Advertising networks want to maximise their revenue. The way they do that is to pick the highest CTR, highest paying adverts. This turns out to be quite simple. Forget about pay per click, and instead calculate adverts as if they were CPM. That’s easy to do. If you have an advert that costs £1.00 per click, and has a 10% CTR, then it is showing at a CPM of £100.00 – because a 10% CTR means that you have 10 clicks per hundred impressions or 100 clicks per thousand impressions, and 100 clicks would cost you £100.00. Now rank your advertisers based on that calculation of the value of an advertisement and you have the first simple AdWords auction – it got more complex, but that’s still the basic premise of how it works, but with a “Quality Score” factor instead of just CTR.

However, while that maximises revenue for the advertising system, it can’t guarantee that it has maximised the revenue of the advertiser. The practical example was when we were working on Thomas Cook. If we offered adverts with “Book now…” then we got an increase of 40% more bookings and they mostly happened a week earlier, than if we used the copy “Browse…” or “Search…” instead of “Book now…”. But the CTR was slightly lower on the “Book now…” adverts. Which did Google choose to optimise? And what was the impact?

Yup, Google picked the higher CTR adverts, which de-optimised revenue for our client. So we stopped allowing Google to choose which adverts to optimise. It’s the only safe defence.

So, very early on, we made a policy decision that we don’t allow the advertising networks to choose which adverts they want to show, unless the client purpose didn’t involve a strict ROI target (e.g. a brand awareness advertising exercise). And we now inspect adverts to see which ones have a lower ROI, and lose them – not based on the CTR, but based on sales.

adCenter doesn’t have the option to show adverts equally, on rotation. That means that adCenter will optimise its’ own revenue, at the expense of an advertiser, by removing the advertisers ability to collect a statistically useful sample of results, unless the advertiser goes to daft lengths (e.g. sequential tests rather than A/B or multiple parallel).

If you are interested in optimising for the new adCenter, see if you can register your interest in having Microsoft permit you to have equal rotation, so you can assess which adverts work and which fail. Note that there are other requirements if you do this – if you have multiple keywords per AdGroup, for example, the performance of each keyword needs to be roughly the same, or you may be selecting adverts to display for high volume low conversion rate keywords, when the right driver to select the adverts is a low impression volume high conversion rate keyword. Making web advertising effective is a bit more complex than just equal rotation – the optimising method has to include how many times each advert is showing for each keyword.

This lack of control by the advertiser has been noticed by other adCenter users. I was going to add this comment (or one rather like it) to the adCenter discussion group, until I found out that to do so, I’d have to stop using this computer, and use a Windows machine. I’d rather not, thanks. I have picked this operating system for a lot of reasons – and Microsoft’s tedious obstacles don’t make me more likely to use Windows, it makes me less likely to do so.

I should point out that this observation about optimisation is not unique. In fact there’s quite a lot of other people making this or a very similar point. Here’s some:

• Use Split Testing to Create Killer Ad Copy
• Ad Copy Testing: Systematically Improving CTR – Microsoft’s own blog for advertisers recommends split copy testing
• Is it possible to make ads rotate evenly? – Microsoft adCenter discussion forum with some strong points made by advertisers

This is an issue that Microsoft needs to take seriously, It can have a substantial impact on performance, and the value of the tried and true technique of split testing is significantly reduced when Microsoft chooses which advert to show. It may help Microsoft’s revenue, but not that of advertisers. And offering “just create a new AdGroup because they start without optimisation” – is just not do-able. We tried. Merging the data from multiple attempts so that you capture the period before Microsoft has optimised is a data horrorfest, especially on a large account. And testing like this is yet another management overhead on a smaller fraction of advertising than Google – Microsoft need to reduce the administration overhead, not increase it, relative to AdWords’s cost of operations.

However, the rate of goal completion is only one of the factors involved in optimising a web site. If a source has a 25% goal completion rate, but only brings in four visitors in the measurement period – there may be reasons to suspect that the single conversion might be less important than the half dozen conversions from a lower converting source that sends a lot more visitors. The likely maximum number of visitors from a source is important – so understanding all sources of conversion is essential.

Google Analytics has recently added a “weighting” switch to the Goal Measurement page. I was hoping that this might then weight according to how many conversions were involved. Instead, it does something quite mystifying. The source with the largest count of conversions is listed *last* in the sort order, below dozens of sites with no conversions at all. I can’t imagine why a positive count of conversions sorts after sites with zero conversions.

Here’s an example of the same data, without and with the weighting.

