Anticipatory Design: The Future of UX Design

The Industrial Revolutions of 1760 and 1820-1840 were com­plete game chang­ers. An eco­nomic and so­cial or­der thou­sands of years in the mak­ing was re­made. The econ­omy, so­ci­ety, cul­ture—every­thing be­came cen­tred around in­dus­try.

Today we are liv­ing through a data rev­o­lu­tion that’s no less im­por­tant than the Industrial Revolutions. In this Information Age”, data has taken in­dus­try’s place at the cen­tre of all things.

Every day data is col­lected and analysed. Collectively, Google, Amazon, Facebook and Microsoft have col­lected more than 1,000 petabytes (1,000,000,000 gi­ga­bytes) of data.

But what good is data? Are busi­nesses and gov­ern­ments col­lect­ing data with­out rhyme or rea­son? No. Data is dri­ving in­no­va­tion and change world­wide, es­pe­cially in user ex­pe­ri­ence (UX) de­sign.

Anticipatory Design

Perhaps the next big thing in the realm of UX is Anticipatory Design: de­sign that’s one step ahead and an­tic­i­pates end user be­hav­iour and pref­er­ences.

Imagine a web­site, which knew you liked read­ing re­views by in­dus­try ex­perts, au­to­mat­i­cally re­spond­ing to your in­cli­na­tions by link­ing to ex­pert re­views. Meanwhile, some­one else, who only cares about price, is shown price com­par­isons be­tween the site and its com­peti­tors. In both cases, the (presumably pos­i­tive) UX is the prod­uct of an an­tic­i­pated and re­spon­sive de­sign.

This is what Anticipatory Design is at its core: a process wherein data-dri­ven de­ci­sions are made on be­half of end users.

Data Driven Design: The Probability of Being Right/Wrong

Anticipatory Design is com­monly used to stream­line processes. You have al­most cer­tainly ex­pe­ri­enced this. When Spotify sug­gests a song, it’s not shoot­ing in the dark, it’s mak­ing an ed­u­cated guess based on your habits and his­tory. It’s an­tic­i­pat­ing what songs you could like.

Anticipatory Design has cousins in se­lec­tion sim­pli­fi­ca­tion and choice edit­ing. Sophie Kleber plots these ty­polo­gies of au­to­matic cu­ra­tion on a carte­sian plane, with one axis rep­re­sent­ing the prob­a­bil­ity of bring right and the other axis rep­re­sent­ing the cost of be­ing wrong.


Anticipatory Design comes with cer­tain risks, which is why it could be cor­rect to en­gage in se­lec­tion sim­pli­fi­ca­tion and choice edit­ing, or noth­ing at all in some cases, in­stead.

However, with the rise of the Internet of Things and ma­chine learn­ing, data and our abil­ity to analyse it is grow­ing at an awe­some rate. Simultaneously, peo­ple are be­com­ing in­creas­ingly ac­cus­tomed to big brother and more open about shar­ing per­sonal data when they see value in it.


We live in a world where ma­chines can beat hu­mans at board games like Chess and Go, and even at trivia games like Jeopardy. Machines can even recog­nise pat­terns in im­ages. For ex­am­ple, SearchInk can read hand­writ­ing. A more main­stream ex­am­ple of ma­chine learn­ing of im­agery is Facebook. Ever won­dered how Facebook can recog­nise your face? The an­swer is ma­chine learn­ing. Tag a face as yours enough times and Facebook will learn which pat­terns make up your face.

Couple the po­ten­tial of ma­chine learn­ing with the equally im­pres­sive po­ten­tial of IoT and Anticipatory Design seems in­evitable. The IoTs con­sists of 4.9 bil­lion con­nected things rang­ing from cars to ther­mostats. Both the num­ber and the range of con­nected things is in­creas­ing. A new sur­vey from Intel and Penn Schoen Berland sug­gests 70% of peo­ple would be will­ing to use a smart toi­let if it would mean lower health­care costs.

With more data and pro­cess­ing power, Anticipatory Design is be­com­ing less a mi­rage on the hori­zon and more a ques­tion of de­sign.

Download our White Paper and see how we com­bine UX de­sign and cloud net­work­ing.


Mitchell Tweedie

Get cu­rated con­tent on soft­ware de­vel­op­ment, straight to your in­box.

Your vi­sion,

our ex­pe­ri­ence

Book an analy­sis call