The first was a photo I took while out for Brunch in Angel for a travel agent with offline advertising claiming to be “No 1 in Google.”

Not too relevant to ecommerce but it generated the most amount of comments the Pod1 blog has seen, and I was especially surprised when it was picked up by a daily digest blog and the guy who created the advert emailed a comment in from the Austrian Alps!

The second was a much more relevant post for ecommerce, about how Analytics and finance data will never match up, and the reasons why. I’ve seen this misunderstanding cost companies a lot of money, but the number of comments …0.

I also had a comment peice on Facebook Places appear in New Media Knowledge, as well as my appointment as an ecommerce strategist in the New Media Age moves. This Clinton-Tarestad is officially no longer an accountant!

Surfing and ecommerce

The world moves not in incremental changes but in leaps and it seems no more so than in eCommerce. The rise of Google from a start up of 1998 to where it is now is a perfect example of this. This rise has given Google it’s current monopoly position, driven by its clean interface and usability to make it synonymous with the word ’search’. With the users have come merchants, gold panning for clicks, building their sites around SEO, investing heavily in Adwords to drive the elusive consumer to them.

However, there are indications that this might not continue indefinately and as Google starts to use it’s monopoly to invade areas that might be considered beyond ‘organising the world’s information’ is it still not being evil?

This isn’t about the smugness with which Google corrects your spelling or knows what you are looking for before you do but seemingly minor changes that may in fact be gnawing away at some of the underlying assumptions of the eCommerce world.

Paid over Organic listings

I recently overhead a CEO at a nameless company saying he wanted SEO to put his own PPC agency out of business in two years, under the assumption that a successful SEO campaign could make up for any loss in PPC revenue. That is laughable. Google wouldn’t let that happen. In fact I suspect the other extreme may happen, Google will know what terms are revenue generating terms and then start pushing organic listings further and further down the page.

This is already starting to happen.

1. They are moving ads closer to the organic listings

2. New ad formats are including sub headings, visual elements, maps – all pushing organic listings down the page.

Organic listings

Google, if you ask them outright, does not provide SEO advice, the quote goes something like “make the site easy to use and navigate and Google will rank it well.” Which is kind of ignoring the elephant in the room. Everyone is aware of the SEO industry, each with there own line they believe shouldn’t be crossed in order to avoid Google’s wrath. However with very few culprits being caught and punished many merchants are no doubt tempted to flaunt the rules as short term gains may outweigh any ‘punishment.’ So no one is sure where the grey line is and what they can get away with. Creating gateway pages like BMW did – wrong, but will you get caught? Buying links? Not sure and very unlikely you could ever be caught. Anyone trying to play by the rules is losing out so with big ££ involved you can bet everyone is pushing it as far as they can.

The result is that merchants who play by the ‘rules’ are missing out, as are genuinely pure white hat SEO agencies, and most importantly are the users who are not necessarily being shown the most relevant content.

New Google markets

Google have just launched the Nexus One. Bet alot of mobile phone manufacturers won’t enjoy paying their adwords bill this month. Same goes for the price comparison and aggregators with the increasing visibility of Google product search. What about analytics companies, cartographers, authors, news agencies? Google is happy to go open source with everything but it’s search algorithm. The profit it makes from its closed source venture allows it to be open source in new markets. As an end user it is great (I’m practically addicted to GA), but let’s not pretend it is altruism.

For example, Chrome is a great browser (my personal choice) but the ability to search directly from the url has the unwitting benefit of turning Google more and more into a navigation rather than a search tool. I predict direct traffic will drop as Chrome market share increases, with one more step before they get to you the chances of a customer being sidetracked to a competitor or a voucher code site increase.

Google dream

One industry to realise the Google dream might not be all it is cracked up to be is publishing. After doing everything required to get traffic from Google they have realised clicks aren’t the same as customers. A lot of people may be reading their news but their product is now seen within the freeconomic model. Rupert Murdoch’s wishes to move away from Google are understandable (if not fundamentally flawed.)

