1. alok jain

    Sr. Interaction Manager at satyam infoway ltd

    27 July 2001 05:38am

    alok jain

    Hi all,

    I am looking for ROI calculation tools on usability (of web applications).

    cheers
    Alok Jain

  2. Alexander Karle

    Vice President, Business Strategy at Thomson

    27 July 2001 09:55am

    Alexander Karle

    ROI of usability or in other words improvements in the user experience of a web application has come to the forefront of the discussion in recent months. Especially when IT and online budgets have been either frozen or cut significantly, design efforts need to be based on measurable business objectives (not that they shouldn't have been based on ROI before...). Still, I currently experience two major hurdles with my clients when it comes to measuring the ROI of web application design:

    * Decisions on design improvement projects have traditionally been based on opinions on the "look and feel" of a design as judged by major internal stakeholders. ROI has almost never been considered in previous efforts, therefore there is a general lack of understanding and reluctance to admit that new initiatives should be based on ROI.

    * Even if clients state the need for ROI calculations, in many cases there is a general lack of data for current site performance, which (at least in the eyes of many web-site operators) makes it difficult to compare current performance/ROI with post-redesign site performance.

    Although I would argue that there is no GENERAL model for determining the ROI of a redesign effort, I believe that for most efforts there are multiple, straightforward ways of estimating an ROI based on reasonable assumptions and the tracking of key metrics. I believe that the major elements of an ROI approach are:

    0.1. Build a general apprehension of ROI as a design criteria in the organisation

    0.2. Start tracking core metrics of you current operations asap - simple tools like Webtrends are a good start, but you might want to get more sophisticated later on. If you don't track anything else, definitely do track where your site visitors come from, which pages they most frequently visit and from which page users currently leave your site

    1. Create scenarios of how you think your high-value users should interact with your site - based on your target user segments, define various different scenarios and then link those scenarios with your business objectives. I.e., for a music site, teen visitors interested in pop music trying to purchase products, techno fanatics looking for the latest clubbing information, etc.

    2. For each scenario, identify barriers on your current site to quickly reach a user's objective.

    3. Identify current metrics, i.e., current conversion rate for visitors is x%, x% of visitors abandon the site after doing a single search (i.e., they did not find what they were looking for), etc.

    4. Build a financial model that reflects key scenarios and links them with key business metrics. I.e., number of overall site visitors -> number of visitors opening a shopping cart -> number of visitors actually buying products -> average value of purchase

    5. Identify costs of redesigning parts of your site/one critical scenario or user path. Cost categories can include: User testing (field studies, user experience lab, expert evaluations, etc.); copywriting; purchase of improved content; costs for additional features (e.g., improved search engine); redesign of key features (e.g., shopping cart, product evaluation tools); estimates of design efforts need to be based on realistic development efforts (team size & costs/day)

    6. Based on your model developed in 4. and cost estimates in 5., estimate the ROI by altering your key parameters for a specific design effort/scenario (i.e., "what if" analysis - if my design project improved conversion rates by x% that would yield an ROI of Y)

    7. Prioritise multiple design efforts using outcomes of 6.

    I just wanted to reiterate that there is no general model for calculating the ROI of a specific design effort. Although key inputs (e.g., hourly rates for members of your design team, costs for user studies, etc.) are reusable, you will need to customise your model for every new design effort.

    Alex

  3. Paul Tinsley

    Creative Director at Agenda Solutions

    30 July 2001 10:22am

    Paul Tinsley

    As Alex says in his excellent post "Start tracking core metrics of your current operations asap".
    A good starting point would be the excellent white paper 'E-Metrics: Business Metrics for the New Economy'
    at:
    http://www.e-consultancy.co.uk/knowledge/whitepapers/whitepaper_view.asp?id=28

