Do #eCommerce #analytics prove #customer #affiliation?

Analytics for e-Commerce can be split between two contrasting and not necessarily companion aspects.

  • analytics of success in terms of market development, profile or response through hit or link ratings
  • analytics of sales or revenue

Dependant upon the purpose of a company’s e-commerce deployment one or other but rarely both of these analytics will seek to prove the status of the investment and allow informed strategic decision making.

According to the United Kingdoms Office for National Statistics (ONS) in 2002 “e-commerce is likely to have a huge impact on the way we do business. It has the potential to lead to dramatic growth in trade, increase markets, improve efficiency, effectiveness and transform business processes”. This notion of increasing markets and improving efficiency and effectiveness through internet technology is underwritten in ‘Blown to Bits: How the New Economics of Information Transforms Strategy’. (Evens, 1997). Here the authors determine that three key factors create success in e-commerce businesses: reach, richness and affiliation.

  • Reach indicates numbers of users interacting in e-commerce.
  • Richness is the depth or quality of information user’s gain.
  • Affiliation is a measure of correlation or affinity with users.

The internet’s ease of accessibility for consumers means there is a levelling of reach between small and larger enterprises. This is due to small companies being able to gain the sort of competitive advantage associated with advertising in other media such as newspapers or television. Richness or detail of information is also easier to provide online with multiple data sources, animations, video and databases linked to internet and extranet applications. Only affiliation with the customer is unchanged and this creates a greater effectiveness and success through tailored services such as the Amazon personal front page (Wolverton, 2000) by making a new store for each user. It is in an attempt to measure affiliation or user buy in, that many commercially based internet statistics are created. This pressure to create statistics often to prove the effectiveness of the method by companies with a vested interest renders many internet statistics unreliable . To quantify affiliation and supposedly shed light on the nature of business transformation, increased markets and improved efficiency of e-commerce as a marketing channel numerous calculation methods are in use.

Credibility and Trustworthiness

There are two distinct terms associated with the effectual usage and profitability of e-commerce, credibility and trustworthiness. James Rosenfields (2001) determined that much of the available internet statistics are “lies or damned lies”. This is because they are “an amalgam of guesswork, wishful thinking, pie-in-the-sky optimism, end-of-the-world pessimism and trivial commonplaces” (Rosenfields, 2001) and therefore lack credibility. Much of the underlying basis of e-commerce statistics are not transparent and in many cases appears to be clouded by commercial purpose. James Rosenfields further suggests that the longevity of the medium, newspaper being old verses the internet being new has a direct correlation in its perceived trustworthiness. This notion of trustworthiness is essential in developing business to consumer (B2C) e-commerce in a way that many companies are used to working by letters of agreement, tacit contracts or unsecured credit whereas consumers are not. There is a general perception by consumers that more credit fraud is created by unsecured internet connections than them not keeping a watchful eye on their credit cards in shops and bars. In fact 5% of all internet sales using credit cards are fraudulent and carried out by consumers against merchants (Sandoval, 2001). In consequence credibility and trustworthiness are often outside the experience of many web retailers and sets an environment of fear in e-commerce.

Foundational Analytics

Given the commercial value of internet analytics governmental websites have been chosen as a prime source of information. While governmental statistics may be open to influences by national economic concerns, the first and most fundamental statistic of total population holds no specific national benefit so may be considered viable. The U.S Census Bureau issues world population data which has been factored to be 6,446,131,400 (U.S Census Bureau, 2004) at a mid point in the year of 2005. Conversely this data is used with an algorithm by David Levin from the University of North Carolina in his population clock which goes up three to four people per second. At 12.30pm 7th April 2005 the population stood at 6,511,632,444 (Levin, 0000). While this format of data presentation is quite compelling it is without consensus and as such flawed. If the population of the world is not an agreed figure it is therefore difficult, if not impossible to determine a percentage usage of the internet as a derivative of world population. Further more to calculate return on investment (ROI) in association with internet usage, sales figures and market penetration by population methods would also be impossibly flawed.

