Moms seek to destroy mass advertising, sales

Saw this quote from an early 20th century advertiser this morning:

Advertising is salesmanship mass produced. No one would bother to use advertising if he could talk to all his prospects face-to-face. But he can’t.

It struck home because my wife was just telling me last night how so many products are being marketed through network marketing plans.
It seems to me the biggest sales competitor to the mass retailer is now the suburban mom. She doesn’t need a major piece of the market it, just enough to make some extra cash.

Mass advertising and sales were a product of the industrial revolution. Higher capacity due to advanced factories had to be sold and distributed to a wider and wider audience. So demand creation became an industry in its own right.

However now, 150 years later, information networks themselves are distributed. Product knowledge is a click or swipe away and distribution costs are much lower, and production processes are evolving to accommodate small-lot output. Public trust of salespeople and advertising in general is waning, especially with the millennial generation who is accustomed to buying with no human interaction at all. And online social networks foster increased communication, mirroring and amplifying offline networks.

All of this means that the sales and advertising functions are ripe for disruption – are actually undergoing transformation this very moment. Companies like stella and dot – cute but relatively benign from a fashion perspective but attractively packaged and marketed amongst friends, are going gangbusters. This has obvious implications for all sorts of businesses.

Friends and family of ours do Ambit energy. Ambit is switching all kinds of people over to it’s service. Energy is a commodity, the same everwhere. TXU’s marketing costs are purportedly 30%, while Ambit’s are 3-5%. If I sign up through my brother-in-law (which I did) and my price is in the lower quartile relative to other offerings, how likely am I to cancel my service?

I know former stay-at-home moms knocking the ball out of the park with this program. MLM programs have been around for a while. But now our truelife networks are being replicated in cyberspace, the rate of network adoption the primary dependency.

The advertiser now can talk to his prospect fact-to-face. Piggy-backing on existing social networks, the sales and promotion functions are being distributed. They are becoming something different than the industrialized birth-child of the early 1900′s.

Using Assisted Conversions in Google Analytics to Drive Content Strategy

This August Google formally launched the multi-channel funnel report. Marketers everywhere shuddered in dearest delight – and rightly so – about the insight it offers into the value attributable to various lead sources, queries, campaigns and more.

I won’t go into the basic stuff you silly, as folk have already done a great job at that like Nick Iyengar’s introduction to Google multi-channel funnels over at Cardinal Path.

I wanted to see if I could tell how we’ve done with growing the long tail this year. Furthermore, can we determine based on the query which searchers are further down in the buying funnel and which are early in their hunt.

From said analysis, could we then derive a content strategy for the targeting of said queries, tailoring the information and presentation to the stage of interest? I proclaim to you it is so. But if I rely strictly on the face-value conversion metrics for a particular keyword in Analytics, i may miss much of the value from eventual conversions. This I cannot do.

Setting up the Report and Advanced Filter

Firstmostly, we derive a measure of the buying interest, disclaiming of course that search intent is a fuzzy monsterbeast that won’t be caged. I’ll list the steps in order then esplain:

  • Conversion Segment: All Traffic
  • Conversions: All. I want to see the value of all goal steps taken, as purchases are not the only valueable goals. Plus it gives me more data to work with
  • Set the primary dimension to Keyword.
  • Advanced Filters:
    1. Include only Keywords of 3 or more words. I used this Regular Expression courtesy of the luxurious Sir Avinash in his post about using RegExp in advanced segments: ^\s*[^\s]+(\s+[^\s]+){2,}\s*$
    2. Include assisted/last interaction conversions with a value less than 1. (Flip this to > 1 to see late-funnel terms)
    3. Include Medium Containing “organic”
  • Sort by assisted / last interaction and filter assisted conversions > some number X (to focus on keywords actually driving significant conversions)
  • Live big or die

Advanced Segment

Key Insights

The result of this excercise is a wonderfully rich list of early-funnel 3-word+ phrases which have brought the company actual value the past year. I really relish a list like this. I’m relishing it right now now I can see more clearly the value brought to the company by early funnel terms and late funnel terms (I’ve displayed the segment for early-funnel above and below)

There are very clearly terms here with more early-stage value

Above are several terms which if viewed in standard Analytics reports would display a number reasonably close the the Last Interaction Conversion Value column. However as you can see there is double the value for many of these terms when viewed from a new visitor acquisition perspective and not merely click – sale / no sale.

Next I sort by lowest assist/conversion number (remember, I’m trying to find the late-funnel terms) and scan the list.

