Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Des Traynor – "Creating Dashboards & Data Visualisations that Resonate"

beyond tellerrand
January 10, 2012

Des Traynor – "Creating Dashboards & Data Visualisations that Resonate"

This talk was held on the "beyond tellerrand – web" conference from November 20th to 22nd, 2011 in Düsseldorf.

Description:

The world is drowning in data. It always has been. Making sense of that data and deriving meaningful insights has been left to professional statisticians. Up until recently that is. Info-graphics & data visualisations are popping up everywhere and they don't always make sense. Many horrible acts have been committed in the name of "Making this chart seem a bit more fun"
Getting the right data, and then getting the data right is a significant design challenge that touches on more domains than a UX designer will initially be familiar with.
If you intend to work with large data sets to answer questions about your company or application, or if you are tasked with summarising large amounts of data, or even if you've just been told to "design the new dashboard" then this is the session for you.

Video of the talk:
http://vimeo.com/34784156

beyond tellerrand

January 10, 2012
Tweet

More Decks by beyond tellerrand

Other Decks in Design

Transcript

  1. PIC  MAG TRO KNOW YR AUDICE KNOW YR DOMA

    KNOW YR DATA KNOW YR VUALS KNOW YR Y CLOSG POTS F TRO
  2. Be clear first and clever second. If you have to

    throw one of those out, throw out clever. — Jason Fried “
  3. If the Gulf of Mexico - the 7th largest body

    of water in the world, containing approximately 660 quadrillion gallons of water (that's 660 with 15 zeros) - was represented by Cowboys Stadium in Dallas - the largest domed stadium in the world - how would the spill stack up? In this example, the amount of oil spilled - if the Gulf of Mexico was the size of Cowboys Stadium - would be about the size of a 24 ounce can of beer. Cowboys stadium has an internal volume of approximately 104 million cubic feet, compared to the just over 50 cubic inches of volume in a 24-ounce can. Just like the can, the spilled oil represents only . 00000002788% of the liquid volume present in the Gulf of Mexico, although as the oil is dispersed, the amount of water affected becomes substantially greater.
  4. If the Gulf of Mexico - the 7th largest body

    of water in the world, containing approximately 660 quadrillion gallons of water (that's 660 with 15 zeros) - was represented by Cowboys Stadium in Dallas - the largest domed stadium in the world - how would the spill stack up? In this example, the amount of oil spilled - if the Gulf of Mexico was the size of Cowboys Stadium - would be about the size of a 24 ounce can of beer. Cowboys stadium has an internal volume of approximately 104 million cubic feet, compared to the just over 50 cubic inches of volume in a 24-ounce can. Just like the can, the spilled oil represents only . 00000002788% of the liquid volume present in the Gulf of Mexico, although as the oil is dispersed, the amount of water affected becomes substantially greater.
  5. PIC  MAG TRO KNOW YR AUDICE KNOW YR DOMA

    KNOW YR DATA KNOW YR VUALS KNOW YR Y CLOSG POTS F KNOW YR AUDICE
  6. WHO IS THE INFO FOR? Surpre surpre,  UX du

     talkg abt knowg yr uss.
  7. CEO Lev tail Stregic view Focus   lg tm

    High v ovview Simp suy
  8. * Qry driv alys * Psn quid * Emps 

    ts,  cs Analyst ro
  9. SALES MARKETING CUSTOMER SUPPORT MANAGEMENT * Satisfaction Rating * Trend

    per quarter * Comparison with competitors ANALYST * My Active leads * Value per lead * Progress towards target OPERATIONS * Active campaigns * Current CPM/CPC * Landing page Role + Department = Information needed
  10. PIC  MAG TRO KNOW YR AUDICE KNOW YR DOMA

    KNOW YR DATA KNOW YR VUALS KNOW YR Y CLOSG POTS F KNOW YR DATA
  11. $ Sas day # Un sas Avg $ p sa

    Th pd vs st pd Us vs Comper Total  m Popur products % Cnge  sas Avg. $ p custom WHICH OF THESE?
  12. WHICH OF THESE? TOTAL SALES $12,240.65 CHANGE 5.32% Top grossing

    items % TOTAL REV. 10 20 30 40 100 200 300 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 400 500 1 2 3 4 5 6 7 8 9 10 Top selling items Item name Unit sales % of total Oak tree (special edition) 803 16% Pet Kitten 607 12% Skyscraper (high rise) 511 11% Sycamore tree 430 9% Dancing disco. 203 4% Other items 2495 52% Change 11.52% 100% 1.52% 5.23% 1.20% -- 100 200 300 400 500 Ybarra Bow Broadsword Dagger Eclipse Mace BattleAxe Magic wand Crossbow Poison Revenue 5 10 15 20 25 Unit sales
  13. 1. COMMUNICATE A SINGLE FIGURE Us wn ctext  obvus,

    psn  quid,  pt/ futu  ievt  us. BALANCE $23.00 BALANCE $11.32 BALANCE $11.32 Examps:  crk wh a wag lt Cckg bk bae Sys adm cckg cut stus Not: Sg numbs c ve st (gd,bad etc)
  14. 2. SINGLE FIGURE WITH CONTEXT “How e we dog ty?

