Friday, 21 December 2012

Some more ratios - and why does wikipedia need my donation?

Having started to delve into the annual reports / 10-K for Amazon and Asos for my last post, I thought I'd extract and compare a few more data points from their published information. Firstly their fulfilment centre density.

Asos have just opened a new 1.1M sq ft facility (for metric readers, 1 square metre = approx 10.7 sq ft, 1:10 is an easily memorable ratio). According to their published figures, they expect to be able to serve £1.2 Bn of sales from this warehouse: a ratio of £1090 per year per sq ft. This fairly high figure presumably reflects the relatively high price points of clothing.

By contrast, Amazon publishes (non-services) sales of $42Bn from 44M sq ft of warehousing (although it is not clear how much of this warehousing space is already open, and how much is secured/under construction for the future). Converting the currency, this is equivalent to around £600 per year per sq ft. Interestingly this figure is remarkably close to the ratio for Ocado, the online only grocer, at around £630 per year per sq ft; this either reflects very well on Ocado - typical price points for grocery are around £1 per item - or rather badly on Amazon.

(It's also interesting to compare with sales / sq ft / year in a typical supermarket of around £1000 in the UK, although less than half that in the US.)

Secondly, sales per visitor. Asos also publishes visitor numbers in its annual report, based on comScore data. Most online retailers seem rather reluctant to publish these numbers, but comScore occasionally publishes data points for huge sites like Amazon (282M visitors in June 2012 worldwide) which can be used for estimation. With these caveats on the reliability of the numbers (and some allowances for seasonality), it's possible to estimate revenue-per-visitors figures.
Assuming the data is reaonably trustworthy, it becomes more evident why Amazon is such an effective machine: the cash it extracts per visitor to its website is quite spectacular. Almost 5 times what Asos manages, and treble eBay - and the eBay figures include the income from paypal. (Note that this is eBay's revenues, not the total value of transactions on eBay).
Although this is nothing to do with multichannel retail, on another tab I am now looking at yet another appeal from Wikipedia for donations. Why? I know it is supposed to be a wonderfully pure site, unsullied by commercial interests, unbiassed in its editorial approach. Would it, I wonder, be compromising its principles too far by becoming the biggest Amazon affiliate in the world? Most Wikipedia entries (should!) have a list of citations and sources; many of these are books. Would it be a commercial step too far to advertise just the books that are specifically in the citations; a wikipedia citation is almost an advertisement anyway? And Wikipedia has visitor numbers that make Amazon look really rather pathetic - at least double. Just a thought...

Tuesday, 13 November 2012

Asos vs Amazon; Delivery vs Marketing

In the P.S. to my last post on delivery charges, I highlighted the gem in the small print of Asos's latest annual accounts: reclassifying delivery costs as a marketing expense. This seemed an interesting enough idea to take a look at what Amazon does, and in fact to generally compare the two. (Especially in the light of rumours that Amazon is considering making a bid for Asos, presumably along the same lines as its Zappos move a couple of years ago).

So, a trawl through the latest Amazon 10-K, and sure enough, Amazon also makes a similar comment:

"While costs associated with Amazon Prime memberships and other shipping offers are not included in marketing expense, we view these offers as effective worldwide marketing tools, and intend to continue offering them indefinitely."

Slightly different approach, but basically the same statement: free/discounted shipping is a marketing cost. How big a cost? Helpfully Amazon provides the information. In 2011, income from shipping fees was USD 1.5 Bn, and shipping costs were USD 3.9 Bn i.e. the net cost of this "marketing tool" was an eye-watering USD 2.4 Bn worldwide! Compare this with the actual spend on marketing of USD 1.6 Bn. Shipping offers cost Amazon 5.1% of turnover (up from 4% the previous year), compared to marketing at 3.8% of turnover. In reality, 5.1% actually understates the figure, because the turnover includes a substantial slice of income from services and fees (such as marketplace and hosting); with these excluded, the shipping offer actually represents almost 6%.

A look at the same numbers for Asos reveals shipping fee income of GBP 10.7M, 2.2% of sales compared to Amazon's 3.5%. Asos doesn't actually directly break out the cost of its shipping, but it is possible to estimate it from the other numbers published, to be around GBP 25.5M, around 5% of turnover. So the net cost to Asos of treating shipping as a marketing expense is GBP 14.8 M, or around 2.9% of sales. Not quite as startling as Amazon, but certainly comparable. However Asos spends a slightly higher percentage of revenues on "true" marketing, around 4%, so it has not YET reached the point like Amazon where free/subsidised shipping is its primary marketing tool. Judging from the statements made in the accounts regarding ongoing developments in this area, however, it won't be too long before this is so.

Of course the other big difference between the two is that Asos can afford it! It is genuinely profitable - net profits are around 8.3% of turnover, despite continuous investment in overseas growth, new IT systems etc i.e. all the excuses that Amazon seems to use to explain its perpetual hovering around the boundaries of break-even: profit was 1.7% of turnover in 2011, and it actually made a loss in 3Q12.

