9.8 Whisky Galore

Dale Abrahams is an entrepreneur, one of an estimated 7 million such in the USA, and now going into the online selling of fine scotch whiskies. Dale will offer free shipping to NY State addresses on purchases over $300, and also offer a premium service: a choice whisky, gift-wrapped and sent express delivery anywhere in the world. Capital has come from friends and family, and all Dale has to worry about is selling over the Internet.

Marketing is largely through the pay-per-click search engines (essentially Overture as was) and Dale is reckoning on paying an average of $0.20/click on thirty keywords to get 2000 click-throughs per month in year one. An Overture presence will feed through into other search engines, of course, and Dale is expecting these extra visitors to amount to 1,000, 3000 and 6000/month in years one, two and three. Conversion rate is 2.0% and average profit per sale is $30. Monthly profit in year one is 0.02 x 3000 x $30 or $1800, rising to 0.02 x 5000 x $30 or $3000 in year two and 0.02 x 8000 x $30 or $4800 in year three. His 'conservative' business plan (all in US$'000) can be summarized as:

	Year One	Year Two	Year Three
Profit on Sales	21.6	36.0	57.6
Site Build	8.0	0	0
Site Maintenance	0	2.0	2.0
Marketing	4.8	4.8	4.8
Salary, hosting, etc.	20.0	20.0	20.0
Profit (Loss)	(11.2)	9.2	30.8

Dale's living expenses, kept to an absolute minimum of $20,000 a year (which includes site hosting at $50/month) is a disproportionately large item in the plan, but Dale hasn't the funds, or indeed the confidence, to lower this proportion by starting bigger. But he will be guided by results, replanning operations if the first year goes better than expected.

What were the results? Here are the figures for the first year of trading, with explanations to follow:

	Year One
Profit on Sales	2.7
Site Build	8.0
Site Maintenance	0
Marketing	3.6
Salary, hosting, etc.	20.0
Profit (Loss)	(28.9)

1. Dale had to pay an average of $0.25 per click and still got only 1200 click-throughs a month.
2. The feed through to other search engines was minimal, and amounted to only 500/month.
3. Supply difficulties reduced the average profit margin to $11/order, and the conversion rate was 1.2%.

Phase Two

Dale decides to concentrate on his most profitable lines, revamping the site to attract visitors from Google, etc. Here is the plan for years two and three:

	Year One	Year Two	Year Three
Profit on Sales	2.7	42.0	58.8
Site Build	8.0	5.0	0
Site Maintenance	0	0	2.0
Marketing	3.6	3.0	3.0
Salary, hosting, etc.	20	20	20
Profit (Loss)	(28.9)	14.0	33.8

Dale still pays $0.25/click at Overture, but opts for only 1,000 clicks a month on more selected keywords. Conversion rate is planned at 2%, but the profit per sale is $35. Dale also counts on getting another 4,000 and 6,000 visitors per month from the other search engines. But the picture after year two is:

	Year One	Year Two
Profit on Sales	2.7	8.4
Site Build	8.0	5.0
Site Maintenance	0	0
Marketing	3.6	3.6
Salary, hosting, etc.	20	20
Profit (Loss)	(28.9)	(20.2)

1. Overture raised their bid prices, and Dale had to pay an average of $0.30 per click for 1000 click-throughs a month.
2. Traffic from other search engines was only 1500/month.
3. An average profit margin of $35 was achieved but the conversion rate fell to 0.8%.

Phase Three: Major Redesign

Dale now gets an ecommerce consultant in, who recommends a raft of measures: redesign to make the site search-engine friendly, search engine promotion by professionals, a newsletter, diversifying into choice wines, etc. The plan now expects 10,000 visitors/month, a conversion rate of 2% and a profit margin of $20. Dale has to hire someone to write the newsletter.

	Year One	Year Two	Year Three
Profit on Sales	6.1	8.4	48.0
Site Build	8	5	5
Consultancy	-	-	3.5
SE Promotion	-	-	5
Newsletter	-	-	2.4
Marketing	3.6	3.6	3.6
Salary, hosting, etc.	20	20	20
Profit (Loss)	(25.5)	(20.2)	8.5

What happened? The site only attracted 6000 visitors/month and the conversion rate rose to 1.2%. Dales's reward for three years of hard work was a loss of $68,200.

