Competitive Analysis in the Web

Daniela Favaretto[*] and Tiziano Vescovi[**]

Ca’ Foscari University of Venice – Italy

Track: New Technologies and E-Marketing

or

Marketing ResearchAbstract

Despite the deep attention that the phenomenon represented by the Internet is receiving from marketing scholars, there are relatively few studies concerning the competitive evaluation in general and the marketing competition in particular.

The problem is to evaluate the relative effectiveness and efficiency of different company Web sites. In order to cope with this problem, an evaluation procedure has been proposed, following six steps, including statistical and mathematical tools.

An exploratory research was carried out in order to test the procedure. The procedure seems to be useful in evaluating competitors in their Web marketing efficiency, even if some further research should be developed to improve the quality of the competitive evaluation.

  1. The problem: competitive analysis in the Web

Despite the deep attention that the phenomenon represented by the Internet is receiving from marketing scholars, there are relatively few studies concerning the competitive evaluation in general and the communication competition in particular. Existing studies focus on simple models that can help to assess the communication performance of a single Web site or Web strategy, but there are quite rare attempts of comparative evaluation among competitors (Berthon, Pitt, and Watson 1996; Novak and Hoffman 1997; Vescovi 1997; Drèze and Zufryden 1997; Vescovi 1998; Watson, Berthon, Pitt and Zinkhan 2000, Elliot, Mørup-Petersen and Bjørn-Andersen 2000).

Anyway the competition in the Web has changed some rules. First of all the competitive dimension has become widely international. Competitors emerge from far countries and enter a common market. Second, the relative reduction in the Internet of financial barriers, quite important in communication contests, lets small companies reach competitive groups once dominated by big corporations. Third, the traditional industries boundaries fall down under the web pressure, including new comers and surprising the incumbents. The Internet influence results in a renewed competition, made by new communication rules and new competitors.

  1. The evaluation procedure

The evaluation procedure we propose can be described in six steps.

Step 1: macro variables definition

A set of macro variables should be defined and evaluated in order to describe the marketing effectiveness of the Web site. We have considered twelve variables (……….) and in particular:

  • Visibility (easy to find the site using research engines, directories, vertical portals, etc.)
  • Usability (structure, navigation tools, and contents organisation making simple the visit of the site)
  • Surfing easiness (text readability, different languages, graphics and icons, etc.)
  • Corporate information
  • Company image (presented by the Web site)
  • Marketing information
  • Transaction quality (e-commerce)
  • Customer service
  • Customisation (tools to customise product, services, and information)
  • Content quality (useful, updated, and complete information)
  • Community (tools supporting communication among customers, as forum, chat, etc.)
  • Entertainment (multimedia and entertainment solutions in order to motivatefrequent visits)

Step 2: micro variables definition, evaluation, and weighing

Each one of the macro variables should be exploded into a set of micro variables, in order to reduce the subjectivity of the evaluation, that can be judge on a 0-5 scale. For example ‘visibility’ has been described using 11 micro variables and each one has been assessed. Moreover, each micro variable has been weighed inside the macro variable by a group of experts in Internet marketing, considering the relative importance of each micro variable inside the macro variable. The total number of micro variables considered in the model is 124.

Step 3: macro variables weighing

Each macro variable should be weighed, in order to define the relative importance in terms of marketing effectiveness, combining three different samples of people. Macro variables would be weighed by a group of managers of the competitive companies, by a group of customers, and by a group of surfers. Managers bring into the model the business point of view, customer the market point of view, surfers the users point of view. Each respondent has to answer to a questionnaire listing the variables on the Likert scale basis. The results of the questionnaire are transformed into a percentage weight.

Step 4: data collection about investments

Each selected Web site has to be evaluated through micro variables, subsequently reassembled into macro variables. These data are used by DEA (Data Envelopment Analysis) model in order to evaluate the marketing efficiency of the Web site. Following the same purpose, further data should be collected, regarding resources used in creating, developing, and maintaining the Web site. The resources considered in the evaluation procedure are the amount of the financial investment used in developing and launching the site and the amount of money normally spent to update the site and keep it working.