Obviously this difference is not unconnected with the difference in gross margins. Asos has gross margins of 49%, Amazon only 22%, a situation it dismisses with the explanation:

"We believe that income from operations is a more meaningful measure than gross profit and gross margin due to the diversity of our product categories and services."

While this is probably plausible, measuring gross margins for an online (or mail-order for that matter) business is not meaningful in another way; I believe the right measure should anyway be delivered margins i.e. including delivery fees, shipping costs and returns processing. And given Amazon's mix of categories, the underlying implication in these numbers is that quite possibly Amazon is operating some categories below break-even delivered margin.

Getting off-topic a little, investors seem to still believe in the Amazon go big or go broke strategy, and don't require it to make reasonable profits, presumably on the assumption that if it eventually stops investing in growth then actually the underlying business is profitable. If you are the punting type, you might fancy a bet on the dual scenario that the US as a whole follows the trend in some states to put purchase taxes on an even footing between offline and online, and then that BestBuy (and others) accept the logic published recently by Media Markt (see my previous post) and go for a big cut in gross margins themselves, thereby putting themselves on a more even footing on price with Amazon. Whither then Amazon?

What then should a true multi-channel retailer do with regards to this whole "shipping as marketing" idea? Firstly, take a look at this photo, taken in a London Underground station recently:

Amazon Locker, Hammersmith Station

Yes, it's one of Amazon's attempts at click-and-collect. But... there's no in-store additional sales to help the business case along. As I suggested recently, customers like click-and-collect, but retailers like it even more because it leads to incremental sales. And a certain lack of convenience doesn't seem to hamper customer take-up: Marks-and-Spencer stores, for example, are not exactly handy in general - they tend to be in town-centre locations not residential areas, especially outside of London.

My proposition then, is that multi-channel retailers should reinforce click-and-collect, using the same mindset that leads Asos and Amazon to treat fulfilment as marketing cost, but focussing on stores as a competitive advantage. The most obvious way to do this is some sort of coupon/voucher that is valid for further spending when associated with a collection. I'm not aware of this being done yet, but I'm sure someone somewhere is already on to it - the results will be interesting.


Monday, 15 October 2012

Delivery pricing - an afterthought

I've just been reading the latest Asos plc accounts. And in the small print, I found this gem:

"From 1 April 2011, the Group has reclassified delivery costs from cost of sales to operating expenses to reflect their increasing deployment as a marketing expenditure. Prior year comparatives have been reclassified accordingly."

Great advice from an online retailing star!

Delivery pricing

Firstly a big thank you to those who sent me feedback on my last post about click-and-collect (and especially to those who "liked" it or forwarded it to all their contacts in turn)!

And now, a warning...

You cannot proceed because you have not reached the minimum order value of €19. Eh?? And this is only the German site. If you want to try the same thing in Belgium, then it's €25.

Congratulations to C&A on possibly the most unorthodox way of avoiding "sticker-shock" at checkout I have yet seen! Doubtless they can be extremely confident that their online customers are not going to abandon their carts due to being unhappy with surprise delivery charges. And on the other hand, delivery is free if you spend more than this minimum. But this does seem a very  strange way of emphasising their very strong "free delivery" message - by hiding it competely on the site homepage, and then jumping on you later if you try and checkout.

Better (best?) practice is demonstrated here by John Lewis. This is the top-left on their homepage (the red circle is my addition):

The free delivery message is considered so important it takes pride-of-place just below the navigation and above the hero product offer.

The C&A "alternative shock" approach seemed unusual enough to prompt a bit more research, and at least validate that my shock was not reflective of some British bias. I've taken a quick look at a few top British, French and German sites:


And no, nobody else is trying this "minimum cart size achtung" approach! No surprise there then... However the first surprise is how hard it is to find this information. Consumer unhappiness with delivery pricing is THE top reason for cart abandonment (assuming you have a basically clean-functioning site). From Forrester's 2010 cart abandonment reasons study:

And the top consumer expectation of a website is that pricing and shipping information is clear:

So why hide it? Customers demand this information. No points to, whose help pages were simply not working (it's a priority guys, not an annoying bit of the website that doesn't matter much). But particularly on the German sites, it is remarkably difficult to find the facts. A standard footer would be a strong recommendation (example from John Lewis again);

Second surprise is how few sites (and not just top sites) offer free delivery above a threshold level. Free delivery over threshold is a very good idea for a few reasons:
  1. checkout conversion rates are known to be lower at psychologically critical price points (it's the old $9.99 thing again), especially at the critical 3-digit point in dollars, euros or pounds. If you want those €95 carts to convert - a figure remarkably close to the average cart size on many sites - don't slap a delivery charge on which takes it over the €100 mark.
  2. customers will add an extra article to their cart to get above the free delivery threshold. Set your free threshold to just above your typical cart size!
  3. any free delivery message is a very powerful messaage
  4. turning away orders (like C&A) is turning away all distress-purchases. Given that a primary driver for customers to use online is convenience, eliminating all those potential customers having a panic-buy moment for that item they desperately need for their summer holidays is a big loss of trade. OK, servicing small orders is expensive, but customers will pay for this convenience. Make the charge standard, waive it over any reasonably threshold.
The ubiquity of delivery charges also implies another misconception (I see this quite often when I work with clients new to online): multichannel is not free, and is no more/less profitable than stores. There is a cost to having stores: expensive space in town-centres and shopping malls, presentable staff, distribution networks. The customer comes to this expensive space. In effect you have paid a high cost per unit sale to have the customer come to you. In a non-store/online model, you pay a cost per sale to take yourself to the customer - fulfilment centre, picking/packing, shipping. The costs are (or should be) comparable per unit sale.

