienced designers estimate lower than experienced designers.
These are just the main variables that can drive up design effort and overall costs significantly. I am sure, you can probably add ten or twenty more, but next time you predict price tags and ranges for your UX design services, keep at least the ones identified above in mind.
Remember, the real task of a UX/CX designer is to change perception, attitudes, and behavior: The experience of a website, the usage of an important new application feature, the image of an organization, the handling of a customer service request, and so on. The work you do has a great impact and goes way beyond the more frequently cited design deliverables that most often make it into an estimation sheet, while those more important ones are being left unframed and unnoted.