	Year One	Year Two	Year Three
Profit on Sales	6.1	8.4	17
Site Build	8	5	5
Consultancy	-	-	3.5
SE Promotion	-	-	5
Newsletter	-	-	2.4
Marketing	3.6	3.6	3.6
Salary, hosting, etc.	20	20	20
Profit (Loss)	(25.5)	(20.2)	(22.5)

Unrealistic? Many fledgling emerchants have worse experiences. Dale's mistakes were:

General Errors

1. The business plan wasn't sufficiently researched.
2. The competition wasn't assessed. The days of 'having a go' are long past, and anyone who operates on the Internet today is up against professionals with extensive experience and large budgets.

Specific Errors

1. Profit margins weren't really known.
2. Too much reliance was placed on Overture, ignoring the cheaper pay-per-click machines, paid placements, eBay and price comparison machines.
3. The natural search engines were largely ignored.
4. The site was over-expensive and inflexible: content was database fed (invisible to the natural search engines) and no accountancy package was integrated (book-keeping was time consuming, and complete picture was difficult to see).
5. A newsletter was introduced too late and in the wrong way.

Phase Four: Realism Dawns

A more realistic picture would have been:

	Year One	Year Two	Year Three	Year Four	Year Five
Profit on Sales	18.0	30.0	48.0	86.0	115.0
Research	2.3	0
Site Build	0	0	4.0	2.0	2.0
SE Promotion	-	-	5.0	3.0	3.0
Newsletter	-	-	2.0	2.0	2.0
Marketing	4.0	8.2	20.0	24.0	24.0
Salary, hosting, etc.	10.0*	10.0*	30.0	30.0	30.0
Profit (Loss)	1.7	11.8	(13.0)	25.0	54.0
	eBay sales	eBay sales	own website	own website	own website

* Part-time only

Profit Margins

In our worked example, Dale Abrahams expected an average profit margin of $30 per order. The assumption played a critical role in his business plan, so how did he arrive at the figure?

Simple. He spent weeks checking prices at the Internet stores and at local shops, 'finding':

1. Collecting Scotch is a hobby, indeed more a religion. Some labels are difficult to find, and price is not the customer's first consideration. A $10 difference in prices between Internet retailers was common, and the differential could be up to $40 for the choicer malts or blends. Bargains could sometimes be found in off-line stores.
2. Even the largest online stores stocked only a selection, leaving many labels unrepresented.
3. The price differential could be widened to $50 if choice whiskies were marketed as special gifts, i.e. offered as an all-in service with 'free' overnight shipping.
4. There were many books and even Internet sites, providing information that could be 'repackaged' on Dale's site, enhancing the perceived value of his products.

Presentation

Dale had a two-pronged approach. His site, whiskies-online.com, was advertised as the specialist site for Scotch whiskies, and carried good photos and interesting write-ups. What wasn't in stock — and Dale in fact didn't carry anything in stock — could and would be sourced without delay. As an extra service (the second prong) a choice whisky could be gift-wrapped and rushed overnight to anywhere in the world for a 'small' charge.

Naturally, Dale opened correspondence with the distilleries and the larger outlets, though without committing himself to purchase. He kept watching briefs on Internet prices, and made sure that while his site stocked some popular labels at competitive prices, it also featured labels that were not generally available, allowing Dale to add a hefty markup.

Moreover, because he wanted the site to show continually changing stock, with 'just in at a special price' promotions, Dale agreed with the web-design company (who charged $5,000 more for the feature) to have the page content dynamically generated. Dale keyed information into a database, the content of which was then uploaded to the design company's server and fed into Dale's pages. The company charged $50/month for the hosting service, but Dale otherwise had control of his site.

Mistake 1

The plan seemed foolproof, but Dale soon found that:

1. He received comparatively few orders for the popular labels: the conversion rate was just over 1%. There was nothing special to attract customers, and they stayed with their trusted suppliers.
2. He got good orders for the unusual labels, but lost them when he couldn't supply within a week. He was finally obliged to buy from local retailers, obtaining a 10% trade discount, but rarely achieving more than a $10 profit margin.

Dale was beginning to realize what should have been obvious: customers and other retail outlets knew far more than he did about the spirits trade. Good whiskies have to be bought as they become available, and Dale eventually and reluctantly spent some $10,000 on a selection, renting a small lockup, and assiduously promoting just these labels on his website. One outlet offered him a commission to sell their lines, but Dale wanted to do things his way.