Very small online orders do need a delivery charge to be applied, because they are disproportionaly expensive to handle. The cost per unit sale for larger online orders is most likely comparable or lower than the cost of the equivalent store sale. Attempting to pass this cost onto customers, when they would not have paid the cost in store (has anyone tried charging a customer £3.95 to take their purchase through the store exit door?) is an artificial charge that is costing you sales.

This conclusion is starting to become particularly clear when you look at the cost of delivery for large articles such as white goods. To ship a washing-machine to a customer costs around £30-£35 in the UK. John Lewis charges nothing for white goods articles over £50, Tesco Direct charges a flat £7. Customers won't tolerate having the high cost of shipping passed through to them: it's not added in store, why should they pay it online? Sites such as Carrefour electricals (France: €59.99 (!!)), Baur (Germany: €39.95), MediaMarkt (Germany: €34,95) need to rethink their model sooner rather than later.

Monday, 24 September 2012

Click & Collect - Calling Time On Store Space?

Philip Clarke, Tesco's CEO, made a speech last week stating that: "...we’ve called time on the old retail 'space race'. We’ve recently opened our 1,000th click and collect collection point..." Is click-and-collect really quite so game-changing?

Well actually, maybe... But it's multichannel game-changing, not necessarily traditional retail-space game-changing by itself.

Not all top UK retailers publish the data, but from those that do, we can see that customers really like this clicks-then-bricks option:

** of eligible General Merchandise sales (i.e. excluding food and impossible items like washing machines and sofas)
* John Lewis state that 34% of John Lewis sales are collected in Waitrose Stores! So actually their figure is probably higher in total. Click-and-collect was offered at 97 Waitrose's and 35 John Lewis's

The latest published data from IMRG indicates that over 10% of all online transactions are now collected in store (up from 7.4% in the previous quarter): more and more major retailers are introducing it, and the take-up is often dramatic.

Some published commentaries are also quite illuminating. First of all this quote from Halfords annual report:

"Our product mix lends itself to a multi-channel offer as customers often want further advice, a demonstration or fitting. Online purchasing patterns reflect this, with 86% of sales on reserved and then collected from a store"

In other words, these customers are basically using the Halfords website as a place to guarantee that the item they want is definitely in stock in a nearby store when the customer makes their visit. Once at the store, the store has little value as a showroom or place to transact, but very high value as a place to get added services impossible to execute online. Interestingly this tends to help validate the KnowHow based strategy for PCWorld and Currys - but it does require your products to be difficult for customers to just point-and-shoot in the first place. These Halfords customers aren't using the store-advice to choose their item, they are using it afterwards to configure it. This is a great differentiator against the online pureplays for bikes, and probably therefore also for laptops, but less good news for e.g. TVs.

Secondly, a note from Sainsburys:

"Customers use Click & Collect for about half of all online general merchandise orders – a figure which rose to 75% for the week before Christmas 2011."

Or in other words, if you could only trust the postal service, then click-and-collect might be less attractive. An alternative explanation is also possible: click-and-collect is almost always offered for free. Customers hate paying delivery charges, and faced with the option of a free collection service compared to a paid delivery, have a natural bias towards the perceived free service, even when there is a hidden travel/time cost. Of major UK retailers, only TopShop seems to be attempting to charge the same for click-and-collect as it does for home delivery. Unfortunately they don't publish statistics indicating how the take-up varies with these fees.

Another quote from John Lewis tends to suggest this TopShop approach is seriously misguided anyway. This, to me, is the most significant data point in the whole click-and-collect space:

“We are seeing about 34% of those visits translating into additional sales in shop and that number is growing exponentially at the moment. It’s typically or increasingly for purchases that the customer didn’t think they would make. So it is quite outwith whatever they were going to collect.”

Customers want to click-and-collect, and then when they do, they find themselves buying extras in the store as well. Sounds like the ultimate retailer win-win.

It's noticeable that fashion retailers are lagging behind this curve. M&S, New Look and TopShop do, but surprisingly few others. Possibly this reflects the in-store space challenges, but Tesco's 1000 Click-and-Collect points now include quite a few Express and Metro stores, especially in central London. If Tesco can fit a collection point in an Express store for awkward boxed items like PCs, then surely it can't be so hard to fit a few parcels in e.g. a River Island store. In fashion, with its high returns rates for home-delivered orders, bringing the customer to store to a) spend more; b) try on, and return or exchange in the store environment must surely make sense?

Perhaps the most interesting experiment in this area in the UK right now is House of Fraser's Buy & Collect only store in Aberdeen.