In favour of Google

Despite any appearances in the above article to the contary I am a massive fan of Google; the company, the technology, the people (and especially Google’s breakfast buffet.) This isn’t about reducing advertising with Google or ignoring the value of your organic search channel, the opposite is probably true and the winners will be the ones who recognise these changes as an opportunity rather than a pure threat.

But just like people are now beginning to see a bank isn’t a social enterprising piggy bank with nice free services, people need to recognise that Google is a large corporate enterprise with shareholders like any other. They don’t owe you anything and what you do get certainly isn’t free (even if it looks like it.)

Everything around doing ecommerce is therefore also in its infancy, the methods, the marketing, the merchandising and, of big interest to me, the KPIs. It sure looks complicated and advanced with all the data and tools out there but I’m confident we will look back in even just 10 years and be astonished by how far we had to go. But that’s what makes it a fun place to work….

I’ve already mentioned how conversion on it’s own isn’t a great KPI due to maths trickery. Now I’ve got my eyes set on availability. Michael Ross has already pointed out that online you have unlimited shelf space so the use of availability to see how efficiently you are using your shop floor space isn’t so useful.

And so a much better KPI, one that addresses the underlying value of availability (i.e Lost Demand) is “Availability by Pageview.” I.e. Michael suggests weighting availability by pageview. This makes sense as if you have 90% availability but no-one is looking at those pages you will be losing a lot of sales. Availability on its own really has very little use.

Implementing that KPI is another challenge, especially using free technology such as Google Analytics but below is as far as I’ve got using the custom variables functionality.

1. Create two page level custom variables

- The first should identify those pages which are product pages. So we have product / category / etc

- The second should identify whether a product is in stock or out of stock.

2. Create a customer report:

- Metric is pageview

- First dimension is the key value of your first custom variable (product / category / etc)

- Drill down dimension is then custom variable value of your second custom variable (In Stock / Out of Stock)

- Final drill down to page title.

This should show you which products have been viewed the most that are out of stock. This is not the ultimate goal of availability by pageview but, like I say, were still at early ecommerce days. I suspect this would be possible in other (paid) analytics software but not in GA as custom variables are on visits not on content.

This can be automatically produced in a weekly report to merchandising to prioritise replenishment or change a previous decision to discontinue a line.

There are obviously challenges in faster moving goods such as fashion or if the product is like our most viewed out of stock product the Pad-Dock iPhone to Tablet Converter – IWOOT

If Google don’t send you relevant traffic then that keyword isn’t working for you and you’ll change how much you are willing to pay for it. The more disagreggated the campaigns are (exact matches) the better results you are likely to see.

However with other online ad serving methods such as content networks and syndicated PPC you are often advertising through several or, more likely, many publishers and therefore the CPCs are aggregated across these publishers.

This aggregation leads to a disconnect between the sales motivation for the merchant and the motivation for the publisher. The merchant wants high quality converting traffic where the publisher wants to send high volume cheap traffic.

From the merchant perspective, the merchant cannot often choose CPC by publisher (although Google’s smart bidding technology on their content network goes some way to solving this) and for the majority this is like having one CPC for all the keywords in an adwords campaign. Some publishers might be sending you great traffic, some will be sending poor traffic.

From the publisher perspective, I know that I have to send a certain quality of traffic in order not to get kicked off the network. Should I send lots of high quality traffic at the most cost to myself or should I send the least quality traffic I can? The model motivates me to send volume. The optimum position for me is the most amount of volume at the minimum cost to me while exceeding network minimums. That is not the same as the best ROI for the merchant.

An analogy to me is the same as the financial crisis. Packaging up good and bad debt and selling it for an average. When merchants, like the banks, realise the bad debt element is worthless will this model also collapse?

The key challenge is interpreting the data and understanding it and there is more than one occasion where the Maths appears to play tricks.

Take a look at the following tables

Table A

Table B

Between Table A and Table B I have increased conversion across all channels, but due to a change in the number of visitors and the channel they are attributed to, my overall conversion rate has dropped.