    Any probms  hiz?” Examps: How we  ms sas? Is  netwk pfmg wl? Hows r us figus lkg? Not: Spk-l c save space,  provi ctext READERS 12,247 CHANGE 0.32% READERS 15,231 CHANGE 9.52%
  15. 3. ANALYSIS OF A PERIOD “Show me a  key

    momts  m” Examps: Lkg f ptns  lg da sets Lkg aad b  cut da Comp wh pvus pd
  16. 10 20 30 11 12 13 14 15 16 17

    18 19 20 21 22 23 24 25 26 27 28 29 30 31 40 50 1 2 3 4 5 6 7 8 9 10 Wk bt wh pse da (e.g. day  day) GOOD LINE CHART
  17. 4. ANALYSIS OF PERIOD, WITH TARGET Did we h r

    sas figus? A we fulfig r five n quota? Examps: A sas we y shld be? A a r employs pfmg okay? Is r spse me be  dustry std?
  18. 0 12500 25000 37500 50000 Jan Feb Mar April May

    June July August September October November December Actual Tget BAD LINE CHART
  19. A co r  vualn  avg a  procsg

      ar. At a ge  lks like we’ dog okay . In  ce, we’ talkg abt a lta, but we’ not showg  lta...
  20. -40% -30% -20% -10% 0% 10% 20% Jan Feb Mar

    April May June July August September October November December FOCUS ON THE DELTA
  21. 5. BREAKDOWN OF A VARIABLE “Wt age grps e buyg

    r stuff? Wt cri e we big ?” Examps: Who e r customs? Wts r aws like  ch mographic? Wt browss e peop usg se days?
  22. 5. BREAKDOWN OF A VARIABLE “B cs ’t sexy, but

    y ly   ne ski, foowg a le. ”
  23. 6. BREAKDOWN OVER TIME “How s  compn cng ov

     st yr?” Examps: How s  brows mket cng? H r v src shift ctly?
  24. 0 17500 35000 52500 70000 Jan Feb Mar April May

    June July August September October November December I U.K Amica STACKED BAR CHART
  25. 0 17500 35000 52500 70000 Jan Feb Mar April May

    June July August September October November December STACKED BAR CHART A-"*&,# 6"34"/ &$ J789? America peaked in July?
  26. 0 17500 35000 52500 70000 Jan Feb Mar April May

    June July August September October November December STACKED BAR CHART A-"*&,# 6"34"/ &$ J789? How has U.K . done?
  27. 0 17500 35000 52500 70000 Jan Feb Mar April May

    June July August September October November December LYING WITH DIMENSIONS Lots more yellow pixels here now...
  28. LET ’S TRY A LINE CHART 0 12500 25000 37500

    50000 Jan Feb Mar April May June July August September October November December I U.K Amica
  29. LINE CHART OF SAME DATA? 0 12500 25000 37500 50000

    Jan Feb Mar April May June July August September October November December A-"*&,# ($ 2!" :6. ;K $"."* <!#$)"/? Same data. Different story.
  30. 0 12500 25000 37500 50000 Jan Feb Mar April May

    June July August September October November December I U.K Amica BAR CHARTS AGAIN?
  31. 0 12500 25000 37500 50000 Jan Feb Mar April May

    June July August September October November December I U.K Amica BAR CHARTS AGAIN?
  32. 0 12500 25000 37500 50000 Jan Feb Mar April May

    June July August September October November December I U.K Amica BAR CHARTS AGAIN?
  33. 0 12500 25000 37500 50000 Jan Feb Mar April May

    June July August September October November December I U.K Amica BAR CHARTS AGAIN?
  34. 0 12500 25000 37500 50000 Jan Feb Mar April May

    June July August September October November December INTERACTIVE, REMEMBER? You can adapt based on Interctions
  35. 0 17500 35000 52500 70000 Jan Feb Mar April May

    June July August September October November December STACKED BAR CHART Why is it so hard to follow the U.K here?
  36. PIC  MAG TRO KNOW YR AUDICE KNOW YR DOMA

    KNOW YR DATA KNOW YR VUALS KNOW YR Y CLOSG POTS F KNOW YR VUALS
  37. Y’ve just tak ov a hot. Y’   as.