"...its new 1,500 sq ft House of shop is the first to offer purely a Buy & Collect service... The new format, merchandise-free, store has opened in Aberdeen’s Union Square. Instead of stocking goods that shoppers can take away with them, the emphasis [...] is on personal customer service. Goods ordered from the more than 1,000 brands it stocks can then be delivered the next day to either the customer’s home or to the store for collection."

Another similar shop (in Liverpool) is on the way, suggesting that the first pilot must be going pretty well. Other retailers, watch this space!

Wednesday, 19 September 2012

Ocado interim statement: not much to get excited about

The Ocado 12 week statement is out this morning.

And there isn't very much to get excited about. I argued in a previous post that the fundamental problem with Ocado's business model is that it is designed to break even, not make a profit. As the business scales up, the break-even scales up with it: the profit margin on each delivery is almost exactly equal to the cost of making it. Sales continue to grow steadily (9.9% year-on-year), but there are no obvious pointers to profit growing with them.

There are some small hints for those who want to take an optimistic view. "We have been able to preserve our margins during the quarter" is essential news for a business model so finely balanced between profit and loss. "The Hatfield Customer Fulfilment Centre continues to operate with improved efficiency... enabled us to remove all remaining... trolley picking" means that costs-per-order should reduce slightly.

More interesting is a rise in the average basket size from £111.08 to £112.44 (1.2% year-on-year). On the one hand, this is at least reversing the downward trend. On the other hand, this compares with Food CPI rise of 4.5% (in the latest available figures from the ONS i.e. March 2012). The implication is that the number of items-per-basket might actually have dropped. The quarterly interim statements don't quote this number, so we'll have to wait for the end-of-year results. If the items-per-basket figures really has dropped (again!) , this is very bad news.

For this business to ever become profitable, items-per-basket must rise. On a rough calculation, for profit to rise to a reasonable 9% of sales, average items-per-basket needs to rise by about 25 items. This is a huge hurdle, and the signs continue to point the wrong way. A drop in average items-per-basket also tends to suggest that the strategy of moving into non-food is not exactly flying. While this is not exactly a surprise, given how competitive online non-food is in the UK, it's a pillar of Ocado's published strategy and it doesn't look like it's working yet. Another key item of data to look for in the year-end figures.

So all-in-all: no news is bad news.

Sunday, 2 September 2012

The Multichannel Halo Effect: Faith or Fact?

In my last post, I reviewed some of the (many!) published data points supporting the argument that there is a Halo Effect from multichannel, or in plainer English, that multichannel customers spend more. There are claims from perfectly respectable sources that this can be as much as 300% more than single-channel customers.

As any statistician will tell you, however, the problem is that correlation does not imply causality. Is it not just as likely that your best customers prefer to use all your channels anyway? They were your best customers when you only had stores, they are still your best customers now you have lots of ways to shop.

This research study, which unfortunately doesn't say who the retailers in question were although there are a few big hints, contains a perfect illustration:


Yes OK, multi-channel customers do spend more. The trouble is they really are simply your highest spending customers anyway, and are probably spending more everywhere they shop. This tends to be borne out by the subtext of all those other data points suggesting multichannel customers spend more. If you read more closely, you will also find those same studies mention that they tend to be on average more affluent i.e. they have more to spend in the first place (notably on things like laptops, smart-phones, iPads, high-speed broadband etc which encourage them to use all those channels): the argument gets circular. These customers expect you to have all the channels available to them - and they'll start to go elsewhere if you haven't.

We need to go back to some basics. In another previous post I used data from the UK Office of National Statistics to illustrate a very important point: eCommerce (etc) has NOT created any new money in customers' wallets. These wonderful multichannel customers do not magically have more cash to spend.

Removing food from the picture (online grocery tends to distort UK data for comparitive purposes with most of the rest of the world) and looking at just non-food, this point becomes even more apparent. Take a look at the overall "growth" of non-food offline-only non-Food sales over the last 10 years:

Non-store sales have essentially remained static over that time. Any growth has come from Online. If you factor in some measure of inflation (I've used RPI because the data is easily accessible), then the picture is that non-store sales have seriously declined:

Basically, allowing for complicated effects like cheaper prices of manufactured goods from China, the big picture is of an approximately zero sum game (no magic extra money in customer's pockets), with Online taking an ever increasing share. And factoring in the "multichannel customers spend more" data, the share that Online takes is a disproportionate share of your best, and probably most profitable, customers.

I often get asked "won't setting up a new channel cannibalise our store sales?". The answer is yes of course it will. But if you don't do it, someone else will - and what's more they'll cannibalise your best customers.

It's not really a halo effect - more a set of horns and a long pointy tail effect.

Why bother with multichannel? - answer #3: multichannel customers spend more

As I mentioned in the last posts, a challenge I am often posed when I work in less developed internet-retail markets (Poland or Belarus being recent examples) is "why should we bother with multichannel? It is organisationally challenging and expensive, and we are doing fine right now."
So here's yet another answer: multichannel customers spend more. This is actually a very well documented statistic with a long history (the earliest data point I found while writing this was from 2004, but I'm sure someone will point out an earlier one).