Why the largest volume source of conversions is weighted so that it doesn’t show up at all… I have no idea:

Notice that with the weighting enabled, the 5,000 visitor source just vanishes from the listing altogether. Useful though the idea of weighting the converting sources is, if you have a lot of sources, watch out – you might inadvertently lose track of some valuable sources if you simply enable this switch without paying attention.

I poked around for a while and found a few examples that did part of what I was trying to achieve – but nothing that converted a UTF-16LE TSV into a UTF-8 CSV (strict). I believe, on the basis that it works on the programs that I’ve been having difficulty with, that the following program stub example (fixed input and output file names) is capable of being tweaked to do the right thing for you.

#!/usr/bin/env ruby -KU

outfile.flush
outfile.close


I hope you find this useful. If you’ve suggestions to improve the core function, please tell me. This is used inside a project that I have – so I’m not fussed about making this a fully competent command line program. I’m most concerned about the horrendously ugly and clumsy “split(2,100000000)”.

I logged into our banks internet banking facility. I wanted to make a payment on Friday, 5th February 2010. I went through the identification of the recipient, and the payment and changed the offered date from Monday 8th, to Friday 5th. It refused the payment request, saying that I must choose a date between Monday the 8th and some date in June.

So I picked the 6th (Saturday). Still no go. Then the 7th (Sunday). Still wouldn’t accept it. Finally, I selected the 8th February, a Monday. Bingo! Payment will be processed.

It then processes the payment and tells me that the payment will be taken from the account on Friday the 5th February and deposited in the target account later that day. Which was exactly what I wanted in the first place, and it kept telling me it wouldn’t do.

Imagine my annoyance at having, for ten to fifteen minutes been told I couldn’t do what I wanted, and then being told that despite offering a different date for the payment, it would be paid earlier than I’d actually requested? What if I *had* wanted to pay on the Monday, 8th Feb? Then the funds would have been out of my account on the 5th, and not the 8th, as requested.

I’m disturbed by my Banks inability to work out when they are prepared to act, and the misrepresentation of the dates of actual activity, versus requested. It makes me less confident that they’ll do as I request. and I’m now more likely to pick up the phone or drop into the Bank. Or to change banks.

Purchase Processes – Especially For Special Offers

Our client had a similar issue – there was a particular offer they made, which couldn’t be fulfilled through the order interface, until after another action had been completed. But this wasn’t expressed clearly to users. They wanted the offer, signed up for what they were told to buy, and then couldn’t see the mechanism to access the offer – until they’d completed another stage in the the checkout. There were technical reasons why this order of actions was important – but the user could have been given a text notice that this is what the system would do. It’s no good when only *your* staff know how the system works – it is the customer that needs to be told, and told clearly, or conversion rates online suffer and the customer service phone lines heat up.

The only diagnostic that this company identified was an increased dropout in the funnel and decrease in conversion rates – when they were expecting total sales to increase.

This is a classic example of “you have to understand your business the way that a customer sees it, not how you know it works”.

This article is part of a series of short meditations on user interaction and the extension into social marketing, following on from a look at reducing the barriers to interaction, and some of the pros and cons of allowing user generated content, in “SEO: Remember Relevance?“

An Analogy

Recently, restaurant revenues have been generally in decline. Is this because people are finding restaurants misdirect them to food?

Or is it because of another, perhaps slightly more obvious, factor?

What about people optimising their budgets?

Is there any reason to think that businesses do not also look to their operations in a recessionary environment? Would the shareholders be out in lynching gangs if a business was to ignore a recession?

Why Is Advertising Really Decreasing?

It’s a recession.

That’s it. When the marketing budget goes down, what do you do? You focus money on the most effective items.

Paid Search typically is icing on the cake. Not always – it depends on the business and I can cite companies where this isn’t true, too – but we’re looking at industry wide factors. IME, Paid Search is most often used to add a chunk of business for an established site. They could drop the advertising, and still see leads and purchases. So when they are keen to save money, it is an easy line item to drop.

In a recession, users spend more time looking for bargains, and click more. Conversion rates typically decrease, and ROI decreases. What is the right response? Decrease spend, and improve the site.

Instead of spending more on advertising, companies will direct effort to Search Engine Optimisation. SEO is an investment. Over months, you keep working on improving relevance, increasing user satisfaction, helping users to link to your business, developing low cost business relationships, and perhaps using grey and blackhat techniques – depending on the type of business you run and the risk you will take.

You also invest in site improvements – for a large company, that is cheaper than paid search. By optimising the site you improve user satisfaction, you can improve search engine rank, and you can decrease the need for paid search.

Strategic Impact

Two deceptively simple-seeming numbers tell a strategic truth.

In January 2008, about 90% of our leads for new business were primarily people wanting AdWords management.