    Exc . W do we stt? HOW TO USE ALL THIS?
  38. Q: A we makg y mey? Pr   dtce

    betwn cts  v. Let’s s  f  yr. -€9,000.00 -€6,750.00 -€4,500.00 -€2,250.00 €0 €2,250.00 €4,500.00 €6,750.00 €9,000.00 Jan Feb Mar April May June July August September October November December Profit and loss
  39. Q: Wt s do  mey come from? Rms Wdgs

    Cfc Buss Rtaurt B Gym/Spa 10% 20% 30% 40% 50% Let’s compe  pctage  v g by ch cegy.
  40. King suite Junior Suite Standard Room Hostel €50 €75 €100

    €150 €175 Deluxe Room Q: Wt st  pric do we crge p rm? Let’s lk   price rge  mn val
  41. REPORT REVENUE TYPE ROOMS & EXTRAS ROOM TYPE KING SUITES

    PERIOD LAST YEAR MIDWEST HOTELZ PROFIT LOYALTY INCIDENTALS GUEST REPORT WEDDINGS CONFERENCES GUEST TYPE ALL GUESTS Dign  sut alyst qri...
  42. PIC  MAG TRO KNOW YR AUDICE KNOW YR DOMA

    KNOW YR DATA KNOW YR VUALS KNOW YR Y CLOSG POTS F KNOW YR Y
  43. Top products Product Orders $ Revenue Books Electronics Magazines Appliances

    e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON
  44. Top products Product Orders $ Revenue Books Electronics Magazines Appliances

    e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON Let’s use this strawman
  45. Smat Effecve Diffce:  t y c do  highlight

    These colours would get very loud. Unnecessarily so.
  46. Top products Product Orders $ Revenue The girl with the

    dragon tattoo 11 88.50 Inception 9 72.50 The girl who kicked the hornet's nest 15 54.05 Books Electronics Magazines Appliances e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON Gradients, shadows, colors, gridlines. All non-content
  47. Top products Product Orders $ Revenue The girl with the

    dragon tattoo 11 88.50 Inception 9 72.50 The girl who kicked the hornet's nest 15 54.05 Books Electronics Magazines Appliances e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON Let’s kill the gradients
  48. Top products Product Orders $ Revenue The girl with the

    dragon tattoo 11 88.50 Inception 9 72.50 The girl who kicked the hornet's nest 15 54.05 Books Electronics Magazines Appliances e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON Let’s kill the colours
  49. HTML has a <strong>tag but no <weak> tag. As a

    result, we forget to think about what’s less important on the screen. — Ryan Singer
  50. Top products Product Orders $ Revenue The girl with the

    dragon tattoo 11 88.50 Inception 9 72.50 The girl who kicked the hornet's nest 15 54.05 Books Electronics Magazines Appliances e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON Let’s adjust the shading.
  51. Top products Product Orders $ Revenue The girl with the

    dragon tattoo 11 88.50 Inception 9 72.50 The girl who kicked the hornet's nest 15 54.05 Books Electronics Magazines Appliances e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON Let’s add the necessary differences
  52. Top products Product Orders $ Revenue The girl with the

    dragon tattoo 11 88.50 Inception 9 72.50 Books Electronics Magazines Appliances e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON From here we could begin to style
  53. Top products Product Orders $ Revenue The girl with the

    dragon tattoo 11 88.50 Inception 9 72.50 The girl who kicked the hornet's nest 15 54.05 Books Electronics Magazines Appliances e-goods Other 10 20 30 40 Revenue per product Sales Report Jan 2012 ORDERS 12,247 CHANGE 0.32% ACCOUNTS 7,343 CHANGE 4.32% SITE LIVE PAYMENT LIVE FULFILLMENT ON
  54. PIC  MAG TRO KNOW YR AUDICE KNOW YR DOMA

    KNOW YR DATA KNOW YR VUALS KNOW YR Y CLOSG POTS F CLOSG POTS
  55. 1. THEY HAVE TO SAY SOMETHING T wst vualns e

      y lk  just k “Heh.”
  56. 3. PRESENTING AN ARGUMENT It’s okay  a vuals if

    yr goal  me  htly psng fo
  57. The world is not filled with professional statisticians. Many of

    us would like a quick glance just to get a good idea of something. If a graph is made easier to understand by such irrelevancies as a pile of oil cans or cars, then I say all the better. — Don Norman
  58. 0 5 10 15 J Feb M Apr May Jun

    Jul Aug Sep Oct Nov Dec Get your data first.
  59. Usability is not everything. If usability engineers designed a nightclub,

    it would be clean, quiet, brightly lit, with lots of places to sit down, plenty of bartenders, menus written in 18-point sans-serif, and easy-to-find bathrooms. But nobody would be there. They would all be down the street at Coyote Ugly pouring beer on each other. — Joel Spolsky
  60. 4. THEY’RE NOT ALL FIRST TIMERS Like c pys ust

    cbods, peop c n  ust vualns
  61. Bks Stepn Few - “Dhbod Dign” & “Now y s

    ” Brn Suda - “Digng wh Da” Edwd Tufte - T first two. Blogs Stepn Few -> hp://pceptuadge.com Icom (me) -> hp://blog.tcom.
  62. PIC  MAG TRO KNOW YR AUDICE KNOW YR DOMA

    KNOW YR DATA KNOW YR VUALS KNOW YR Y CLOSG POTS F F