How much more? This is rather more difficult to quantify. Laura Wade-Gery, Multi-channel director at Marks and Spencer, can be seen in today's Sunday Times (business section 2nd Sep '12) claiming multichannel customers are worth 4x store-only customers. Anecdotal figures I've heard talking to colleagues and clients are usually in the +50% to +100% range. John Lewis, generally considered one of the most sophisticated multichannel players around, seems to side with M&S appear to be quoting a multiplier of 3.5x, in this study with IDC:

"These multichannel and omnichannel shoppers: spend 3.5 times more; purchase across more categories; shop more frequently; are more loyal and have a higher retention rate"

What more could you want?

And it isn't an effect confined to upmarket UK retailers like John Lewis or M&S. A 2010 study for the US National Retail Federation found that: "39% of retailers describe cross-channel customers are significantly more profitable than single channel customers. In 2007, this figure was 18%." Of course this might be that US retailers have become better at measuring it, rather than any trend change, but nevertheless the effect is still reported consistently.

OK, these customers are more profitable, but by how much? IDC research from late 2009 apparently provides the answer: "findings show that, while multi-channel shoppers spend, on average, 15% to 30% more with a retailer than someone who uses only one channel, omni-channel shoppers will spend 15% to 30% more than multi-channel customers." Ah yes, it's those omni-channel customers again.

Accenture always likes to get in on the act, so here's the data from their more recent (Jan 2012, and European not US) equivalent study: "more than three quarters (76%) of the retailers surveyed report the multichannel customers spend more than their single-channel customer counterparts, and one third (32%) said that multichannel customers spend at least 26% more than single-channel customers."

Deloitte seems to think this is all a bit too conservative: "Multi-channel customers spend 82% more per transaction... The average expenditure for multi-channel customers across the clothing, home and electrical categories is £116 per transaction compared with £64 for store-only customers".

Everyone likes a graph, so Deloitte helpfully provide one in this report. Unfortunately it's not apparently supported by any referenced data/survey, so it's completely unclear if these figures are just plucked out of thin air. The impression from the rest of the study is that they may just be "illustrative" to use a polite word. But it's a nice graph so I'll show you it here:

Lots of nice growing bars!

Incidentally there is the ominous (if dubious) deduction possible from the Electrical bars that stores are a complete waste of time... See my previous post on this topic.

There's a nice piece of jargon for all this: the 'multichannel halo effect', which I think has made the leap from brand-marketing to retail. I rather like the pseudo-religious connotations of this term: it would certainly make convincing my clients easier if the question "why bother with multi-channel?" could be answered "because God says so"!

Monday, 20 August 2012

Why bother with multichannel? - answer #2: it's where the growth is

As I mentioned in the last post, a challenge I am often posed when I work in less developed internet-retail markets (Romania or Belarus being recent examples) is "why should we bother with multichannel? It is organisationally challenging and expensive, and we are doing fine right now."

A second answer is that, even in a rather stagnant economy, multichannel just keeps on growing. An interesting illustration comes from Office retailers in the US. Both Staples and Office Depot sell to a mixture of B2C and B2B customers. Although they have each had varying fortunes, non-store sales as a percentage of total have just continued to grow and grow:

At over 40% of total sales, non-store is pretty important!

Perhaps even more significant, however, is a closer look at recent numbers from Staples, comparing sales via brick-and-mortar with sales via online:

As the credit crunch hit, store sales actually declined by almost 7% between 2008 and 2010. But online sales grew by 32% in the same period, more than enough to ensure that overall Staples kept on growing.

Office Depot is something of a horror story in general:

But at least online sales declined by only 16% compared to a whopping 30% of brick-and-mortar.

Sunday, 29 July 2012

Why bother with multichannel? - answer #1: follow the money

A challenge I am often posed when I work in less developed internet-retail markets (Romania or Ukraine being recent examples) is "why should we bother with multichannel? It is organisationally challenging and expensive, and we are doing fine right now."

The simple answer is, that's where your customers' money is going, and if your country follows the trends in others, going quickly. This is 2011 data published by Ofcom for various European countries.

Much more telling is a look at the trend in retail spending in the UK. (Data sourced from the Office of National Statistics).

One observation is obvious: the growth in eCommerce sales has been spectacular. The second observation is probably more important: eCommerce does NOT represent "new money." Overall retail sales are either down (non-food) or merely on-trend (food). What we can see is a switch in spending, not additional spending.

This is the channel customers are choosing to spend their money. If you're not on it, your competitors will be, and your customers will go there.

eCommerce now represents over 10% of all retail sales in the UK, so it is genuinely meaningful to draw such charts. The interesting question for the next few years is whether it will soon become relevant to draw similar graphs charting the growth of Mobile. Countries with well developed landline broadband such as the UK or US may be less interesting studies than those such as India where over 50% of the population has access to a mobile phone, but less than 3% of households have landlines or broadband.

Sunday, 15 July 2012

Not all channels are equal - at least in the eyes of the finance team

One of the more bizarre obstacles I have encountered when working with clients adding eCommerce as a totally new channel to an existing brick-and-mortar retail operation is the accounting treatment. OK, I'm asking for accountants to use their imagination - not always an easy stretch.