In January 2009, about 80% of our leads for new business are primarily people wanting SEO and Conversion Improvement assistance.

Most of our AdWords clients have reduced spend. A few have increased their spend. But nearly everybody looking for extra business is asking us to find more economical ways to win business. That means looking at other channels, such as FaceBook and Twitter. It means exploring the use of blogs, and discussion forums. It means asking for feedback on the site content. It means looking at what users do and don’t do, and working out how to provide them with the best answer. It means improving site navigation and usability. It means offering up to date content.

Misdirection and Satisficing

The *worst* part of the article was a complete misunderstanding of how search engines, especially Google, work.

Google makes money because it doesn’t focus on making money. It focuses on satisfying users. Because Google has a large number of users, it is more appealing to advertisers. If Google presented search engine results that *didn’t* satisfy users, then users would seek other search engines that offered a better result. Advertisers would seek other opportunities. That has a direct impact on revenue. How?

The more advertisers there are in an auction, the more Google makes in that auction. If just one company bids, and users like the advert in large quantities, the advertiser can pay $0.01 per click. If there are two advertisers and users think both of the adverts are pretty good, then the average cost per click is higher – how much higher depends on the bidding strategy of the two companies.

However, if only one company bids on a keyword, and search users hate the advert – either the CTR is low or the advert is misleading – then Google will disable the advert. With a sufficiently unpopular advert you can bid $100 and your advert won’t be shown, even if no other advertiser is advertising on that keyword.

This is called “satisficing“. Google has created an auction mechanism that allows businesses to compete with each other. It has also created a system that allows users to vote on the acceptability of adverts. Unacceptable adverts – things that users hate – will eventually disappear even if the business is prepared to pay outrageous amounts.

There is no misdirection in that system. None of the parties may be completely happy, but none are unwilling to be a part of the game. That’s satisficing in a nutshell.

Misdirection Can’t Satisfice

So a central plank of the argument, that advertisers misdirect, simply isn’t borne up. If you have any real experience of paid search, you’ll know that you can decrease your Average CPC on Google by delivering improved adverts, better landing pages, easier purchase paths, links to supporting information, better customer service – *NOT* just by bidding more. Bidding more, in general, reduces ROI. Bidding management as a sole strategy is a failed idea. It was true in 2002. It is obsolete as a strategy now, and has been obsoleted for years.

Ignore The Rest

If someone pontificates at such length about something they don’t understand, I can’t read the rest with any sense of belief.

I *do* know something about PPC. I’m an AdWords Help Forum Top Contributor – I’ve read *thousands* of advertisers questions about why their adverts don’t work. I have free tutorials on how to improve adverts – which people say work for them; all the tutorials focus on increasing search user satisfaction. I’ve spent, on behalf of paying clients, budgets of up to $500,000 per calendar month – each advert leading to a specific landing page that answers the search query, with up to 4000 unique adverts flying at the same time, to bring users to the highest satisfaction landing page.

I’ve advertised, on behalf of clients, on competitor names and seen low CTR and low conversion rates and reducing Quality Score. I *know* that Google pays attention in organic and paid search results, to user satisfaction.

I’ve seen too many examples of clients in distress, asking for help, where they haven’t understood that they can’t just pay to appear.

Summary

Paid Search is declining because you get one sale from about 1% of clicks; it costs for each of the hundred clicks (or whatever the rate is – we see between 1:5 and 1:500 for various clients). In a recession, you can save costs by reducing items that are per-sale. SEO is not a per sale factor, it is closer to a capital investment; you can weather an economic crisis by living off the organic search results or investing further during the downturn so that as the economy recovers, you are better positioned for increasing user interest – but you aren’t paying more while users shop around.

Almost every single client that we have is working on improving their site, and improving their SEO results as a consequence. A year ago, they’d have spent money to bring in leads, and spent less time improving the site. That’s why PPC spend is down. Not because users are misdirected, but because advertisers are actually focusing on improving performance in a downturn. It is often cheaper to address user needs than to advertise and an improved site is improved for some time – whereas a click is transient. Pay Per Click appeals less than Conversion Improvement and SEO. That’s it. That’s the story.

Pretty flaming obvious, IMO. Oh, gee, I wrote about 2009 strategic advertising responses months ago. In a prediction. Hoop de doo.

Update

Danny Sullivan versus Eric Clemons in TechCrunch.

1 – Ignore Other Opportunities

You can appear on Google Map listings. In Image search. In Video search. On YouTube. Blog searches. Mobile search. News sites. News aggregation sites. Bookmarking sites. Press Releases. Social networking (FaceBook, My Space, Twitter, LinkedIn, etc all have their markets and uses). Human moderated directories. Professional and trade association directories and many, many other places.