Analytics of Success

There are many methods of analytical calculation in use on the internet e.g. click through rates (CTR) which are derived from using coded links which log activity and flow direction, website visitors which is an extrapolation of site hits, data requests and site session logs. However the underlying data is captured in only two arenas. Data is derived from either locally assessed user activity or remotely assessed communication packets or logging. These two data sources then form the basis of calculations in association with national and worldwide population statistics.

Activity Analytics

Activity statistics are derived through the use of pre-defined or targeted user groups surveys or reviews, cookies, banners, html email, adware and spyware.

Predefined groups are used by Nielson NetRatings to produce commercial guidelines for internet design. They include recommendation of functionality, content design and creation and accessibility. Commonly, internet experts are used to assess websites, they perform heuristic evaluations based upon defined criteria. This method is highly favoured in commercial arenas as it provides absolutes. An unfortunate aspect of this sort of evaluation is a level of condescension is required to dumb down the findings to allow for non technical usage of the internet. While this kind of study may produce insights into professional aspirations for internet usage it says very little about mass consumption. The motivation for this kind of study is consultancy or publication fees. The use of this sort of data to describe national or global interactions is highly suspect.

Targeted user group are used to perform ethnographic reviews commonly in users office or home that study interactions based upon goals, walkthroughs or scenarios. This type of study is highly effectual in producing notional interaction behaviour. Until recently the use ethnographic reviews has been associated with a response to an existing website rather than the underlying philosophy or ethos.

References (some links require fees)

Cranor, L., Langheinrich, M., Marchiori, M., Presler-Marshall, M. (16th April 2002) The Platform for Privacy Preferences 1.0 (P3P1.0) Specification. W3C Recommendation. Published on W3C. [Electronic version]. Retrieved March 10th, 2005 from

Sandoval, G., (5th October, 2001) As Net fraud grows, so do e-tailers’ fears. CNET Published on ZDNet News. [Electronic version]. Retrieved March 10th, 2005 from

Wolverton, T. (13th June, 2000). Amazon, others add personal touch to home pages. CNET. Published on [Electronic version]. Retrieved March 10th, 2005 from,+others+add+personal+touch+to+home+pages/2100-1017_3-241869.html

Rosenfield, J. R., (November 2001). Lies damned lies, and internet statistics. Direct Marketing. Published: Garden City. Vol 64, Iss 7 pg 61 – 64. [Electronic version]. Retrieved March 10th, 2005 from Proquest.

Mullarkey, G., W., (2004). Internet Measurement data – practical and technical issues. Marketing, Intelligence & Planning. Vol 22, No 1, pg 42-58. [Electronic version]. Retrieved March 10th, 2005 from Emerald Fulltext

Frangos, A. (16th September 2002). E-Commerce (A Special Report): Selling Strategies — Search Engines: A Question for Google — Why do you think your Q&A site will succeed — after so many others have failed? Wall Street Journal (Eastern edition). Published: New York, N.Y. pg. R.8. [Electronic version]. Retrieved March 10th, 2005 from Proquest.

Phippen, A., Sheppard, L., Furnell, S. (2004). A practical evaluation of Web analytics. Internet Research: Electronic Networking Applications and Policy. Vol 14 No 4, pg 284 – 293. [Electronic version]. Retrieved March 10th, 2005 from Emerald Fulltext

Hallerman, D., (29th November 2004). Why online advertising matters. Marketing. Published: Toronto 2004 Vol 109. Iss 39. pg E34  [Electronic version]. Retrieved March 10th, 2005 from Proquest.