Cue desperate calls from your CEO about why conversion rate has dropped and the usual conversion specialists pitching their shiny wares.

Just maths trickery but it might catch some people out, especially with the emphasis on conversion rates that there has been in the last year.

What has happened is that my profit has gone up. Increased conversion does not necessarily mean increased revenue. I could have kept the same number of visitors and just assigned them to a different channel to show the same affect.

Why diversification increases upside risk

Lets look at the same table but with some year on year %age changes.

If I increase my 3 key profit drivers (visits, conversion, profit per order) in my email channel by 1% each, I get a greater increase in profit than the decrease in the same profit drivers in the affiliate channel. Seems weird, but it’s just the maths.

It is also of greater benefit to either gain or lose 1% in each of the 3 areas than lose 3% in one area.

Is the maths suggesting you spread your marketing investment out? It appears so but how well does that translate to the real world and is it relevant?

Comments below……

Thanks to @manross for opposing aggregated conversion as a KPI and pointing this one out.

Super cookies, permanent cookies that stay on a user’s machine allow for a user’s journey to be tracked and therefore revenue to be attributed correctly across channel’s or across keywords.

ConversionWorks have already integrated this concept into Google Analytics and seem to be using it to good effect.

Theoretically this could be tied in with offline data about that user to really gain insight in to consumer behaviour. And, for me, this is where potential problems arise. There is a difference between using data on an anonymous basis to correctly attribute revenue recognition and being able to see the specific shopping habits of every single customer.

The use of cookies is being looked into in both the US and Europe, hopefully everyone will see sense and realise that the internet experience as we know it is largely reliant on cookies and getting rid of them would create a much less efficient market. But perhaps something does need to be done to protect anonymity on the internet.

Apart from privacy issues, the data will never be 100% accurate due to clearing of cookies, increasing use of multiple machines i.e. smartphones, work and home PCs. However, this is a potential solution to one of the fundamental problems in internet marketing at the moment and I can’t wait to try it out.

K

At a recent Hitwise round table discussion it was brought up that Voucher codes are making it more like a souq market than an efficicent market. Searching for voucher codes are the equivalent of haggling with the market trader for the best price you can. Even more than that, if you find a better code later it is as if you’ve returned to your hotel room to find someone else got a better deal.

I don’t imagine anyone in online retail wants to see it go down the route of the entertainment industry where the face value of tickets means nothing, just hunt around for the best deal and never pay more than half price. This disguising of prices only aids the middle man, devalues the product and reduces customer satisfaction. Brits, especially, feel hard done by if they have paid 1p more than the next person. Not to mention that this makes inefficient markets (many codes don’t work, some are public, some aren’t) and therefore resource that would be more productively employed elsewhere isn’t.

I favour a more open and honest approach from retailers. Online customers aren’t stupid and if your prices are cheaper than your competitors are including a discount then I think they are likely to shop with you. This should be achievable as you are not paying 3 times for a discount (customer, affiliate, affiliate network.) Any revenue lost from discount sites is likely to be made up from price comparison sites. I’m not ruling out a place for voucher code sites and offer sites but they should be short term and product targeted.

Having said that – retailers have created a monster, voucher code searches have only increased YoY and consumers are expecting a voucher code. Is it too late to turn back?……

The flip side of that is that until we do have a method of correctly assigning revenue across all marketing channels (offline and online) than those channels that appear to have a clear positive ROI are likely to get more investment than those without. This explains how Google can be so successful as both SEO and PPC both seem to have clear ROI.

One channel that may struggle amongst this is affiliates and that is because of the apparent lack of transparency. It’s hard to tie up affiliate data with Google analytics or PPC agency data. Have you ever tried de-duping the sales you are attributing to PPC and the sales you are attributing to affiliates? I suspect, like I Want One of Those, you’ll be in for some interesting results.