Step 5: statistical and mathematical models application

Collected data are used in two frameworks to reach expected result, i.e. the competitive analysis of the companies on the Web. One set of tools includes some multivariate statistical devices as factorial analysis, makes the researchers able to identify the competitive groups and the positioning of the companies considered in the research. The other tool is a mathematical programming model and in particular it is a DEA method (Charnes et al. 1994) to identify and analyse data in order to study the relative efficiency of different companies on the Web.

Step 6: competitive evaluation

Results of the mathematical programming model are analysed in order to assess the competitive positioning of the companies included into the evaluation process, considering the efficiency level, the relative comparison, the points to be improved. Factorial analysis has been added to complete the competitive evaluation.

3.Exploratory research

An exploratory research was carried out in order to test the procedure, using 154 Web sites referring to companies of the gold & jewellery industry, leather industry and industrial electronics industry. Considering a traditional competitive analysis, we can use the factor analysis to define a positioning map (exhibit 2). We used 154 web sites to outline two main explanatory factors (48.4%) that can be named as customer orientation (45.3%) and company image (10.9%). Considering the average value of each factor two lines have been drawn, dividing the map into four quadrants. In the north west quadrant there are companies using the web site as a brochure, while in the south west quadrant you can find companies having the worst position (low customer orientation and low company image); they have only a simple presence on the Web. In the north east quadrant there are companies focusing their Web strategy on relationship variables, keeping attention on customer orientation and communication. In the south east quadrant there are companies whose strategy is mainly based on customer service.

Ten Web sites, out of 154, where selected on two conditions: being industry homogeneous, being available information about Web site investments. In exhibit 3 you can find the evaluation of each of 10 industrial electronics industry sites, using the micro and macro variables. The last column contains the final rating, weighed using the opinion of the three communities, as explained above. Investment data in building and maintaining Web sites are used as inputs in the DEA model. The evaluation of single macro variables are contained in the columns from the fourth to the fifteenth; these data are used as outputs in the DEA model.

Exhibit 1. Web site competitive positioning – all industries

Exhibit 2. Web site competitive positioning and efficiency – 10 web sites of industrial electronic industry

Factor analysis and positioning map can help in defining the competition, i.e. which are the competitive groups, the distance among the competitors, the different basic strategies followed by the companies. However, they are not able to evaluate the efficiency of the invested resources in developing a Web site, and in explaining where (on which variable) to improve the Internet marketing strategy.

Exhibit 3. Assessment of 10 Web sites in industrial electronic industry.

company / Develop. costs / Updating costs / Visibility / Usability / Surfing easiness / Corporate informat. / Company image / Marketing informat. / Transact. quality / Customer service / Customi
sation / Content quality / Community / Entertainment / Total weighed score
A / 1,600 / 900 / 2.13 / 2.55 / 3.38 / 1.53 / 1.97 / 1.00 / 0.00 / 0.00 / 0.00 / 1.16 / 0.00 / 0.00 / 1.320
B / 3,500 / 1,000 / 2.20 / 2.26 / 3.30 / 1.00 / 2.17 / 2.54 / 0.00 / 1.14 / 0.00 / 1.14 / 0.00 / 0.56 / 1.540
C / 14,000 / 1,800 / 2.42 / 3.68 / 2.61 / 2.83 / 2.70 / 2.44 / 0.00 / 0.63 / 0.53 / 1.49 / 0.16 / 0.00 / 1.828
D / 3,000 / 800 / 1.69 / 2.15 / 2.30 / 1.62 / 1.77 / 1.72 / 0.19 / 0.79 / 0.53 / 1.38 / 0.00 / 0.00 / 1.327
E / 8,500 / 1,800 / 0.78 / 2.11 / 2.13 / 1.45 / 2.47 / 1.67 / 0.00 / 0.46 / 0.00 / 1.54 / 0.00 / 1.00 / 1.244
F / 8,000 / 2,000 / 2.04 / 2.28 / 2.66 / 2.07 / 3.02 / 1.48 / 0.00 / 0.70 / 0.00 / 1.22 / 0.00 / 1.11 / 1.514
G / 16,000 / 9,500 / 2.62 / 2.68 / 3.45 / 1.00 / 2.93 / 0.80 / 0.00 / 0.07 / 0.00 / 2.03 / 0.00 / 0.56 / 1.533
H / 16,000 / 10,500 / 1.22 / 2.94 / 3.38 / 2.24 / 2.08 / 2.54 / 0.00 / 1.44 / 0.00 / 1.00 / 0.00 / 0.56 / 1.626
I / 1,500 / 500 / 1.31 / 1.70 / 2.48 / 1.00 / 1.22 / 1.48 / 0.00 / 0.21 / 0.53 / 0.51 / 0.00 / 0.56 / 1.018
J / 26,000 / 8,000 / 1.84 / 2.30 / 3.46 / 1.43 / 2.22 / 1.54 / 0.00 / 1.37 / 1.05 / 1.68 / 0.00 / 0.56 / 1.617