McClelland, S., (Febuaray 2004) Mobile internet statistics flatter to deceive. Telecommunications International. Published: Norwood. Vol 38. Iss 2. pg 15. [Electronic version]. Retrieved March 10th, 2005 from Proquest

Office for National Statistics (2002). e-Commerce And Internet Why we measure web business and internet access. Office for National Statistics [Electronic version]. Retrieved March 10th, 2005 from

U.S Census Bureau (2004). Total Midyear Population for the World: 1950-2050
(Data updated 9-30-2004) [Electronic version]. Retrieved April 7th, 2005 from

Levin, D.(00). Population for the World counter based upon Data updated 9-30-2004 University of North Carolina [Electronic version]. Retrieved April 7th, 2005 from

Nielsen/NetRatings (March 2005) U.S. Internet Usage Shows Mature Growth, Forcing Innovation of New Web Offerings.  Nielsen/NetRatings [Electronic version]. Retrieved April 7th, 2005 from

U.S Census Bureau (2004) Quarterly Retail E-Commerce Sales. U.S Census Bureau. [Electronic version]. Retrieved April 7th, 2005 from

E-Stats (2004) Statistical Methodology – ASM. U.S Census Bureau. [Electronic version]. Retrieved April 7th, 2005 from

Foley, P. (2001) Internet and e-commerce statistics. European Business Review. Vol 13, No. 2. [Electronic version]. Retrieved March 10th, 2005 from Emerald Fulltext.

Clayton, T., Tsvelik, M., Waldron, K. (2002) International e-Commerce Benchmarking Experimental Statistics Database. Office of National Statistics: New Economy Branch.  [Electronic version]. Retrieved March 10th, 2005 from National Statistics Online.

Clayton, T. (2002) Towards a Measurement Framework for International e-Commerce Benchmarking. Office of National Statistics: New Economy Branch.  [Electronic version]. Retrieved March 10th, 2005 from National Statistics Online.

Newman, Eric J; Jr, Donald E Stem; Sprott, David E (2004) Banner advertisement and Web site congruity effects on consumer Web site perceptions. Industrial Management & Data Systems; Vol 104. No. 3.  [Electronic version]. Retrieved March 10th, 2005 from Emerald Fulltext

Unknown. (22nd November 2004) Value of Internet sales doubles in 2003. Office of National Statistics.  [Electronic version]. Retrieved March 10th, 2005 from National Statistics Online.

Unknown. (16th Febuary 2005). Internet sales surge 21% in January. Interactive Media in Retail Group (IMRG). [Electronic version]. Retrieved April 7th, 2005 from National Statistics Online.

Unknown. (17th March 2005). Retail Sales: Underlying sales are in decline. Office of National Statistics. [Electronic version]. Retrieved April 7th, 2005 from National Statistics Online.

Tagged : / / / / / / / / / / /

#Definition of #UX

Philosophy of UX

User experience is about making people’s lives better not just changed

A persons experience is based in their mind and their emotions and can be established by both actual interaction and reflective (biographical experience) inputs.

In UX we define inputs in digital or real world frameworks which enable the creation of solutions that have meaningful impact and that can be measured.

Overview of UX

The current approach to UX is that it is the practical implementation of audience drivers, cognitive acuity, usability standards and accessibility laws with ergonomics (physical, contextual use) and anthropometric (digital behaviours analytics) measures. Creating an integration of business context into user context, to facilitate alignment, transactions and communications.
Definition; A user is a representative of the target audience. They are not involved in the project in any way. They will use the final product or service either as a customer or as an internal business user.
UX is not involved in the Look and Feel associated with GUI’s but rather delivers the human solution that can be accessed through any user interface this is why UX is closely associated with assistive technologies used in accessibility which are in turn derivatives of technologies developed for the military and space exploration.
While UX is not rocket science, it has been involved in the space program

The UX Process

We first try to ‘Understand the Problem’ from the user perspective (user research) so that we can create User Requirements, these combined with Business Requirements and Implicit Requirements create Project Requirements.
This process often called Discovery and can find new requirements, challenge business requirements or redirect the entire project along a route that delivers the business or organisation what they want but in a totally different way.
To de-risk human error (needing to be right) we work from researched archetypes (persona modelling) which creates the opportunity to ‘think like a user’ a great support tool if users are not always available.
This post is republished from an earlier blog from 2001.
Tagged : / / / / / / / / / / / / / / /