I ended up posting the below on an affiliate forum as on open letter to see how one guy at one merchant was viewing the situation:

Dear Affiliates,

I wanted to write a response to some of the posts I have been seeing around the affiliate forums. Perhaps a view from outside the usual affiliate arena (finance) might be useful.

Removal of voucher codes
It is clear that the use of voucher codes is only an increasing phenomenon and the number of new affiliates entering this area is only increasing. After careful thought we decided it was not beneficial to continue working with them for the following reasons:

SEO – voucher sites were ranking for brand terms and therefore taking our brand traffic and selling it back to us. With the increased use of Google as a navigation tool this was only ever going to increase.

Cannibalisation – a look at time spent on our checkout page made it apparent that people were seeing the voucher box and then, understandably, going off to find a voucher code. Again this does not provide any incremental revenue.

Brand – to continually have voucher codes in the market gives the impression that the brand on its own does not provide value.

Virality – any codes we did want to release, say to high performing segments, were soon on every discount site.

Middle man – we were paying commission on passing on discounts. We felt this wasn’t necessary with our own DM channels and cashback sites.

Hurting value add affiliates – affiliates that were working well with us in a true value added partnership were having their cookie trumped at the final stage by a voucher site.

Ultimately, like all affiliate programs it comes down to adding incremental revenue.

Were these sites sending customers that would not have come anyway – we didn’t think so. Were the conversion rates higher, and offsetting the loss of margin – we didn’t think so.

We are not pretending we know it all and we are fortunate enough to be in a position to test this. No doubt people will suggest other options, quickly expiring codes, codes on individual products etc but as a small company we weren’t sure whether it was worth the time spent on this or focussing on other areas. We decided to focus on other areas but may revisit this.

We believed we communicated the removal of voucher codes to all affiliates and apologise if this was not clear.

Sibling sites
I have also seen a couple of affiliates complaining we link to other sites and then they would not be rewarded for these sales. This is an understandable concern. What I would say is that we have never paid commission on low margin products (i.e. experiences) and are not in a position to do so. These represent a tiny proportion of our sales and if you are not pushing the products I really don’t believe these will impact you.

Again this could have potentially been communicated more effectively.

MVC
I generally wouldn’t name and shame but this takes the biscuit and unfortunately reinforces any perception that some affiliates are happy to play more than dirty.

It came to our attention that, despite no longer offering commission to voucher codes, MVC have been forcing click out through syndicated PPC links. They continue to rank highly for our brand terms despite having no relevance to the brand.

Obviously this has been raised with Yahoo, Google and ValueClick and we await whether this grey action has crossed the line. We have of course paused working with search123 and are investigating the scale of this.

Other discount voucher sites continue to rank highly for brand terms due to SEO manipulation and then pass that traffic to competitors. This is perhaps not ethical but as long as Google continue ranking discount voucher sites with no relevance to the search term then I suspect we should expect it to continue.

PPC Brand bidding
Although tempting to name and shame in this case as well I don’t think it would be beneficial. It was however disappointing to note that affiliates were bidding on our brand terms despite clear and long understanding this was not allowed. More disappointing was to find some larger players masking their URLs ( a pretty ineffective tactic which only reinforces an element of sneakiness.)

Our affiliate relationship relies on trust and it feels that some people are letting the rest down and eroding that trust. We must obviously continue to reassess the most effective ROI across all channels and I have concerns that a couple of players in the affiliate marketing space are disregarding all codes of honour and giving the rest a bad name.

Overall we are more than happy to continue to work with affiliates. Our creative team works really hard to produce innovative and effective content and we like seeing both parties benefit from this. We welcome contact by phone or email if you have any ideas or suggestions.

Anyone who provides incremental traffic through new visits or increased conversion rates deserves to be rewarded and we are happy to pay industry best commissions, provide prizes such as our weekly £250 and give you the chance to come on holiday to Thailand with us.

We have a strong brand and are benefiting from increased conversion rates so it really is all to play for, just play fair!

Best regards,

Kieran Clinton-Tarestad
Head of Finance – I Want One Of Those.Com