Now sorted by lowest Assist/Last Interaction

There should be fewer surprises here if I have been looking at keyword reports already. One way I use this report often is to check whether a higher volume PPC term is generating high assist value before dropping the bid, if it’s running at a high cpa. If I add a couple of filters likeso:

I see a list of terms with last-click conversions and low traffic.

I’ve used this for keyword research to find queries with low numbers (yet higher than 0) in the assist/conversion column which are especially good low-volume but valuable terms with conversions that we may not have recognized or bid on. There are lots of specific questions, geo-related phrases, 4,5, 6-word questions, people asking for reviews on a particular product, pricing information and such and so.

There are great variant suggestions here for our copywriters and bloggers to add to existing content targeting similar terms, and i’ll pass them to our ppc manager to target in campaigns.

Of course I see a lot of familiar terms here, but I’m looking here for the outliers and long-forgotten ones – ones that highlight questions people are asking that we may not address in a systematic way on the site. I see terms from the senior segment we’re not targeting, jargon of the street we’re not using, etc. This is a great, quick way to add valuable keywords to your target list which may have slipped past.

A scan of these lists shows me query themes that we may be overlooking because we see no attributable value to them in the standard Analytics keyword reports (excepting one variation), i.e., they have no last-interaction conversions.

Many of these show value only in the Assist column. For those of y’all looking to sensationalize the long-tail, this is a haven. In our report, I see enough queries to fill a vat – a spacious vat with room to spare – of specific long-tail content which is not expensive for us to produce at scale. So we have terms which we know to draw converting traffic, are less competitive and thus should rank with little link-building.

That’s it – would love to hear how you’re using multi-funnels to destroy marketing as we know it.

The focus of accurate self concept

There is an eastern saying that gives us the admonishment to think of ourselves with sober judgement. This saying was reverberating through my cortex this weekend as I watched the new series of TechStars, a new Bloomberg News reality show. Think Hell’s Kitchen meets a Microsoft quarterly analyst briefing. Actualmente, it was highly entertaining. The drama of young startup hopes and tribulations mixed with the sober judgement of the stars (Fred Wilson or somesuch of Facenovel fame among them) was intriguing.

However during one particular dialogue David Cohen raised the question, ‘is this team capable of bringing this particular solution to market, and are they in a position to do it faster and more competently than others. I think there are others better positioned to do this.’ Immediately I thought of this eastern saying, and recalled so many ideas I have personally floated which later occurred to me as outside my areas of competency. The man of the east knew then what David knew as an investor: accurate self assessment will save a lot of time and heartache – both our own and others’.

It also forces us to focus my ideation. Now I don’t want to play the idea czar here. You may be creative, outside even one of Rothko’s boxes-type of creative; but if your creativity is ineffectively focused on areas where success is rather unlikely, then it is a hobby. Hobbies can be great – good fun and cathartic, but they may take a long time to become profitable. In a commercial endeavor, if the goal is make a profit and create a sustainable enterprise, accurate self assessment shortens the timeline to success by eliminating the time necessary to acquire a skill or core capability.

Dreaming is great, and teams can be assembled, and capital raised; but others have these at their disposal as well. Ask the question: do I have an angle of attack I can succeed at? Have I tightly defined a niche I can understand better, faster? Do i have the wherewithal to raise funds for this? Will investors trust me? Is this a realm I can demonstrate proficiency in, one to which my demonstrable skills transfer? If the answer is no – pivot your idea. Or else, go get experience. Now you have a track to follow.

1-Question survey insight

A couple of weeks ago I was pouring through the results of a follow-up email we send to new members two days after they’ve signed up. We wait two days because they’ve already received a confirmation email as well as a follow-up email with information they need to use their membership. It was a Survey-Monkey questionaire, and one of the responses was an unstructured question – I can’t even remember what it was now. The respondent was a woman, over 60 years-old. She had taken the time to detail how she used to drive 50 miles to see a doctor and how our plan had enabled her to save money while using a doctor close-by. Great. So I write a quick email to her thanking her for taking time out of her day to help us understand her situation better.

I got an email back from her. An angry one. Problem was: she thought we were providing one service, when we were actually providing another. Even though we say explicitly we are not providing this type of service in above-the-fold locations throughout the site. She was upset. The terms we used in the ad copy meant a particular thing to her, while to others it was a general category which we clarify on the website. She felt irresponsible that she had not read the entire site. So she risked the time to buy + the application fee.

