Add to bread basket…

My local bakery in Stepney looks as if it hasn’t changed for decades, so it’s pleasantly surprising to see them embracing the online world with a modern-looking website, in fact two websites for trade and retail customers, and a Facebook page updated several times a week.

When I first discovered the site they were actually selling bread online, with a “add to bread basket” feature a nice touch… Although they don’t seem to be selling online currently, it’s a lovely example of a local business using a website and facebook to win new customers and build loyalty with existing customers.

Facebook has just updated me about the “Almond macs coming out of the oven…” – I might wander over and try some…

Is adwords ever “not cost effective”?

I was talking to the owner of Sunspel, a website selling luxury menswear both online and offline, who came out with the statement “Adwords is not cost effective for us” (not for product-related keywords, only for brand-related keywords).

My main experience with Adwords was for Viners.co.uk where my adwords campaigns generated 50% of sales at a cost per conversion of around 10% of the average order value – so very cost effective, in fact the most cost effective marketing channel in that particular case.

Of course an adwords campaign will not be particularly cost effective if the keywords chosen are not relevant, if ads are poorly written, and landing pages are poor. But the point made by Sunspel was that there were too many competitors bidding on keywords like “T Shirt” or “mens clothes”, and with so much competition it’s difficult to push a luxury message vs. a price-based message.

To some extent I had it easy running campaigns for a site selling mostly cutlery, where the majority of traffic came for a handful of “cutlery”-related keywords. And there are fewer Adwords competitors in the world of cutlery than T shirts. But to a much bigger extent Adwords was cost effective due to constant optimization; stopping keywords and ads which were not cost effective, refining keywords and ads which were more cost effective; constantly looking for trends and opportunities in the long tail of search terms.

Stop designing websites, start designing posters

This is an unusual use of .gif product images on a patterned background in an online shop – perhaps inspired this trend for web design inspired by poster design.

I also like how the price (and all product text) is only shown when you mouse over the product. Not very search engine friendly, but a bold attempt to make the product the hero, and sell the sizzle not the steak.

I Like:

Zappos.com (the biggest online retailer of footwear in the US, now owned by Amazon) has this nice widget to encourage you to “blog about this item”.

I Like:

Ramona by Lassen at Zappos.com
Ramona by LassenZappos.com - Powered by Service

It’s a great example of building engagement with the brand AND helping SEO by building links to their site. Apparently Zappos has been doing social media well for a while, as this chart of inbound links since they embraced Twitter shows:

Incremental sales, or cannibalising from other channels?

Reading about the imminent launch of a Gap ecommerce site in the UK got me thinking about to what extent online sales are incremental, or merely cannibalised from existing channels.
Gap in the UK has been a notable latecomer to online retailing (along with Primark, H&M, Zara, Claire’s Accessories, Morrisons…). In the US it has been trading online since 1997; in the UK it been testing the water with an online offering through asos.com.

It’s interesting reading discussions about why brands aren’t trading online. This one assumes that online sales have the potential to bring in 10% of total offline sales. Whilst this seems a reasonable ballpark for any retailer, it can’t be assumed that the 10% will be incremental and not simply customers who would have bought in store switching to buying online instead. It’s difficult for most brands to assess whether or not online sales are incremental, because there are so many competing factors determining retail sales and customer behaviour. Perhaps in Gap’s case the uplift would not be significant – as Gap’s stores are fairly widespread and I suspect the customer base is fairly loyal. Perhaps this has been their experience in the US, hence the slowness to launch online?

In some cases there’s a more compelling case for incremental sales – for a retailer with stores only in one part of the country, online opens up a much wider market. In my own experience working for a retailer with 15 or so stores located in factory outlet centres I don’t believe there was much cannibalisation from the stores, however there was almost certainly cannibalisation from other consumer and trade channels.

Of course there are many more benefits to trading online; lower cost of sale, brand building, more marketing channels, more flexible pricing & merchandising and the multichannel benefits (customers researching online, purchasing in store and vice versa). It is always better to be selling online yourself rather than through a third party, to be able to control pricing and how the brand is portrayed (on other sites and in their online advertising).

I did find a survey from 1999 predicting only 6% of online sales would be incremental. I would guess the experience of most brands is even lower than this. My conclusion? Not selling online is a missed opportunity, but the opportunity is complex and about much more than just sales.

What is the secret to a successful website?

This is a fantastic article – So true, I could have written it myself…

I particularly like the closing paragraph:

Management must realise just how complex a job it is [managing a website] and ensure you have time dedicated to its execution. Perhaps you should print off this article and give it to them.

What is the secret to a successful website?