Mistake 2

Some of his stock sold well, but a lot did not. Some labels were five times as popular as others, and Dale did not know why. He was tempted to join an enthusiasts club, but realized that, as a retailer, he was supposed to know the answers anyway. He carried on, but suffered from funds tied up in his inventory, which he could not easily dispose of.

His site was also rather different now. Far from being the site, carrying a world-beating selection, it was a collector's site, carrying a small selection of choice items. And that created its own problems. Internet shoppers do not search with 'Glenfarclas Whisky' or 'Buchanan's Deluxe' but under 'whisky'. And the bid price for that search word on Overture was over 30 cents. Visitors came in droves but bought little. If his site had really some two hundred malts on offer, Dale might very well have achieved a decent conversion rate. But he had only 30 on offer, so that 85% percent of visitors went away disappointed: the conversion rate was 0.8%. Customer acquisition cost was therefore $0.30/0.008 or $350: not an exorbitant figure in today's climate but not one Dale could accommodate on a slender budget and with so much capital tied up in slow-moving stock.

Solution

Dale's problems were:

1. He didn't have sufficient experience to anticipate difficulties with the plan,
2. He didn't understand customer preferences, and — most serious of all — he didn't test to ascertain them.

Money was tight, but the $8,000 spent on the site build would have been better allocated as:

1. Basic website built online under some test URL with dummy ordering facilities: $300 in fees for an all-in ecommerce hosting service.
2. 5,000 test click-throughs with Overture and the smaller ppc search engines. Search words, and software/services (split testing and bid management and monitoring) for 3 months of testing would have cost $2000.
3. Stock purchases of whiskies with the right profit margin which could then be marketed through eBay, i.e. without any website at all. (Dale assumed that his customers needed educating with his whisky information pages, but in fact purchasers of $100 plus items have already done their homework.)

Two years later, with a good track record, and sales figures to guide his trading, Dale would have been well placed to get the funding required to stock and run a proper whisky store.

Using Pay Per Clicks Effectively

A proper study would look at the various advertising options more fully, not neglecting banner ads and the natural search engines.

Comparing PPC Search Engines

But if we just look at the top bid prices for 'whisky' in the better-known ppc search engines:

	Overture (now Yahoo)	Google AdWords	Kanoodle	GoClick	Enhance	FindWhat
	average $ per click	average $ per click	average $ per click	average $ per click	average $ per click	average $ per click
whisky	0.34	0.10	0.06	0.03	0.08	0.06
conversion rate	3.0%	6.0%	0.8%	0.8%	0.6	1.3
marketing cost	$ 11.3	$ 1.7	$ 5	$ 3.7	$13.3	$4.6
monthly traffic	261	1020	100	100	300	100

The conversion rates are notional figures, based on MEC's studies, as are the traffic figures for the smaller ppc engines, but what this simple study suggests is:

1. The various ppc machines give very different marketing costs.
2. The smaller ppc engines do not always provide better value.
3. Better traffic is supplied by the larger ppc machines, especially Google AdWords.

Paying For The Top Ranking

Whether you should pay for the top ranking depends on the traffic and conversion rates. Both tend to fall off with decreasing rank, which is reflected in the bid prices. How it will work in your particular case can only be discovered by testing.

	Overture #1	Overture #2	Overture #3	Overture #4	Overture #5	clicks/month for #1 ranking
	average $ per click	average $ per click	average $ per click	average $ per click	average $ per click
scotch	0.25	0.15	0.14	0.14	0.13	320
single malt scotch	0.45	0.25	0.23	0.21	0.11	338
glenfiddich scotch	0.29	0.28	0.27	-	-	11
whisky	0.34	0.29	0.23	0.21	0.20	261
chivas regal whisky	0.13	-	-	-	-	13
glenfiddich whisky	0.26	0.16	-	-	-	1
single malt whisky	0.45	0.20	0.18	0.17	0.16	17