I told her we would take care of the fee, and we will also try to help her find the service she was looking for. But I learned something; for some people, when they type in a query what comes back is the answer. Google is giving them sites which answer it, period. For others, for her in particular, the ad describes specifically the service. The particular risk of buying a product: time + monetary + data risk is evaluated and acted upon. Some are more analytical. Some are more skeptical. Some are intuitive. For some reason, she formed an impression at the ad level, the risk did not seem too great to her based on her previous experience, and the reward seemed very promising (nearby doctor for less), so she bought. This is where we imprinted a new experience on her. Now, since this effort failed, she told us she will be more careful to read the fine print carefully, and would advise her friends to do the same.

What I learned:

We have to balance our ad copy – to make it sensible to as many people as possible so we don’t misrepresent the service or product, while still gaining as much exposure as possible because terms carry different meanings for different people. We must write on our site to appeal to different types of thinkers: intuiters, thinkers, feelers, innovators, skeptics, etc. If we care about customer experience (and if we don’t, there is someone behind us who will), we will do more than a disclaimer stuffed at the bottom of the home page or in an faq.

We imprint by the experiences we create for folks. This woman will now approach the next website differently, and maybe she should. But how many people are we losing because our communication is inaccurate, or because we’re unaware of the effect it’s having on a subset of our customers or would-be customers? By spending 10 minutes getting to know one customer’s experience, I can help thousands of others have a better experience with us and leave them with a positive imprint..which means possibly more repeat business and referrals, and hopefully less disappointment in humanity generally. The very last thing I want is a person walking away from an interaction w/ us feeling that they have failed because they weren’t watching us closely enough.

Protecting the baby

“Customer Service is not a unique value proposition.” Reading this oft-quoted refrain again t’other day, I was prompted in my heart of hearts to respond, “Sometimes yes, sometimes no.” Clearly servicing ones clientele should be a de facto component of the day-to-day business. However, there are companies who leverage a particular talent for service into UVP status. Take the well-known case of Zappos. Tony Hsieh claims to have simply encouraged and guided the culture which was already inherent in the people he had hired, building it into a driver of customer value. The employees are first trained, then empowered to act in the customer’s behalf. But there are two keys:

The right people are hired, rigorously
The company systems are constructed around the principle of service:

  • owned logistics operations so orders are received within 2 days
  • 365 day return policy
  • free shipping both ways

Not in a position to re-engineer your shipping operations? Don’t even own the customer relationship, you say? Affiliates or resellers who toss the new customer over the fence to the primary merchant are unable to claim that business is even theirs, fully. That is the position of the front-end marketer- the special forces recon marketer: seen once, and never again. Well friend, maybe customer service is not meant to adorn your calling card. But I am seeing that defending our core customer segment is becoming more and more important as competition stiffens and broadens. From well-heeled competitors to ones formed of new technology and complimentary offerings, our core segments are being stalked.

So I’ms thinking we need to implement a little something I like to call…a “conversation”™. If that’s the first time you’ve seen that term then…I invented it. I may not know how to talk real efficiently with them over the tweetbook or freindster portals, and I may not have learned Excel yet, or Word or DOS. But I can pick up the phone and talk with them. I can post a phone number on the homepage and ask them what they like, don’t like, what hacks them off, what they’ve been surprised by. I can write it down on the back of a milk carton or in my trapper-keeper, tweak my questions as I see responses are biased or confused as to intent. I could learn to chat with them on Liveperson (whatever that is) and ask them how I can help, is there anything they don’t understand. I can throw up a KissInsights survey on my website (if I have a mainframe coder handy, gaw) and ask all sorts of questions right at the point of content interaction for maximum relevance™ (mine too).

El punta es que we have all of the resources we need to develop our customer learning mechanism. No, it may not rival the TI81 in its sophistication, but it will get us to ground zero, our ears to the grindstone, nose to the metal. We’ll be talking to those very important people who trade for our wares, and we can thus begin our marketing efforts, at the beginning.

My niche’s pie size

I have been working for my company now 18 months. As I look back at the ppc account history in Adwords, I see a steadily upward trending line representing clicks & conversions. Nice. When I toggle the graph to show cost/conversion, there again I see the same upward trending line. Not so nice.

When I overtook the account, there was quite a lot of optimization to be done and so naturally, I looked great in spite of my frightening lack of experience and know-how. Now however, the improvements are more difficult to attain.
We are doing the normal things: optimizing and testing landing pages, testing new ads, new creative on the display network, adjusting bids, adding longer tail keywords, testing fandangled new targeting as it rolls out in beta.

Here’s the thing – we can get more impressions and more clicks, but conversion doesn’t improve at those higher positions. So we pay more for each customer.