Building usability into ecommerce design

With a specialist website, selling one product or category of products, thinking early on about the best way to present products can be beneficial. Not only will you improve usability and conversion rate from the start but it’s an opportunity to make the site stand out from competitors, both the not-as-well-designed specialist shops and the more general retailers using the same layouts to sell a huge variety of different products.

Two contrasting examples I’ve worked on are cutlery and umbrellas.

In the first example, Viners cutlery, I was responsible for the design. Through looking at other websites selling cutlery (good and bad), and understanding patterns of sales through the company’s other sales channels, I made sure each cutlery range was presented with a means of ordering cutlery sets and loose pieces all on the same page (without having to go back and forward to the shopping basket).

Hence this page…

product layouts - Viners

… is the same in terms the product hierarchy as this page…

product layouts - Viners 2

The difference being, a customer is much more likely to purchase more than one product from a range of cutlery (e.g. some knives and some forks, or a cutlery set and some extra teaspoons). With cookware, it’s less likely a customer would want to purchase more than one saucepan. And there’s more detailed information on a saucepan on the product detail page a customer would want to see before making the purchasing decision.

In my second example, Fulton Umbrellas, my involvement was much further down the line, when the design was almost complete. The first design I saw showed around 10 umbrellas on the womens page, mostly black and with technical descriptions like Superslim-1 and Open and Close – 3. This was the way the company presented the range to trade customers, but obviously not the optimal way on B2C online shop; because consumers don’t know the difference between a Superslim and an Open and Close, and because a consumer looking at that page wouldn’t know unless they clicked on it that the black umbrella pictured came in 10 different designs.

Unfortunately the ecommerce platform chosen was limited in the filter and sort views available; the ideal solution would have been to allow the customer to view by type, by colour, by pattern, by price etc. however, the least-bad option for launch was to show all the products.

product layouts - Fulton

What to believe?

“Whizzy imagery out-performs ratings and reviews in Adobe survey”

A survey by Adobe concluded the following are effective in increasing conversion rate.

  • Product tours or multi-media viewing which combines guided spin, zoom imagery, videos or animations with copy (36%
  • Visual filtering and advanced search on product features including size, color, and price (33%)
  • User comments and reviews (32%)
  • Search landing pages (32%)
  • Product comparisons (28%)
  • Zoom (28%)

Well, looking at the numbers not significantly, and it does seem like a convenient conclusion for Adobe, who sell “whizzy imagery”…

Also today I was interested to read this discussion questioning the value of “trustmarks” such as McAfee, with plenty of examples where adding such logos actually decreased conversion rate.

The key lesson from these two examples is to take claims to improve conversion rate with a pinch of salt. Work out what’s most appropriate to your business. Tackle the low-hanging fruit first. And test everything.

Lessons to learn from invite-only online shopping

I became a member of Gilt.com, via an invitation from Gwyneth Paltrow no less!

Gilt.com is a members only shopping site set up to recreate the buzz of a New York sample sale but online – you have to be a member to join (by invitation only), each “sale” of a particular brand lasts 36 hours, starting at a fixed time of day with limited availability – first come, first served. Each sale is previewed a few days before and there are 5-10 different sales live at any one time. Typically a lot of products will show as sold out, but you can join a waiting list.

Prices are typically half the RRP, and I suspect the same bargains are probably available elsewhere online. In addition you may have to wait 2-6 weeks for delivery. But that’s not the point – the site is using exclusivity and limited availability to create a real buzz, and avoiding the need for traditional online marketing. Very clever!

Thumbnail images; Two ways to cope with different aspect ratios

Most online shops standardise the dimensions of thumbnail images – for obvious reasons of neatness. In fact, by default most thumbnail images are either square or slightly portrait.

But what happens if you have a wide range of products and hence images – long and thin products, and short and wide products? How do you maximise the impact of the images whilst avoiding a lot of white space?

Two solutions I’ve found are:

The Holding Company – all images are 170px width, but heights range from 100px to over 300px. This works because the images are vertically-centered on a background colour, balanced with a large text box at the bottom.

The Holding Company

Moss.co.uk – images are all the same width 185px and either 429px tall (full length – suits and trousers) or 215px (tops and accessories). Within each category images are likely to be the same type, so there’s a uniformity – you either look at all the suits together, or all the shirts together. An exception is the search results (e.g. I searched for “blue”products), but the designers have obviously weighed up the benefits of large, full thumbnail images vs the likelihood of seeing slightly messy search results. Besides, search results by default group the types of products together, and even price-based results will group similar products e.g. suits together.

Moss.co.uk