How can we establish the best design under technological constraint ?

19 Jun 2011 - 7:03pm
5 years ago
5 replies
2118 reads
Agradoot Ghatak

I am working with an approach of solving strategic design decision under various constraints. Usually there is always a challenge for design managers to establish a design concept however revolutionary it considered. Often these revolutionary designs can not be fitted in to the scheme of implementation due to various constraints such as technology revamping, large amount of framework re-engineering, engaging cutting edge skill workers etc. Finally these outstanding pieces of creative effort become an overhead on budget. Worrying about the ROI, we eventually compromise with a mediocre or substandard design.  

Can we proposes a scientific method of choosing the best design out of finite set of  concepts, that fits best under the limits of affordance.  Can we offer a scientific model to avoid this practical deadlock?


19 Jun 2011 - 9:05pm
Steve Baty

There are a few points to this question worth drawing out separately.

Working with constraints

Design is an exercise in working within, and challenging, constraints. Some constraints are very fixed and cannot be shifted. These might be due to the available technology - not currently within your organisation, but available anywhere - or the limits of the materials with which you have to work. Other constraints are fixed for economic reasons - that is, the ability of the organisation to afford a change; or the ability (or willingness) of the market to pay a higher price that would result from such a change. Still other constraints are simply assumed to be fixed through habit, belief, or myth. These latter constraints are often where design, and the design process, can provide a great deal of value. 

Throughout a design process, challenging assumptions should be a high priority. Asking whether some 'truth' actually holds, and testing that with our intended audience/market and against our internal capabilities. Do customers really not care about colour? Do customers really make purchasing decisions solely on functionality or technical specifications? Is the market really price sensitive? Does it really take 3 days to process and approve an application for a credit card?

It is also important to recognise that a 'revolutionary' design may or may not require a radical shift in capability for an organisation. Innovative ways of working within existing constraints can yield surprisingly high returns - although obviously not always.

From a design process perspective, the existence of a technology or capability constraint shouldn't come as a surprise to the design team when they're delivering their final designs. Early stages of the design process should include a clear understanding of existing constraints, and the challenge of those constraints. In fact, one very accessible source of innovation for a design team is to ask: "What if x constraint didn't hold?" and exploring the direction that leads.


It is especially difficult to demonstrate a clear ROI for a breakthrough product/service concept. You can't easily estimate the market impact of your concept, and speculation doesn't help make a business case.

A case can be made, however, for design-as-risk-mitigation strategy within any project. The design approach, with its inherent acceptance of multiplicity of ideas, and the testing and refinement of those most promising ideas over the course of the process, provides organisations with an increased confidence that the design is heading in the 'right' direction; and that the execution of that design is a good one. 

Risk mitigation, from a business perspective, is hugely important as the project reaches points of increased investment in an idea. Design's ability to point back to a multitude of 'failed' ideas demonstrates the strength of the few ideas that the team is continuing to explore. Gradually increased investment - in prototyping and testing - minimises the risk to the organisation of over-investing in the wrong idea, or a bad execution of a good idea.

A scientific approach to evaluating designs

This idea feels incredibly fraught to me with the potential for de-humanising the design process. In many respects, design works best as a qualitative exercise for as long as possible. That said, a more 'scientific' (which I am taking to mean quantitative) evaluation of a design makes sense. After all, a design is only really successful if people are willing to use it and, in a business context, pay for it. The question is really: At what point does it make sense to conduct such evaluations? I would propose three options:

i) A/B testing of a new design against an existing one
ii) A pilot program in some limited subset of your overall channel; or
iii) A public 'beta' program (i.e. launch it and see).

All three options share a common trait: they all involve the target audience/market in the evaluation in a 'live' sense. This is really the only viable 'litmus' test for a design concept. This is not to say that review and critique aren't valuable - they absolutely are. Design critique is an essential part of the process of refining, enhancing, and pushing the boundaries of a design concept. But that won't win over the business, which I think is the point you're after.

So, the short answer to your question is no, I don't believe a scientific method of choosing the best design that fits best 'under the limits of affordance' is a viable option. I think the question pre-supposes an economic constraint which in itself needs to be challenged. And, if it has been challenged and shown to be fixed, then that constraint should have formed the basis for your designs from the outset. In other words, I think the problem you're attempting to solve is not the right problem. Rather, how can we identify economic constraints early enough in the design process such that we can work within them; show them to not hold if indeed they do not; whilst not unduly narrowing our creative options in the process, thereby missing opportunities for 'revolutionary' designs along the way.