Our niche is a very specific component of a larger industry in the finance market. We push out into targeting tech such as audiences, interest groups and toss in with ad networks like Adknowledge, Adbuyer, and others I can’t remember right now. And maybe seams of traffic will open up like rifts in the earth over many millenia. But I can’t help thinking that the suspicion crawling into my cortex is plausible: are we close to maxing our reach at current profitability? In other words, with our current offering and presentation, will we be able to profitably convert the next orbit of traffic as we push further out into the atmosphere of potential customers? Those not actively searching for us. Those not actively on sites relevant to our offering. Those flitting around content hubs passing our banners like billboard decoration at 75 mph. Maybe some of them will be “hotter” according to their recent search behavior, but will it be enough?

I guess what’s I’m saying is: it’s time we took a hard look at our product, at ourselves- who we are and what we provide. The competition has caught on and most of the easy money has been gotten. Now it’s time to evolve or reinvent.

Geo Snafu cont’d..

So I talked w/ an associate who does a lot of advertising on Adwords and he says if you target locally based on google maps or mobile, then someone living in NG near Boston, whose ISP provider routes their traffic to a Boston datacenter, then Google will think the user lived in Boston, not NH.
He reiterated that geo targeting isn’t always by address and state; it’s by the user search query and/or IP adress; and added that IP address mapping is good overall but no mucho.
So my bottom line for now is: if I have a metro area that spans state borders which gets conversions and is close to my threshold on cpa, then I will leave it included. If it is dramatically above my threshold I will cut and watch it die.

Meaningful words coming soon..

these are not them

This weblog is a place for me to process the discipline of online marketing:
seo, paid search, display, html, css and the like, as well as how people communicate and understand. And other-type stuff as well.

Not a seasoned veteran, i have made a leap from an 11-year career as a homebuilder and remodeler to the online, and as I make haste to learn as much as I can as quickly as possible, i will regurgitate what I’m dealing with day-to-day here in the hopes of getting feedback, correction, witty banter, curses or commiseration; all in the hopes of learning faster and hopefully helping at least one other person to do the same. Oh – and I want to make a million dollars by December. Fair enough?

Day 545: Adwords Geo-targeting Snafu

It’s good to see everyone. Actually, I forgot to mention I’ve been in this position now as seo/ppc..person for about a year and a half. So the other day I decided to pull what I call the Full Monty Geo Report from my dimensions tab in the Adwords ui, and slap it in a pivot table and try to mine some specific regions or metro areas where my CPA is higher than our targets.

I set my date range to the past 6 months to get enough clicks and impressions in each ad group and location (this campaign has been running a while at roughly 800-1200 clicks per day, so i was good there), went to Dimensions tab, and made sure i had set columns to include state, metro, region, conversion, impressions, costs – and that’s it. Once in the pivot table, i started by selecting region for the rows and impressions, clicks, conversions and cost/conversion for columns.

pivot table adwords export
this is how I layed out my pivot table - pretty simple

What I found was that I had a handful of states w/ high cpa, so i added the metro area beneath the regions to see if there were some metro areas which were the culprit and not the entire state. In fact there were, so i highlighted them + a few states which had 4-500+ clicks and cpa higher than my goal. We sell a product in the finance vertical, and i have a good idea that certain states don’t produce as well as others already.

So, I excluded about four of these areas: 2 metro areas and 2 regions (states) in the campaign settings. Then I went away with the thought that I would come back on Monday and begin to see an improving cost/conversion. However, what I found after about 5 days, since I neglected to check it Monday, was a drop in impressions of about 20% and a steadily falling conversion rate. Now, this campaign only gets 10-20 conversions per day (online), so possibly this is simply in line with its normal fluctuations – however maybe not..

I went to Google analytics and checked the ad groups for monday compared to monday one week ago to try and isolate the drop – and then i went back to adwords and looked in those adgroups at the change history link: the drops were indeed correllating with my geo changes. Soooooo, going back to excel, i pivoted again and realized that while Boston-MA-Manchester/NH was a dog, Boston-MA-Manchester/Massachusets was performing quite well. I had not accounted for the fact (I suppose this was it- I actually haven’t validated this) that many metro areas, which are actually composed of hosting provider ip’s (right?), stretch accross state lines. So while my campaign was a dog in NH, Mass folks were still buying. From what it looks like to me so far, excluding the metro area in one state seems to have excluded it in the other state. Another example was the Mobile MA, which stretches across [3] state lines – Missip, Albam and Fla.
anyway, i haven’t checked back yet, but this is as it seems to me right now.
out