But actually the equivalences between brick-and-mortar and virtual channels are very important for two reasons. One, they are essential to properly understanding the business case. Two, they are even more essential to moving from multiple channel to true multichannel in a later phase of the strategy.

I'll start with the less controversial ones. Of course it's all a matter of opinion, but this is my starter for 10 (well OK, 6):

1. both channels probably have warehouses (strictly DCs - distribution centres - for stores, and FCs - fulfilment centres - for online). Some retailers even manage to combine both operations in one location, despite this being actually quite a tricky blend of operational processes.

2. both channels have a store operations team. In a store, there is a store manager, sales assistants etc. Online, there should be a web-site operations management team, organised in various ways. They perform approximately equivalent roles.

Now the going gets a bit trickier.

3. Brick and mortar has store fit-out. Online has a website. If you are going to compare (and eventually seamlessly blend in true multichannel) they need to be treated similarly from a business case perspective. Quite often this means equating store-fitting opex with IT capex, although as SaaS models such as Demandware increase in popularity maybe this distinction may become less acte.

4. Brick and mortar has stores. Online has delivery/fulfilment costs. In the former case, you operate places on the high street to which you distribute your goods for customers to receive. In the latter, you take them to customers' homes for them to receive. The processes are logically equivalent. Fulfilment costs are typically a high part of the cost of an online channel. Store rental is typically a high part of the cost of a brick-and-mortar channel.

And now for the really controversial part.

5. Standard delivery fee income is NOT sales. It is a contribution to your marketing budget. If you accept that delivery/fulfilment costs are equivalent to store rental costs, then delivery fee income is not a way to reduce the fulfilment budget. Imagine the store equivalent: suppose you could charge each customer a few pounds service-fee to be allowed to use the checkout in your stores. Surely you would not really treat this as "sales", (and count it into gross margins)? In the same way as free delivery is a huge driver of trade on a website, allowing your customers to checkout in stores for free does actually help persuade customers to make a purchase.

OK, you might book it into your general ledger as income. But for business-case purposes, and for targetting the web channel, don't treat it as sales income.

6. Non-standard delivery fees are somewhat different. They fall into 3 kinds. Firstly expensive charges for expensive deliveries such as for pallets or 2-man products. In this case, the fees should genuinely be offset against the delivery cost. Secondly, delivery-related services such as installation. These are sales. You have persuaded the customer to buy an extra (service) product. Thirdly options like express delivery, where you may feel it is appropriate to treat these as a (service) product with an associated cost-of-goods/service and resulting margin.

For most retailers, looking at things through this "equivalence" perspective has a number of major benefits:
  1. it allows a level playing field comparison of channels, and avoids distorting behaviour leading to artificial channel conflicts.
  2. it makes the trade-off between the different capex and opex elements of the business cases for each channel far more transparent
  3. it reduces the sense that margins are being distorted by multichannel
  4. it makes a seamless transition to true multichannel much easier to implement later on, especially for truly cross-channel activities such as click-and-collect
Even if you don't like my particular equivalences, select some that you feel are appropriate to your particular business model.

Sunday, 8 July 2012

Eastern Europe anyone? Ofcom statistics about Poland

With its economy closely coupled with that of Germany, Poland appears to have escaped the worst of the credit crunch. Economic growth in 2011 was over 4%. It should be fertile ground for retailers looking for new markets, especially with its strong links to the UK.

One of the most interesting charts produced by Ofcom recently (2010 data) is an analysis of the enthusiasm of online consumers - how often do you buy online? Poland is a very interesting edge case:

OK, spend per head is not so high, reflecting its relatively low GDP per head, the value of the Zloty, and its economic development stage. But just look at that spend frequency: second in Europe only behind the UK! (and the UK is a weird edge case anyway, nowhere else are consumers as enthusiastic about online retail). There is anecdotal evidence that other very developed E.European markets, especially Czech, show similar behaviour, but unfortunately no data.

And online customers are growing fast, as this data from the 2011 Ofcom report shows:

Already ahead of Italy and Spain (where customers seems to like to browse online but not purchase), catching up rapidly on Ireland, and showing similar growth to France.

So if you are looking to reach out to another customer market, have a serious think about translating your site into Polish... There are some other advantages to Poland too. Delivery infrastructure is well developed, with major players like DHL well established, and payment methods are less "eccentric" than some other apparently easier potential markets (e.g. Netherlands, France, Germany, Belgium or Denmark all with challenging local schemes).

There's just one snag... in some categories everyone else is there too, especially the leading retailers from Germany and France. If you fancy somewhere a bit easier but with similar characteristics, try Czech. There are only 10.2M Czechs compared to 38.5M Poles, but there's a lot less competition as a consequence. Tesco are, however, using it as their pilot country for eCommerce in E.Europe, so move fast before it's taken!

Sunday, 1 July 2012

Is taking a page out of Booker's book an opportunity for Ocado?

Another set of Ocado sales figures, another half year of no-profit growth.

The latest figures continue to underline the basic issue with Ocado's business model. Quite simply, on a cart-by-cart basis they've just about made it break-even. But on that same basis it's extraordinarily difficult to see where the profits are going to come from.