19 Jun 2011 - 10:26pm
Agradoot Ghatak

Thanks Steve. Thank you for your analisys about constraints and how to handle with those contrains . This realy opens a lot more thoughts. But I am trying to solve in a bit logical rather being subjective. Please share your thoughts.. Again thanks a lot. 

19 Jun 2011 - 10:18pm
Agradoot Ghatak

Usually in the stage of a design process, conceptualization happens much later than the behavioural research, contextual inquiries, and field researches etc.  This conceptualization is always iterative. Some organizations continually spend more time on research to ensure their designs align to the objective. Indeed whatever SDLC model they adopt, creating a design concept always happens first, presented later to the cross functional teams who actually facilitate transformation of the concept into a tangible product. 

Now the question arises:

1- How can a constraint be defined before conceptualization of a design?  Assume it could be identified partially; do all constraints discovered at the level of conceptualizing a design? 

2- Do designers aware of these technological constraints before putting his/her pen into conceptualization?  As none of us expect a designer to be a master in technology, therefore it is practically impossible to know about the whole framework by a designer then work – ‘within certain constraints’ in parallel.  

3- In a given circumstance if a design fails to establish due to those constraints, can we able to establish a measuring method in place that would quantify the loss for expelling it?

The answer could be putted in a manner saying about participatory / collaborative conceptualization so that the designer aware of the technical or economical difficulties about his proposed design. Now this opens two question –

1) Are we taking a defensive design approach (Read: Strategically) compromising with economic and technical constraint? If so, can we draw a quantitative metrics about the amount of USER EXPERIENCE suffers from its threshold benchmark of optimum experience?  I am here trying to draw a safe zone and the amount of deviation a designer has to take because of those constraints. Can we use any mathematical model to frame it ?
2) The second question is rather behavioural - How do designers drive (or fight) for negotiating with those constraints in order to shape the product? Since design is subjective, could have many solutions, many approach. Can we first create usability metric before applying these constraints and then later apply those constraints and find out the amount of dispersion? 

20 Jun 2011 - 3:55am
Yohan Creemers

The only valid constraints before conceptualization are the constraints defining the strategic goal and scope of the project. E.g., is the project about making short term profit - or about setting a new industry standard. In daily practice many projects also start with technological constraints - unnecessarily hampering the conceptualization. In the worst case these constraints are not stated explicitly, but taken for granted implicitly by designers who are too much aware of technical constraints.

As Steve said: designing is challenging constraints. To play a fair game, the constraints must be made explicit. Not in a parallel process, but iterative, the design challenging the constraints, and vice versa.

Answering your last two questions:

1) Whether to take a defensive approach depends on the strategic goal and scope of the project. You will have to bring all variables under the same denominator If you want to measure the degree of compromise. The common denominator probably being dollars, euros or yen. I guess this is possible, but in my opinion useless. Calculating the value of anything is subjective too.

2) If everyone on the project endorses the strategic goal, and all team members respect each other's expertise and responsibility, then there is no reason to fight. Good professionals love to be challenged. Indeed, there are many solutions to a design problem, making it possible to search for an optimum between functional (ux), technological and economic constraints. Somehow it seems easier to capture technological and economic constraints. But yes, one can and should also capture the requirements for usefulness, usability and desirability in order to evaluate the design.

- Yohan

20 Jun 2011 - 11:26am

I'd like to add a few points here:

  • Be careful of relying too much on numbers and metrics, you get a lot of false accuracy.  Subjectiveness can be objectified using surveys, etc. but statistical accuracy is expensive, especially because you'll need to do this both wide (many tests) and deep (many people).  Most orgs can't cost justify this.
  • How to deal with "strategic decision making" depends upon what you mean by it, and how big and how revolutionary it is.  My terminology: if you mean a cool new table design, then that's tactical design.  If you mean "the next iPad", then that's strategic design.  I assume you mean the latter.
  • Strategic design and tactical design need to be treated differently.  Most people understand the latter (the output being concrete designs, and you can do user testing, etc.).  Strategic design, I found, more relevant to enterprise-class complex software, with lots of inertia and technical and/or design debt  (e.g. existing technology or designs that were done poorly).
  • With strategic design, the key is how far out you are looking and understanding that the design is only as good as the understanding of the functional problems that it solves (e.g. "requirements", although how you can come up with good requirements quantitatively is questionable).  The problem is that such "requirements" changes with time.
  • I deal with the above using a multi-phase, multi-work stream approach.  The farthest streams uses more marketing or general user studies, rather than prototype user testing.  Mostly I use a simple NPS type of metrics to track general direction.




Syndicate content Get the feed