(We come across companies with an obsession to show in precisely one place – usually Google organic search results. There’s a risk to that. If Google change their ranking algorithms, all the traffic can disappear; spread your risks by appearing in all the places you *should* be. They don’t usually have the volume of traffic you can find on Google, but some have high quality visitors, even if of low volume, and some have really high volumes of variable quality – if you don’t consciously consider these, your competitors will have no competition. )

2 – Redirect Everything, All The Time

Sick of those boring old URLs that everyone else uses? Evade the search engine results pages by changing your mind about what the page is called. Why stick with just serving the page you were asked for – deliver an excitingly renamed page on every request!

Use a custom Web Content Management System with a unique tracking code in every page request! Or, use Microsoft’s incredible “This’ll Fool The Spiders” Cookieless Mode. Then, when a spider comes to visit, the pages that ordinary users see will be renamed dynamically. Now you can decorate your URLs with “?AspxAutoDetectCookieSupport=1″ *AND*, in a 2 for 1 Daft Deal, insert unique strings like “(X(S1)(hjkgjhgw2823tku))” into all paths recorded by search engines. That’ll split the weight between multiple competing pages and drop your rank. Super move!

For an extra trick, change the page content after each crawl, so that spiders will take years to recrawl frequently enough to work out that the page is actually the same, whatever the wierd bit in the middle, or at the end, does.

(This seems to be a misguided attempt to help people who don’t allow cookies or to provide some other kind of session interface. However, the public part of the site doesn’t critically need cookies and tracking – so for the sake of a tiny handful of people who want stateful operation but don’t want to allow cookies, they stopped spiders from crawling efficiently, and spread pagerank to thousands of “different” pages. The new “canonical” link tag should help reduce this perverse usage, eventually.)

3 – Only Idiots Need Descriptions

Your site is famous for what you do. You don’t need no stinking meta description. Besides, people say that the meta stuff isn’t important any more. Dump those weaselly bits that say ‘name=”description” content=”some boring old rubbish”‘. It’ll make the page smaller too, so it loads faster. That *has* to be good!

Alternatively, forget boring old HTML and build a Flash Site. Sure, you need a stinky piece of HTML to *load* the Flash, but you don’t really have to do anything with it, like Titles or Descriptions. That’s *old* stuff.

(We still come across far too many sites that buzz with graphical excitement, and are either unfindable, or if you stumble on to them in search, have a dull as dishwater title and no description. It’d take 400 characters to put up a meaningful title and description that would help users to find and recognise the site as useful. And not a lot more to make Flash content into a decoration for users that want to engage with it, while the real site is delivered in search engine friendly AJAX.)

4 – Optimise *All* Pages for the Same Keyword

Well, if getting *one* page in the search engine results is great, think how much better it would be if every page in your site was showing all over the search engine results pages? Make sure that you use all the keyword optimisation tricks you can think of, to make sure every page ranks for the same keyword, including your contact page, privacy page and terms and conditions page. Everyone wants to see snippets from your privacy policy, anyway, no matter how irrelevant to the keyword.

(If you don’t optimise each page for different keywords, you confuse the search engines about which page to rank – reducing the likelihood of any page appearing at all. Don’t repeat titles and descriptions, either. Don’t link to other pages with the text “click here”, either – a nice descriptive link text helps you and users.)

5 – Consistency Is The Hobgoblin Of Small Minds

For ease of development, and to give that creative freedom, just let anybody name any page anything! So if you sell hot water bottles, we could have pages called:

• hotwaterbottles
• hotwaterbottle
• hot_water_bottle
• hot-water-botttles
• HotWaterBottles
• Bottles/hot-water

Each page makes a slightly different claim about what you do. Each is linked to a few times. So there’s no particular page that has all the weight for your product. They are all good pages for your product. They just don’t rank very highly. But that’s OK, because they are creative.

(Creativity is great. But a creative site you can’t find and reliably navigate, is… a work of art? A lost opportunity? An investment without a return? Your choice of what you call it – but “high ranking” and “popular” shouldn’t be in your list of descriptions.)

6 – What Call To Action?

Write effusively about your company. But don’t let anyone know what the products are or what they have to do to get it. Avoid phone numbers. Don’t offer contact forms. No shopping baskets and storefronts. No email addresses. It’s a web site. They sell themselves, automatically…

(We’ve come across sites that manage to completely conceal what the product is, and have no mention of any contact information, whatsoever. Companies that say they deliver results… no idea about even what sector they work in, and no idea on the page what to do to proceed further. Tell users what you’ve got and what you want them to do; they’ll do it if they want what you’ve got.)