This picture tells the story:

Ocado enjoys pretty good margins for a Food retailer, but these are entirely swallowed by the costs of delivering each order. Margins are being squeezed, and delivery income is being squeezed by increasing competition.

The only way forwards is to increase cart-sizes. But year-on-year, cart-sizes actually fell  by almost £2 between 2010 and 2011 as customers cut back.

One option to increase cart-sizes to to increase the range of products on offer, which Ocado continues to do. The problem with this approach is that eventually you start to extend into categories which are not naturally part of a weekly grocery shop. Health & beauty: maybe. Consumer electronics or fashion: probably not. And as Ocado themselves observe in their analyst presentations, non-food requires a different distribution model. In other words, it doesn't really play to their strengths. Worse still, it sets them into online competition with Amazon, John Lewis, Tesco Direct, Argos, Debenhams, Marks-and-Spencer...

But while Ocado has been slugging it out unprofitably with the supermarkets, another online Food retailer has been quietly growing its online business even faster than Ocado: Booker Cash & Carry.

And while Ocado has a minimum cart-size of £50, Booker has two minimums (depending on whether you are a restaurant or corner-shop customers): £100 and £500. What would Ocado give for some £500 shopping carts? Maybe a bit of that 26% gross margin - cash-and-carry margins are inevitably lower. And e-commerce in the B2smallB sector is a tough nut to crack. These customers expect B2bigB service, B2bigB prices, and B2C charges.

But with all that automated infrastructure, you can't help wondering if there's an easier way to profitable growth than slugging it out with Tesco and Sainsburys.

Tuesday, 26 June 2012

Are shrinking stores inevitable in consumer electronics?

"Best Buy has set plans to reduce its store square footage by 10% in the next five years."

The continuing threat to the business model of consumer electronics retailers posed by the internet seems to have elicited a consistent response: "honey I shrunk the stores." Or in the case of Comet/Kesa, sold the stores.

These retailers are determined to protect margins. Protecting margins means that price-based competition from pureplays eats into store sales. Ergo, fewer/smaller stores.

Protecting margins is rather understandable when you look at the panicky 15.5% drop in the share price of BestBuy when margins took a 0.9% hit during the last holiday season.

And yet curiously gross margins are historically quite high. US GAAP helpfully forces retailers to publish true gross trading margin figures. And Best Buy's margins look like this:

Pretty steady in other words, and despite the recent drop due to trying free-delivery for the holiday season, historically really rather high. Take a quick look at the longer historical trend:

In other words, since the dawn of the internet age (and of course the dawn of the Chinese manufacturing age), Best Buy has enjoyed historically high margins.

Dixons and Comet are rather coy when it comes to stating gross margin figures. When they are down, they tend to say things like "significant margin pressure" while when they are up they crow about "0.4% increase in gross margin". Solid data is rather absent, but a trawl through the financial reports for the last few years seems to imply a reduction of somewhere between 2% and 3%, recovering slightly recently, hardly a disaster.

Dixons and Comet's sales, of course are another story (although this doesn't - yet - seem to be such a problem for BestBuy). Effectively they've traded preserving margin for lost sales. But is this the only approach?

German giant Media Markt-Saturn thinks not. OK, they've got an extra problem - their stores are partly owned by franchisees, so announcing a programme of store closures is not really an option. Instead they've decided to go straight to the heart of the matter, and announced that they will reduce prices... by 5%-6%.

In other words, they've decided that they are actively going to try to protect sales (and their franchisees), by accepting a loss of margin instead, and planning for this strategically.

It has to be said that they've had a rather torrid time trying an alternative approach. This horror-story is from their 2010 pilot in Austria (where they tried it out before risking their core German market):

My online customer journey:
  1. I'm attracted to MediaMarkt by their slogan "I'm not stupid, (because I shop at Media Markt)"
  2. I go onto the site, where I have to declare my home store (instant loss of 50% of customers)
  3. I find that a) prices are cheaper at a different store; b) their online price is not cheap at all
  4. I decide I am indeed stupid because I shop at Media Markt. I go somewhere else, probably online
  5. I decide that Media Markt are also stupid, because their pricing strategy is a disaster area
The net result of this monsterpiece was the departure of the CEO and Finance Director, and a fresh think.

And now they've had a fresh think, they've reached a completely different conclusion from Best Buy. Quite simply, they have recognised that store price = online price is mandatory, (which is a conclusion all true multichannel retailers eventually reach: the customer experience is just absurd otherwise.) And then they've drawn the logical consequence from this - prices will have to fall:

You can find the full presentation at this link.

The summary is actually quite simple:
  1. we have to have one price across all channels
  2. in order to do this credibly, we have to reach down to median internet price or lower
  3. to do this prices have to fall, by about 5% from shelf price, and 3% from customer price (customer note: it's probably worth trying to negotiate a bit in a MediaMarkt or Saturn store)
  4. in order to make this possible, we will reduce the cost base (but not the store base) and also simply accept lower margins
It will be interesting to see if BestBuy follows suit. And if so, what it does to their share price if a 0.9% margin drop can reduce the value of the company by 15.5%... But then they are currently enjoying historically distorted high margins!

