Tuesday, September 25, 2007

What's your point of reference?

Whenever you analyse results from a marketing campaign, you need to frame them in relative terms - that's the great thing about ROI metrics - they are relative - i.e. what did you get out vs. what you put in.

Very often, analysts make simple mistakes in their calculations and miss simple errors. Because they forget to use a term of reference before they report, they send our garbage.

I learnt this lesson early on when my then boss, Jeffrey Merrihue (http://www.lostartofmarketing.com/) suggested some numbers i'd given him were likely to be b****ks. He took out a ruler and showed me just how wrong my pages of spreadsheet calculation must have been. Because i'd been up for 36 hrs straight (it was a bad week in Madrid) and had become lost in the detail, i'd lost my point of reference.

The biggest mistake analysts can make is not seeing the woods for the trees - is the result sensible and understandable? If not, take a step back and reframe the question. If you're not sure, check your steps!

One more thing - don't always trust your calculator. Excel 97 had issues running regression calculations using the analysis toolpack and now it appears Excel 2007 can't do a simple multiplication - try asking it to calculate 850*77.1. The answer shouldn't be 100000 should it?

Friday, September 21, 2007

Will Procurement eat marketing's lunch?

On Tuesday this week, I was lucky enough to briefly present at this weeks Marketing and Purchasing Conference organised by the Chartered Institute of Procurement (CIPS). I was invited to speak by Prof Robert Shaw from Cass Business School and the subject was "how we measure marketing effectiveness".

What was interesting was that the "we" in the question involved the procurement teams as well as the marketing and agency teams. For many larger organisations, this makes tremendous sense. What was especially interesting for me was to see that so many procurement professionals were clearly being challenged to assess the question of "what was returned" as opposed to just "what was spent" and it seemed like well established practice for many.
Of course, the reaction wasn't homogeneous and the audience appeared to divide into two groups. The first group knew that were only interested in focusing on cost control - even if they acknowledged that this was only 1/2 the story. The second group were open to looking at accountability but appeared to lack the tools or training to do so (bingo). What was clear was that the procurement function had lot's of tools and discipline to bring to the marketing function and this left many wondering whether the future of internal marketing teams was as a smaller coordinator of effort rather than as the owner of all marketing services for an organisation.
Another interesting discussion was the perennial one about over reliance on historic marketing assessment at the expense of creativity over future campaigns - clearly the agencies don't want to loose their creative authority. However the value of measurement may often be about measuring just how bad things are. There's no need to throw ones self over a cliff to prove gravity and the same is true in marketing. You don't have to spend fortunes to work out whether a message is having an impact.

Monday, September 10, 2007

Online Banner Blindness

Online marketing presents marketers with a real wealth of data and info about what performs and what doesn't. A vast industry has grown up around analysis of online data - this industry is disproportionate in size compared to the amount spent in the channel but the reason is the ability of marketing organisations to "flex" campaigns in real time - hence the need for more analysis.

Jakob Nielsen is considered by many to be a guru when it comes to analysing web sites and their usability and in a recent study, his company turned their attention to online banner ads to try and define the key characteristics of a successful campaign. His conclusions?

"Users rarely look at display advertisements on websites. Of the four design elements that do attract a few ad fixations, one is unethical and reduces the value of advertising networks."

The unethical element that works is making your ad look like native content - i.e. trying to "con" the user into believing the ad is really native content. Sounds a bit like PR doesn't it....

If you can artificially persuade users to click your ads (that seems to be the metric people look at a lot) then the metric itself becomes a waste of time. What does click through mean when you're getting bad clicks. Better to stick with real "end of line metrics" at the end of the day.

I recommend a quick read of his article that can be found here.

So my conclusion - online ads are valuable when actioned. Advertisers can use tricks (some dirty) to persuade people to click on the ads but it's not always going to work out in the long run. Therefore, be cautious when looking at metrics such as click through rates - they can be artificially inflated using some clever formatting and a picture of a naked woman.

Tuesday, September 04, 2007

Cadbury's Update - it's now viral

It was obviously going to happen - the Cadbury's ad has gone viral and already the imitations / mashups are coming thick and fast.

Cadbury's vs. Total Eclipse of the Heart

Cadbury's vs. A skillz

As of 9am this morning, I calculate that 162,794 people had watched these ads on youtube. I'll keep an eye on that number over the coming days. Let's see where this goes.

Monday, September 03, 2007

Words that need to be banned from marketing research - Optimisation

Optimisation - I hate the word optimisation (or optimization depending on which side of the pond you are on) because it implies some degree of certainty in outcome. When dealing in statistics, it's a brave person who puts their neck on the line and says they know for certain the outcome of an event.

In marketing analytics, many people talk about "optimising" the mix, optimising price points, optimising mailing lists. In reality, they mean "probably improving" but it's not such a sexy term is it.

I've used the term myself and will continue to do so on occasion (because I'm weak no doubt) but we must be able to find another way to explain that there is little certainty in analytics.
Mu suggestion - "Improvement" or "Best Suggestion" both sound a bit wet. Anybody got any ideas on terms I can use?

Sunday, September 02, 2007

Cadbury's "'Glass & A Half Full" Campaign

Cadbury’s are desperately in need of some decent marketing. As has been pointed out elsewhere (http://blog.johnbillett.com/2007/06/20/cadbury-gets-it-plain---dairy-milk--wrong-again--again.aspx) their recent track record in marketing isn’t the most successful and it appears to have been that way for a little while now.

On Friday, they launched a new campaign and this time I’m wondering if they’ve gone nuts or they are onto something truly great.

This is what Cadbury’s say about the campaign – “Drumming gorillas, Phil Collins and Glass And A Half Full Productions? Well it just seemed like the right thing to do. There's no clever science behind it - it's just an effort to make you smile, in exactly the same way Cadbury Dairy Milk does. And that's what we aim to continue to do; simply make you smile. So if a drumming gorilla's not enough, wait until you see what else we have up our sleeves. Sign up if you're curious...”

OK – I’m curious to know where this is all going. According to Creative Review (http://www.creativereview.co.uk/crblog/cadburys-dairy-milk-gorilla/) the film comes from Juan Cabral, the director of the Sony “Balls” commercial. It appears that there is more and Comparing this commercial to their previous effort is like comparing a 20 yr old single malt with generic 89p lager – the generic lager is pap consumed simply because it’s there whilst the whisky isn’t to everyone’s taste but will resonate with those who appreciate these things. The big question of course is whether this “Glass and a Half” message will resonate.

On the strength of what I’ve seen so far, I’m a sceptic but impressed that they are trying something different. Clearly their current marketing strategy hasn’t been working but whether this campaign marks a real turnaround is anyone’s guess. Maybe this is a bit like the Vodafone “April showers” campaign that I highlighted previously (see http://marketing-works.blogspot.com/2007/06/will-new-vodafone-campaign-add-do-more.html) – wonderful creative idea waiting for someone to finance it. Let’s wait and see – I’d love to be proved wrong. I’ll let you know what I get via e-mail having signed up.