7 – Hide The USP

When someone buys from you, they pay slightly over the standard price… But unlike your competition, you include free delivery and on-site service for three years. Wouldn’t it be a good idea to *mention* that? Somewhere? Or are you really just going to let visitors guess that’s what you might be offering?

Start off by assuming that you’ll only be seeing your regular customers and your staff on your web site, and you could be right.

(We’ve come across sites that have great reasons to pay a little more than the lowest market price – but who seem embarrassed to admit they have a reason why their product costs more – whether that is functionality, quantity or quality. If you have a Unique Selling Point – tell the visitor what it is. Clearly. )

8 – Abandon Shop, Abandon Shop!

Your visitors have added items to the shopping basket. They click on the purchase button. They aren’t signed in!

This would be a great moment to empty the basket for them. If they are a new customer, they probably need a nice new basket, not that old, used one with stuff in it. Do them a favour and just clear it out. Now they can order with confidence …. wait… where did they go?

(Focusing on parts of the process can destroy the whole point. If you create a great login system, that’s interesting. But lose the basket, and you’ve lost, no matter how good the login system or the basket are. Even better – don’t make users remember whether they were registered before and force a password retrieval process – if they aren’t on your site every day, to buy something, they *will* forget registration details – so focus on collecting shipping details and match those to an ID, if you have to, as late in the process as you can. It is more important to get a new, satisfied customer, than the registration details of someone you’ve annoyed, who won’t buy. )

9 – Never Diminishing Circles

Special! Discount For New Customers!

In order to purchase this item, you must be signed in. Register Now!

Item Discount Only Applies To New Customers. Please select another item.

Purge cookies.

Special! Discount For New Customers!

Rinse and repeat.

(We’ve come across many variations of this. Special offers you can’t reach, technological obstacles to users completing a purchase. Test your site; monitor for small creeping changes having a huge impact. It is tedious to pretend you are a real user, and try to look at the screen as they do, instead of just *knowing* that at this screen you ignore the big graphic link that says “Buy Me” and must click the small text link in the footer, or whatever the trick is. If you want customers, you have to hit the problems and solve them, before would-be customers hit them and become dead basket cases. )

10 – Unique Navigation System

So many stores have boring links to look at, with categories of product and links to related products, etc. Let’s make an exciting visual storefront that exemplifies our brand values! We can use special parts of the screen, with unexpectedly tiny and ambiguous icons in pale grey on paler grey to act as store navigation! That’ll be exciting! New! Different! We’ll stand out from the crowd with our uniquely low conversion rate!

(Forcing users to understand new navigation systems can work – if the site is a game or a puzzle and that’s the point of the site; it might work for something like the the Wii, or PS3 sites, but mostly fails everywhere else. Non-standard navigation prevents users from understanding what they have to do to get to the right page and the right activity. Why would you *want* to turn away business for a shop, by disguising the front door as another display window or as a brick wall with a “closed” sign up and the lights off – the visual cues are important user interface design features, not a fashion statement. Usually. We did some work for a company who insisted that would be users solve a puzzle before they could register – and the registration rate declined dramatically; buying users are usually trying to solve their problems, not yours. )

11 – Make Your Business Problems A Challenge For Your Users

So you have a lot of products… Organize them on the web site, the way that your business divisions are organized! Users will naturally understand that, for example, *this* printer is managed by the “Small Office/Home Office” group and *this* very similar printer is managed by the “Small Business” team. Or, organize the user interaction by department. Have marketing offer products over here… and the basket is filled using these links over in the Sales part – and the order is submitted directly to accounts, who have their part on the site. But because each part of the site can change independently of the others, no cross linking between divisions of the company! If users want the products enough, they’ll learn what to do.

(OK, there is some value in exclusive brands, because they keep people out and that can let you drive up the price… but is displaying your lack of ability to organise a functionally effective customer oriented interface, something you really want to tell customers? Customers shouldn’t be aware, from your web interface, how complex a problem it is to manage their order – that should be your problem, not their problem. We’re perpetually astonished that so many businesses reflect their internal issues to customers, and make it harder for customers to buy, not easier. )

12 – Server Response Codes

Stand out from the crowd in search engine results! Why bother with boring messages relevant to your users, when you can show a fashionable and stylish “Planned Server Downtime” or an equally gorgeous “The Page You Were Looking For Cannot Be Found” search engine results title? Hack your web server so *all* pages are “200″ (Page OK). Then, when a search engine spider visits, it can’t tell whether the page is what visitors want, or a temporary out-of-service page.