Therefore the problem is to determine the relative Web marketing efficiency of 10 independent homogeneous companies. A company transforms resources (inputs) into products (outputs). A measure of the (absolute) efficiency of this transformation is the ratio Output/Input. The inputs considered in the research are the amount of the financial investment used in developing and launching the site, the amount of money normally spent to update the site and keep it working. The outputs are the macro variables that evaluate the company’s Web site. To obtain a single ratio Output/Input it is necessary to combine all inputs and all outputs. The absolute efficiency of a company is the ratio between the weighted sum of outputs and the weighted sum of inputs. The relative efficiency of company j with respect to the company r is the ratio between the absolute efficiency of company j and the absolute efficiency of company r. The DEA (Data Envelopment Analysis) models (Charnes, Cooper and Rhodes 1978, Charnes, Cooper, Lewin and Seiford 1994; Cooper,Seiford and Tone, 2000) determine the optimal inputs and outputs weights in order to maximize the relative efficiency of each company.

Exhibit 4. Companies’ efficiency and benchmarks.

Companies / Efficiency / Companies more efficient than the considered company / Company as Benchmark (n° of times) / Superefficiency
A / 100,00% / 1 / 1.7
B / 100,00% / 3 / 1.2
C / 100,00% / 1 / 1.3
D / 100,00% / 4 / 1.4
E / 56,99% / D, I
F / 46,43% / B, D, I
G / 21,44% / A, I
H / 37,19% / B, C, D, I
I / 100,00% / 5 / 5.5
J / 21,40% / B, D, I

Applying DEA model to the data in Exhibit 4 it is possible to determine the relative efficiency for each company. In particular, how Exhibit 4 shows, the companies A, B, C, D, and I are efficient, wile the companies E, F, G, H and J are inefficient and have different level of inefficiency. The third column explains, for each inefficient company, the companies which are more efficient than the considered company. The fourth column explains, for each efficient company, the number of inefficient companies which have the considered company as benchmark. These evaluations are made on the basis of the weights chosen by the considered company. For example company D is efficient and 4 inefficient companies have company D as benchmark; company F is inefficient (its relative efficiency is 46,43%) and, using the weights chosen by company F, that is the optimal weights for itself, companies B, D, and I are more efficient than company F. Crossing the information originated from exhibit 3 and 4 (efficiency and positioning map) is possible to define the threat represented by competitors. In this sense it is possible not only to identify the closest competitors on the Web, but also their efficiency. For example, B and C, are in a better position if compared to G, F, H, and J because of their efficiency in using resources in the Web, while other companies are not competing directly with them. In particular C is in a better position also from a effectiveness point of view (exhibit 2). The score in the fifth column identifies the superefficiency of each efficient company, i.e. it indicates the maximal percentage change which is feasible such that the company remains efficient. Superefficiency measures the strength of each efficient company.

Exhibit 5 identifies the importance of outputs that the inefficient companies would have to reach to be considered efficient (***=very important; **= important; *=not so important). For example, company F would have to improve the usability, the surfing easiness, and marketing information as first steps to reach efficiency and a better competitive position, being the investment equal.

Exhibit 5. Outputs improvement to reach the efficiency.

