"Why South African Retailers Do So Badly On Google and Other Search Engines"

1 Introduction

For good reason well known companies with recognizable brands place great value on their brand names. Over the years these same companies have invested enormous sums of money in promoting the brand so that customers become somewhat conditioned to buying by brand name.

What many of these companies have not recognized is that when it comes to the Internet brand can certainly play a role but the value of that real world brand is discounted.

On the Internet people buy what they are looking for and do not necessarily remain brand loyal.

For more than 18 months I have been observing the web sites of 3 major retailers in South Africa and looking for signs of change that they are truly embracing the power of the web. The retailing giants were

• Woolworths
• Pick N Pay
• Makro

These 3 retailers rely almost totally upon their brand name for any chance of being found on search engines. This point is easily proven.

Think of this if youre looking to buy a wedding present for somebody do you type Woolworths into the search engine box of Google? Probably not. It is far more likely that you will type wedding present ideas or if you know what you want then you would type in say cutlery set or dinner plates.

If you typed any such phrase into Google (www.google.co.za) you would never find Woolworths in your search results and since Woolworths sell a whole load of wedding gift ideas and own a developed e-commerce website it is a pity. The company is losing potential sales for that item and, more importantly, a chance to convert that person to a long-term online Woolworths shopper.

Of course this malaise is not confined to retailers it is symptomatic of most websites in South Africa.

I first wrote about this problem in Real Business supplement of Business Day around June 2004. As a follow up I decided to do some specific research to demonstrate unequivocally 2 things

The absolute power of the Internet to generate new customer revenue

That it is just as important (probably more important) that web sites be found for what they offer (sell) in most cases rather than for what they are (brand).

2 Background to the Research

For the last 4 to 5 years Internet keyword market research has been an important part of my life. Over this period I have gained considerable experience in researching any market online in an efficient, quick and accurate way using an extremely wide variety of tools. Consequently I have built up a capability to discover a vast array of information in terms of what people are looking for on the web when they use search engines.

A great deal of general information on search habits is readily available using free keyword research tools (such as the Overture, now Yahoo, keyword tool) but there is virtually no information available on what South African searchers are looking for. The following was therefore the basis behind my South African research

I identified from my own information sources 1,968 of the most important retail type terms that were being entered into search engine by South Africans in 2005.

These terms were categorized into 8 distinct groupings shown in the table below

Category	Search Terms in category
Appliances	390
Baby	81
Fashion & Beauty	177
Food & Drink	212
Gifts	551
Home & Garden	210
Recreation & Leisure	132
Retailer Brand Names	215
TOTAL	1968

Over a period of 8 days between December 21 and December 28th 2005 I monitored the actual real Google searches performed in South Africa only for every single one of the terms within each of the categories. Why this period? Well I was on holiday and it gave me the opportunity to do the work.

The total number of searches made for the 1,968 terms only was 56,844 (or around 2.6 million per year on a strict pro-rata basis).

Based upon this research some general and, in my view, accurate and valuable conclusions were made possible based upon the information gathered. These are discussed in general terms below.

Of course the period of the research was short and was just before and after Christmas. This needs to be taken into account in assessing the validity in strict statistical terms.

3 Results of Research

In the table below the actual number of searches made for each category is shown. For example the number of searches made for the actual brand name of the 215 retailers was 14,409. This represented 25.3% of all searches made.

Category	Search Terms in category	Number of Searches Made	Percent of Total
Retailers	215	14,409	25.30%
Gifts	551	13,509	23.80%
Recreation Leisure	132	8,758	15.40%
Appliances	390	6,206	10.90%
Food Category	212	5,418	9.50%
Fashion	177	4,067	7.20%
Home & Garden	210	2,645	4.70%
Baby	81	1,832	3.20%
Totals	1,968	56,844	100.00%

3.1 Which Retailers are being searched for by Brand Name?

Incredible connection was top of the list over the 8 days surveyed with the term Incredible Connection being entered into Google 1,593 times. Hi-Fi corporation was second as can be seen in the table below.

Retailer Searched For	Number of Searches Made	Percent of Total Searches
Incredible Connection	1,593	2.80%
Hi Fi Corporation	1,329	2.30%
Woolworths	1,127	2.00%
Makro	963	1.70%
Musica	803	1.40%
Exclusive Books	742	1.30%
Pick And Pay	648	1.10%
Edgars	560	1.00%
Game Stores	365	0.60%
Foschini	298	0.50%
Sportsmans Warehouse	289	0.50%
Hyperama	274	0.50%
CD Warehouse	244	0.40%
American Swiss	243	0.40%

Between all these top retailers they could only garner some 16.6% of all the searches. Yet they must carry an absolute minimum of 100,000 SKUs (stock keeping units) of which I researched a tiny fraction.

Clearly people are not using search engines primarily to look for Brand names after all what do these same people do once theyve found the branded website? It is almost a certainty that they have little intention of buying

3.2 Which Products Were Being Searched For?

Bearing in mind this is not scientific research but nevertheless valuable and meaningful the table below shows the searches per category item in an effort to normalize the data

Category	Category Terms	Searches Made	% of Total	Average Searches per Category Item
Retailers	215	14,409	25.30%	67
Gifts	551	13,509	23.80%	24.5
Recreation Leisure	132	8,758	15.40%	66.3
Appliances	390	6,206	10.90%	15.9
Food Category	212	5,418	9.50%	25.6
Fashion	177	4,067	7.20%	23
Home & Garden	210	2,645	4.70%	12.6
Baby	81	1,832	3.20%	22.6
Totals	1,968	56,844	100.00%	28.9

It can be seen that people are actively looking for information or products to do with Recreation and Leisure.

The table below shows some of the more important products or services searched for

Searched term	Searches Made	Percent of Total
baby names	685	1.20%
barbie	642	1.10%
bikini	533	0.90%
recipes	508	0.90%
lingerie	477	0.80%
hairstyles	404	0.70%
scrapbooking	372	0.70%
cocktails	335	0.60%
garmin	267	0.50%
calendar 2006	250	0.40%
cocktail recipes	248	0.40%
adult games	239	0.40%
golf	224	0.40%
mag wheels	201	0.40%
game cheats	201	0.40%
model trains	192	0.30%
calendar	192	0.30%
lg electronics	187	0.30%
ps2 games	184	0.30%
garden route	178	0.30%
online shopping	167	0.30%
art	167	0.30%
furniture	166	0.30%
kenwood	166	0.30%
food	162	0.30%
restaurants	157	0.30%
bikinis	150	0.30%
koi	143	0.30%

This paper is a very short introduction to practical keyword market research using the power of keyword discovery and online search tools. Given that there are in excess of 150,000 single English words recognized by Scrabble players and that rarely does a single term search give anybody useful information it should be obvious that there really can be no substitution for keyword research and management before entering into any serious attempt to sell or gather potential customers online.

Google reports that of the 500 million world-wide searches made every day that approximately 50% are unique.

4 Conclusion

Those companies that embrace expert keyword discovery, analysis and practical implementation will win the Internet race for customers and sales hands down.

Download the pdf of this page