This is one drastic technique to reduce the number of 404′s in your site reports. Get to the top of your managers’ “outstanding staff reports” by eliminating all 404′s, 301′s, 302′s and 500 series status codes in web server reports. Now you can have 100% good page delivery! Even if the page is rubbish.

(We have no real idea why the guy that did this, did it. We think he wanted to reduce 404′s, but he may just have been having a bad brain day.)

Summary

There’s lots of way to kill your business on the web. Many more than we’ve listed. How many are you using?

A recessionary environment is a good time to take a hard look at how effectively your web site is working for you. Hopefully we’ve tripped a few alarms and you can review whether you are making the most of the web marketing opportunities you have.

Weaknesses in the paper

There are a few embedded misperceptions in the “AdFraud Anecdotes” PDF.

I’ll probably add to this list as I think more about it.

• Advertisers do not react rationally. That’s partially a consequence of information holding asymmetry.
• Google controls Broad Match and Auction Quorum Sizes to improve Google’s returns – and hides the crap search queries in reports as “18,0000 other unique searches”
• Google is in complete control of the quality of matching, by default. Google controls click quality, advertisers choose bids and can forgo volume by selecting phrase match and exact match, or accept Google’s decisions about matching. Most of the variation in CPA & ROI that I find can be directly attributed to Google changing the nature of search queries that are being matched.
• Advertisers are remarkably reluctant to measure – because for a small business the costs of developing the understanding of measurement is expensive; so is hiring in the talent to understand the data. I infer that small businesses have a higher percentage of undetected and undetectable click fraud and greater difficulty in establishing a “useful for management purposes” CPA (internet noise makes data collection long period – I may write up the details of this).
• The paper addresses large advertisers quite well. Typical of economists and governments – but single large entities are *not* representative of the larger numbers of smaller businesses – scale changes impact.

Strengths of the paper

This shows that some people within Google are clearly understanding the role that Google has.

It is interesting that the thinking and insights exposed in this paper *haven’t* percolated to the AdWords Sales Teams, at least in the UK. I’ve recently had a rather unsatisfactory meeting with a major account team, who said that Google has no insight into conversion… Despite our mutual client having AdWords Conversion Tracking enabled for several years. If ROI was important to Google, you’d have thought they’d have noticed, and been rather more interested in how recent change have affected ROI. Wouldn’t you? Instead, I got the usual lecture about increasing spend… instead of them giving me techniques to manage AdWords to improve ROI in the face of Google’s efforts to undermine that. I may write about that more, when I can work out how to disentangle any explanation from any specific clients’ data.

The PDF provides some rather Delphic oracular premonitions of the recent user interface changes, for example, separating metrics between Google Search and the Search Partners and the Content Network; allowing (some) control over domain parks, 404 page advertising, etc. Careful reading is required and may be rewarded.

I may make some predictions about likely further AdWords UI changes, when I’ve had a bit more of a think… Frankly, stumbling on this has been a bit of a shock…

I baulk at throwing out stuff that is perfectly good, if the right use can be found. I tend to pass on old computers to charities, or, as in this case, look for ways to re-use or recycle material. Obviously, I used search – on Google, Yahoo and Live – to look for “charity spectacles glasses” and “charity glasses re-use”. I was immediately intrigued by the different approaches of two sites, and how effectively content match was working.

One site has clear actions to take, and links to the relevant organisations. The AdSense adverts are pretty relevant – though I have an additional observation to make about this, below. Have a look at this screen capture from the UK Reduce, Re-Use, Recycle page on “Where Can I Recycle Glasses“:

Check out those adverts! Vision Express is named in the article, and there’s a matching advert. The link in the article is helpful for searchers, too. I used the link rather than the advert – but the advert helped me with name recognition and awareness, part of the role of content match. I am now more favourably disposed to the chain.

“Maggs Optics” is probably poorly geotargeted. I have absolutely no intention, from here, of going to Wales. It is quicker and easier for me to go to London or Birmingham (which I also wouldn’t do) than to go to West Wales. Geotargeting typically improves CTR and reduces wasted clicks; very useful with geographically restricted services. I’m not going to waste a brain cell on remembering the name… see, I’ve forgotten it already…

“D&A” – not bad, a money saving discount offered against an article on recycling glasses – but do I want a money saving offer just *after* I’ve just bought glasses? IMO, you would be more likely to look for this page, if you had just bought some glasses and were cleaning out old pairs, or if you were doing something unrelated to purchasing. I suspect that this advert was tripped on “glasses” or “spectacles” and “reduce” – as in “price reduction”. So it may have been well targeted, but accidentally stumbled on a “recycling” word. OTOH, flagging your product savings just after a purchase, especially when it is a frequent and likely repeated need, might be regarded as pretty good targeting for awareness. :)

Yell’s “Find Local Opticians” – spot on message, especially with the advice in the article that some local opticians may also be taking part.