Saturday, 23 June 2012

Fix cross-functional before you try do cross-channel

Having singled-out River Island rather unfairly in the last post, another interesting site issue in which they appear to be in rather elite company.

Their home page featured a rather nice promotion "after dark":

Maybe I'm just not your typical customer, but I did an obvious thing and typed "after dark" into the search box.:

and there it is... no search results. Sloppy, but is this just River Island, who to be fair are hardly a top 10 internet retailer? Let's take a look at a few who are.

Marks and Spencer, for example. Here it is, big banner for an event named "shwopping":

But apparently the M&S search doesn't believe in this nonsense. Instead it thinks I've mistyped the word "shopping" and, better still, helpfully offers to sell me a dictionary to help my spelling:

Maybe this is connected with having their site running on Amazon's system. I'm going to spare you the screenshots, but Amazon is currently running a "summer of sport" promo on my version of its homepage; entering "summer of sport" in their search just offers me some obscure Kindle-only book with this name, followed by a digital camera.

Debenhams goes one better, or rather worse, by apparently not believing in the existence of its own sale event. Here's the home page...

and here are the search results for "half price sale":

"Did you mean halo price sale?" Well no I didn't actually. And I certainly don't want to see a pathetic 2 items, both on discounted promo but neither at half price.

The point of all this is that these sites just don't seem to have their internal teams working together. In most retailers online merchandising and marketing, search, and product data admin are owned by separate functional areas. A typical set up would be to have these three tasks owned by web-merchandising, IT and category management respectively, although it varies in each retailer I've ever worked with. And quite clearly in River Island, Debenhams and M&S, they aren't speaking to each other.

By contrast lets look at a couple who get it right, in two different ways. First of all John Lewis's clearance event. Homepage banner:

And search results for "clearance":

Very simple and effective: the search just drops you straight onto the clearance page. Minimum  site maintenance effort, maximum customer journey coherence.

Finally, Tesco Direct, which goes the whole hog. Again first the banner:

and now the search results for "summer of sport":

Someone has gone to the effort of tagging every one of these 73 products in their product master data with "summer of sport", their search algorithm has been adjusted to prioritise the phrase, and the whole thing coordinated with site merchandising. Full marks for both effort and results, for getting the basics right, and evidently for having a proper cross-functional team managing (and testing!) it all.

Full marks also for recognising that site & search admin is a time consuming, labour intensive task and managing it. It's analogous to refilling the shelves and sweeping the floors in your brick-and-mortar stores: difficult to automate, human-intensive, but an essential success factor.

Thursday, 21 June 2012

No Man is a River Island

Or "how to forget the basic hygiene factors when trying to do international ecommerce".

Today River Island has a special international free delivery offer on its site. Special enough to be the banner on its homepage, and in fact to potentially cost River Island £7 or more per order.


So what happens if I try to take advantage of this offer?

Well, first of all I have to register. Why? Why is there no anonymous checkout option? If English is not your first language, have you ever tried to follow in the instructions on one of these forms that insists on a strong password with a weird '*&!()***' character and a number in it? If English is your first language, try registering on a French or German site now, and see how you get on...

In countries which are more nervous about data protection than British consumers, Germany for example, or Holland which is one of River Island's core target markets, this has already cost them at least 25% of their potential new sales, possibly more.

To be fair, they then jump the next hurdle successfully: the site at least manages not to demand a UK postcode/housenumber combo for customers who indicate they are not based in the UK.

But then, they lose the next 25% of the potential overseas customers with one simple, glaring error: you can't enter an international phone number! Their phone number field only accepts the digits 0-9. So if you try to use the standard international convention of +countrycode, your number gets rejected, and you can't proceed any further. Game over.

                "! we think you've mis-typed your phone number - please try again"

No I haven't, I'm just trying to take advantage of your free delivery offer, but my phone number looks like this: +44 (0) 1234 567890.

Then I'd like to pay. River Island has stores in the Netherlands and Belgium, they are key target markets for its international offer. Google for "River Island .nl" and you get directed to the UK site. Fair enough. But...

Dutch customers like to pay using a scheme called Ideal. In fact >50% of all eCommerce transactions in Holland are conducted using Ideal. Does the River Island site support Ideal? Of course not.

Now of course it's quite challenging to incorporate all the possible local schemes into a website checkout, even if you have a worldwide payment gateway. In most of western Europe, however, there is a quite reasonable alternative: PayPal. However their site doesn't support it, and in fact goes so far as to provide a separate help page explaining that no, they don't support PayPal.

Yes, of course they support MasterCard and Visa. But European customers are much less likely to be prepared to use such a card online than the British are, even if they have one (in a country like Belgium, another core River Island market, they possibly don't).

I wonder if River Island's payment gateway has a default fraud-prevention rule in it blocking non-British card BIN ranges! Lots of retailers do this automatically. Given how little else seems to have been tested, I wonder if they've tested this.

It takes more than adding an international delivery address-box and an international parcel-courier to be ready for international eCommerce, and River Island unfortunately falls at the second and third hurdles. Nul points.