Companies / Visibility / Usability / Surfing easiness / Corporate information / Company image / Marketing
information / Customer service / Customisation / Content quality
E / * / * / ** / * / * / * / * / - / -
F / * / ** / ** / * / * / * / * / * / *
G / *** / *** / *** / ** / *** / ** / - / * / **
H / ** / ** / ** / * / ** / ** / - / - / *
J / *** / *** / *** / ** / *** / *** / ** / * / **
  1. Conclusions and future research

The exploratory research leads to some conclusions to be discussed and underlines other opportunities for a future research. The main conclusions can be summarised into two main points:

  • The competitive analysis based only on traditional tools remains an important step, but it fails in explaining some aspects, particularly the efficiency of the company’s choices.
  • The mathematical model can be really useful in comparing competitors showing similar results in marketing strategies. In facts:
  • Conditions being equal in marketing results, some companies are more efficient than others
  • Web marketing budget being equal, low efficiency points can be identified
  • More virtuous behaviours can be suggested to the companies, as well as which employed resources show a lower level of efficiency

For example, companies G, F, and J (exhibit 2) are located in a substantially positive quadrant of the factorial map (relationship) as well as H (service), but they show really low efficiency in using input resources (exhibit 4) or they should improve, investing the same amount of resources, some performances(exhibit 5). “Transaction quality” as well as “community” and “multimedia & entertainment” are evaluation factors reaching a low score for all the companies included in the sample. As consequence they are not considered as improvement points to be enhanced.

Nevertheless, the research underlined the need of further steps. First of all, the integration between the statistical and mathematical tools used in developing the evaluation procedure could be improved.

Another point to be developed could be the industry comparison, considering a broader competitive perspective, quite useful in the Internet context. Close industries can compete quite quickly in the new economy, their borders can suddenly fall down, and their Web communication levels should be kept under control in order to prevent new competitors’ attacks.

References

Berthon R. T., Pitt L. F., Watson R. T. (1996), “The World Wide Web as an Advertising Medium: Toward an Understanding of Conversion Efficiency”, Journal of Advertising Research, January-February.

Charnes A, Cooper W. W., Rhodes E., (1978), “Measuring the efficiency of decision making units”, European Journal Of Operational Research, 2, 429-444.

Charnes A., Cooper W. W., Lewin A. Y., Seiford L. M. (1994), “Data Envelopment Analysis: theory, methodology and applications”, Kluwer Academic Publisher.

Cooper W. W.,Seiford L. M., Tone K. (2000), “Data Envelopment Analysis: a comprehensive text with models, applications, references and DEA-solver software”, Kluwer Academic Publisher.

Drèze X., ZufrydenF. (1997), “Testing Web Site Design and Promotional Content”, Journal of Advertising Research - Mar/Apr.

Elliot S. R., Mørup-Petersen A. S., and Bjørn-Andersen N. (2000) “Towards a framework for Evaluation of Commercial Web Sites”, in: Electronic Commerce: The End of Begin, 13° IBECC, Bled.

Farrel M. J. (1957) “The Measurement of Productive Efficiency”, Journal of Royal Statistics Society, 3, 253-290.

Novak, T. P. and D. L. Hoffman (1997), "New Metrics for New Media: Toward the Development of Web Measurement Standards," World Wide Web Journal, Winter, 2(1).

Vescovi T. (1997), “The Unknown Link. Communication Effectiveness in the Internet”, in: Marketing: Progress, Prospects, Perspectives, Warwick, Emac.

Vescovi T. (1998), "Literature, Metaphors, and the Formula to Advertise in the Internet", in: P.J. Kitchen (editor) The Changing World of Corporate and Marketing Communications: Towards the Next Millennia, Glasgow, SGBS.

Watson R. T., Berthon R. T., Pitt L. F. and Zinkhan G. M. (2000), Electronic Commerce, London, The Dryden Press.

1

[*] Department of Applied Mathematics, Ca’ Foscari University, Ca’ Dolfin, Dorsoduro 3825/E, Venice – Italy;

[**] Department of Economics and Business Management, Ca’ Foscari University, Ca’ Bembo, S. Trovaso 1075, Venice – Italy; ; ph: +39 041 234 87 41