On The Other Hand

There is a temptation, when writing text for SEO purposes, to write from what you know, and do no or little research. It is cheaper, without doubt. There’s also a tendency to avoid off-site links, in the fear of either losing position or of sending people off site before they click on a revenue generating mechanism. This next page looks like one of those “write without research, and avoid outbound links” types of SEO effort:

Although there’s a lot more text, there’s pretty much nothing to do. No links to look up anything. Nothing that connects this site to the rest of the world. Although there were lots of suggestions like retaining my glasses to put on a snowman, there was no practical advice about the businesses that actually take part in reuse or recycling. As a consumer, my personal feeling is that it is a weaker page – I’d need to do more searches or click on other results, to accomplish what I want… or more likely, since this is a “low commitment” intent, where I don’t have strong personal needs to do something, just give up.

And look at those adverts. You’d have thought that if this page was strongly SEOed that the multiple appearance of keywords – evident in the writing – would provoke some recognition by Google’s content network that this page had nothing to do with IT… or Laser radial keratotomy… or magnifying glasses. It’s mildly surprising – how did these adverts end up on this page?

My guess is that for the first advert, “disposal” was triggered by synonym with “landfill”, and the word “it” appears on the page a few times – case isn’t important and “IT” and “it” are the same, though probably a stop word for content match. I’d expect a low CTR and low awareness – because the advert is so far off the intent. Only if you were on an unrelated project involved in IT disposal, would you expect to consider, much less click, on that advert.

“Eye correction lenses” probably got there from mention of lenses in the content. And “Magnify Reading Glasses” probably just tripped by “glasses”.

Google does pick up on content – but it doesn’t always grab what you are really trying for. These adverts certainly aren’t as well selected as for the first site. Brevity helps. Google content match appears to be less confused by a tersely worded, on topic page.

Sending People Off Site

I’ve had several discussions with various web site owners and managers over the years about sending traffic off-site. I’ve even been involved in some experiments, testing whether off-site links really damage conversions. Pretty much invariably, you end up with more visitors when you offer useful links, than the people you lose to following off-site links.

For example, one site targeting users of a specific language, non-UK citizens in London, is more popular as a result of providing useful links for that community. The site is referenced within the community as a resource, and gains links inbound as a direct result of offering outbound links. How much is the specialist services directory used? Much less than the main purpose of the site.

If you run your own blog, and offer something like FeedBurner, or use Google Analytics to analyse inbound and outbound traffic, you’ll probably observe the same. An article scattered with outbound links may see less than 1% of users clicking – so long as the article is moderately interesting. If it isn’t, you’ll see a high bounce rate and low dwell time on the page – users will make up their minds to go back to search or the referrer, rather than click on an outbound link from a site they consider unhelpful.

Charitable Adverts

How much more appropriate and awareness-reinforcing if these advert blocks had carried advertising for Vision Aid Overseas?

The problem, of course, is that these adverts would cost the charity some cash. Or would they?

Googles corporate responsibility program offers a charity grant process for specified purposes – such as global health. I can’t be sure that Vision Aid Overseas would benefit if they applied, but I’d prefer to see an advert for VAO on both of these sites, than one for “IT Disposal”.

Summary

I felt that these two pages characterised two extremes of SEO:

• Researched, brief with specific actions and offering links to related activity
• Unresearched, lengthy, with non-specific elaboration and no links to go further

Don’t fear outbound links, if it helps the user. They’ll remember your help favourably and may refer you to their friends.

If you run a charity, check out the Google Grants programme here in the UK. If you are a charity in the UK, and need help – we’re particularly interested in child and teenage education and music education for all age groups – drop me an email and I’ll see if we can schedule some free effort to help you get some Google Grant advertising going.

Well done Google for that Charity Grant. Can’t say that I’ve seen adverts for it, or as a result of it, though. Is it effectively promoted? :)

For maximum impact, you’ll want purpose-designed landing pages to optimise conversion. That usually means running some experiments. You should be conducting those experiments before the peak selling season starts. For example, coming up to Christmas, your visitors may become more anxious than usual about delivery schedules and packaging – especially if you can handle seasonal gift wrapping and personally imprinted cards. Delivery costs can sometimes decrease in significance, and receipt before Christmas can dominate – meaning that normal factors for your standard pages for the rest of the year are turned topsy-turvey.