    On 09:55:05 27 July 2001 alexkarle wrote:
    >ROI of usability or in other words improvements in the
    >user experience of a web application has come to the
    >forefront of the discussion in recent months. Especially
    >when IT and online budgets have been either frozen or cut
    >significantly, design efforts need to be based on
    >measurable business objectives (not that they shouldn't
    >have been based on ROI before...). Still, I currently
    >experience two major hurdles with my clients when it comes
    >to measuring the ROI of web application design:
    >
    >* Decisions on design improvement projects have
    >traditionally been based on opinions on the "look and
    >feel" of a design as judged by major internal
    >stakeholders. ROI has almost never been considered in
    >previous efforts, therefore there is a general lack of
    >understanding and reluctance to admit that new initiatives
    >should be based on ROI.
    >
    >* Even if clients state the need for ROI calculations, in
    >many cases there is a general lack of data for current
    >site performance, which (at least in the eyes of many
    >web-site operators) makes it difficult to compare current
    >performance/ROI with post-redesign site performance.
    >
    >Although I would argue that there is no GENERAL model for
    >determining the ROI of a redesign effort, I believe that
    >for most efforts there are multiple, straightforward ways
    >of estimating an ROI based on reasonable assumptions and
    >the tracking of key metrics. I believe that the major
    >elements of an ROI approach are:
    >
    >0.1. Build a general apprehension of ROI as a design
    >criteria in the organisation
    >
    >0.2. Start tracking core metrics of you current operations
    >asap - simple tools like Webtrends are a good start, but
    >you might want to get more sophisticated later on. If you
    >don't track anything else, definitely do track where your
    >site visitors come from, which pages they most frequently
    >visit and from which page users currently leave your site
    >
    >1. Create scenarios of how you think your high-value users
    >should interact with your site - based on your target user
    >segments, define various different scenarios and then link
    >those scenarios with your business objectives. I.e., for a
    >music site, teen visitors interested in pop music trying
    >to purchase products, techno fanatics looking for the
    >latest clubbing information, etc.
    >
    >2. For each scenario, identify barriers on your current
    >site to quickly reach a user's objective.
    >
    >3. Identify current metrics, i.e., current conversion rate
    >for visitors is x%, x% of visitors abandon the site after
    >doing a single search (i.e., they did not find what they
    >were looking for), etc.
    >
    >4. Build a financial model that reflects key scenarios and
    >links them with key business metrics. I.e., number of
    >overall site visitors -> number of visitors opening a
    >shopping cart -> number of visitors actually buying
    >products -> average value of purchase
    >
    >5. Identify costs of redesigning parts of your site/one
    >critical scenario or user path. Cost categories can
    >include: User testing (field studies, user experience lab,
    >expert evaluations, etc.); copywriting; purchase of
    >improved content; costs for additional features (e.g.,
    >improved search engine); redesign of key features (e.g.,
    >shopping cart, product evaluation tools); estimates of
    >design efforts need to be based on realistic development
    >efforts (team size & costs/day)
    >
    >6. Based on your model developed in 4. and cost estimates
    >in 5., estimate the ROI by altering your key parameters
    >for a specific design effort/scenario (i.e., "what
    >if" analysis - if my design project improved
    >conversion rates by x% that would yield an ROI of Y)
    >
    >7. Prioritise multiple design efforts using outcomes of 6.
    >
    >I just wanted to reiterate that there is no general model
    >for calculating the ROI of a specific design effort.
    >Although key inputs (e.g., hourly rates for members of
    >your design team, costs for user studies, etc.) are
    >reusable, you will need to customise your model for every
    >new design effort.
    >
    >Alex
    >
    >
    >

  4. Ashley Friedlein Staff

    CEO at Econsultancy

    30 August 2001 18:05pm

    Ashley Friedlein

    Might be worth looking at the piece on usability in this month's newsletter which gives some (bullish) figures on usability ROI. It's at http://www.e-consultancy.com/newsfeatures/newsletter/view.asp?id=126#1539

Reply to this thread

Log in to reply to this thread or join Econsultancy for free so you can post to our forums along with other benefits.