Varying the Keywords

Keyword costs vary considerably, between similar phrases and across the ppc search engines:

	whisky	scotch	single malt	single malt scotch	chivas regal	glenfiddich
Google AdWords
average $ per click	0.10	0.18	0.26	0.36	0.09	0.09
conversion rate	2.0%	3.0%	5.0%	6.0%	5.0%	3.0%
marketing cost	$ 5.0	$ 6.0	$ 5.2	$ 6.0	$ 1.8	$ 3.0
clicks/ month	870	3000	330	72	54	6
	whisky	scotch	single malt scotch	single malt whisky	chivas regal whisky	glenfiddich whisky
Overture
#1 $ per click	0.34	0.25	0.45	0.45	0.13	0.29
conversion rate	2.0%	2.5%	3.0%	3.5%	3.0%	3.5%
marketing cost	$ 10	$ 8.3	$ 15.0	$ 12.9	$ 4.3	$ 8.3
clicks/ month	597	532	338	17	13	11

PPC Engines Generally

1. Ecommerce is competitive, and openings are soon exploited.
2. An effective ppc marketing campaign will employ several hundred bids across ten or more of the ppc search engines. The advantage obvious lies with the bigger companies who can devote staff exclusively to this task.

Price Comparison Machines

The price comparison search engines do not publish their fee structure, and companies have to approach them as a bona fide advertiser with specific products. But if the MEC figures still hold, then using the price comparison engines might yield something like this:

	whisky	scotch	single malt	single malt	chivas regal	glenfiddich
Price Comparison Search Engines
average $ per click	0.10	0.18	0.26	0.36	0.09	0.09
conversion rate	2.5%	2.5%	5.0%	5.0%	6.0%	6.0%
marketing cost	$ 4.0	$ 2	$ 5.2	$ 2	$ 5.0	$ 1.5

Using eBay and other Auction Sites

eBay charge an insertion fee ($0.30 to $4.80) plus a commission (5.25% for the first $25, plus 2.75% of the remaining closing price). Marketing fees (supposing no re-insertions) compare as follows.

$ price of item	$65	$120	$180
Google AdWords	$1.8 - $6.0	$1.8 - $6.0	$1.8 - $6.0
Overture	$4.3 - $15.0	$4.3 - $15.0	$4.3 - $15.0
Small ppc machines	$3.7 - $13.3	$3.7 - $13.3	$3.7 $13.3
price comparison machines	$1.5 - $2	$1.5 - $2	$1.5 - $2
eBay	$2.7	$4.2	$ 5.9

These are very notional figures and assume goods priced attractively and presented well, i.e. achieving fairly high conversion rates. If Dale's poor 0.8% conversion rate applies, then the marketing costs are many times those shown.

Auctions are popular. Here are the top auction and auction-like sites worldwide: quality is rather mixed.

ebay.co.uk	ebay.com	ebay.com.au	ebay.ca
ubid.com	ioffer.com	bidz.com	mightybids.com
ebid.co.uk	deremate.com	tradera.com	bid4assets.com
edeal.com (now Truition)	turners.co.nz	auctionfire.com	ebaycareers.com
bidchaser.com	shopgoodwill.com	auction.com	auctionweiser.com
auctionhints.com	bidchaser.com	business- auctions.com	auctioncity.co.nz
bid-alot.com	auctions4acause.com	bidchaser.com	ebayliveauctions .com
priceline.com	ebaymotors.com	qxl.com	auction- warehouse.com
buyme.co.nz	directmylinks.com	-	bidchaser.com

Points to Note

A simple business model would have identified the weaknesses in Dale's plans from the outset.

1. Customer segments: unknown and untested.
2. Customer channels: Internet-facilitated: Dale had no experience here.
3. Customer relationships: poor to nonexistent.
4. Key resources: none: Dale knew nothing about the spirits trade.
5. Key partnerships: distilleries: Dale has no proper relationship with them, or the resources to buy wholesale.
6. Key activities: selling fine whiskies.
7. Value proposition: resourcing fine whiskies: Dale is again without the necessary knowledge, experience and contacts.
8. Cost structure: few improvements possible: why would customers trust Dale rather than order through their usual supplier?
9. Revenue streams: poor, as Dale has no competitive advantages.

Questions

1. What should Dale's first question have been?
2. What, overall, were Dale's two greatest mistakes?
3. Analyze Dale's performance under the Osterwalder and Pigneur business model.
4. What crucial benefit was offered Dale, which he turned down?
5. Cost an effective research program for Dale.

Sources and Further Reading

1. Pricing Central. Price comparison shopping portal.
2. Recommend.org. Short but useful listing of price comparison machines.
3. Marketing Experiments. Experimental data, courses and partnerships.