We see many sites with poorly configured web analytics, unable to track conversions and hence unable to tell whether changes are having any effect – time to get that sorted out. Site designers will be looking at scheduling Christmas activities, so your opportunities to fix up your site, if you rely on outside help, may be limited if you wait much longer. Some site designers for SME’s can’t, IME, wrap their heads around the difference between the analytics code that goes on every page and post-purchase order conversion code… Working out what your real conversion rate is when someone just stuck conversion code on every page of the site, can make life rather complex – think “300% conversion rates – up from 1%, this week only”. If the designers can’t get back to your site to fix it for a few days or even weeks, it really messes with the stats. Note that these are not invented scare stories, but examples of practical problems that I’ve experienced in the last year.

IMO, you should now be working on extracting key management data from your web analytics, in order that testing new landing pages for what is, for some of you, peak selling season, is as effective as possible.

Oh, and one more piece of advice… Don’t rework the store coming up to your peak season, especially if you rely on SEO. I’ve been called in for more rescues this year, than ever before, from store owners who have accidentally killed their own sites in the run up to Christmas with a steroid-pumped video game full-Flash site – and no search engine presence at all, as any keyword loaded links now get a 404. Switching site technologies and making major information architecture changes needs some serious planning before the switch, especially if you get any volume of traffic from organic search.

IME, there is a great urge to switch the site just before the peak of activity. If the new site turns out to be less effective, you’ve just shot yourself in the foot – and new sites can be less effective, easily. Why does a new site have decreased conversion? Because it usually forgets customer needs that were baked-in to the old site, emphasises graphical branding over the ability to work out what the offer is, fails to reduce purchase fears or to make the next steps obvious. Note that good graphical design can enhance sales – but the graphics designer has to be suitably briefed and the CEO mustn’t get overexcited by a heading graphic featuring something irrelevant to customer needs, like video walkthrough of the new extended warehouse and good receiving area, that consumes two thirds of the page on a 1280×1028 monitor (but looks quite small on the agency’s 1920×1200 screen). Again, all experience within the last year – though I changed the description of the “missed the point” headline graphic to avoid embarrassing a client ;)

Summary

Make sure your web analytics are capable of measuring conversion rates and sources (down to search queries in paid search) with some accuracy.

Start designing landing pages now, using experience from previous Christmas selling periods – not your normal year round experience. What special offers and discounts can you offer that will trigger a higher CTR than your competitors, and generate higher conversion volumes?

Schedule time for testing responses – looking at bounce rates, time on page, the conversion path, heatmaps, etc.

Avoid switching your site now. Unless you have a dedicated team and a fast cycle time with well organised analytics, you are probably too late to change and fix up a site ready for a Christmas rush, with optimised conversion paths and effective paid search. It took time to make the old site work – it’ll take time to raise the performance of the new site. Honest.

Focus on what customers are trying to achieve, not what you are trying to sell – you can square the circle by making them want what you’ve got, by pointing out how what you have matches their needs, rather than flogging the features. And make it a good reason to buy from you, not a competitor – returns policy, unwanted gift policy, accreditation, endorsements, testimonials, guaranteed on-time delivery, official reseller logos, etc…

Do it now. Even waiting another month will likely lose traction.

I’m expecting a busy September, troubleshooting.

I’ll be thinking about this for the next while… results when I get some!

Speculation…

Hmm. reading between the lines, this is probably for optimising medium to larger accounts (more than tens of clicks per day). Why?

Wild guesses, since I don’t work for Google… I’ll bet that they are using some kind of multivariate real time data analysis technique. Possibly something like Stochastic Perturbation with Simultaneous Annealing (SPSA) or the Taguchi Method. I’ve certainly had problems using these techniques on low volume accounts. When you get down to a few tens of impressions, it takes a long time to establish any statistical basis for action.

Making optimisation more complex, AdWords operates in a number of different regimes. An advert may be shown on Google’s properties before editorial review, and on the Google and Content Network after review. That substantially modfies the performance. Google knows exactly when an advert is reviewed… But advertisers have to guess based on seeing impression volume changes and clicks from non-Google properties. That means you really want to tie either a third party redirector, a JS Web Beacon/Page Bug to for landing pages, or hook into the web server log files, just to infer when the review has been completed… If you don’t, you get skewed results.

Small bid changes can also affect the regime – so bids can affect not only position, but also whicj searches show the advert. And other keywords interact, too. Drop the bid on one broad matched keyword, and you may find hat impressions ramp up on another broad matched keyword, for the same set of search queries.

I can imagine that Google *could* do a better job of optimising, using data that they uniquely hold, than any kind of external analysis. I expect that their main method of optimisation will be to optimise their inventory rather than your ROI, though. Yeah. I’m sceptical. Still.