The right paid search platform can give you a competitive edge – you can spend more time optimizing your paid search programs and less time dealing with repetitive issues.
Choosing the “right” solution isn’t as easy as it sounds. No one knows that better than PPC Summit. They conducted primary research to understand the pitfalls to avoid, and the strategies to succeed, in buying technology to help you manage paid search. They interviewed over 50 marketers and surveyed 14 vendors.
In this webinar, you’ll discover what’s going wrong in the buying process, issues peers have faced, and how they’ve dealt with it. Mary O’Brien from PPC Summit will cover the top 10 mistakes to avoid, including:
Why RFPs don’t always lead to the best decision
Getting stuck in the feature trap
Failing to define and prioritize your business needs
Every webinar attendee will receive a free copy of the Executive Summary from the research and a discount code to purchase the full report!
Forrester Research Inc. released their latest coverage of paid search platforms, “The New Paid Search Automation Landscape” February 22, 2011. ClickEquations was one of eleven enterprise vendors profiled in the research.
”The paid search automation landscape has grown to include more vendors that provide enterprise-level bid optimization, campaign management, and reporting. And automation can help marketers control their keyword programs directly while also cutting costs and saving time,” says Forrester Research.
Forrester Research’s report was based on interviews with vendors and end customers. The report covers:
How Tools Support Bid and Campaign Management
Why Use a Paid Search Automation Partner?
The Paid Search Automation Technology Landscape
Recommendations for who should buy paid search automation technology
Forrester Research’s Paid Search Automation Technology Landscape includes a list of all companies, revenues, customer base, estimated revenues and who the company is best for. The landscape highlighted ClickEquations’ “guidance to prioritize the most important tasks” as a key differentiator and noted that “ClickEquations helps marketers stay on top of their programs by alerting them when campaign conditions change and they should take action.”
Next week is the first Search Marketing Expo of the year in San Jose, aka SMX West. If you haven’t registered, you can use the code smx10click to get 10% off, register here.
I’ve lost count, but this will probably be my 9th SMX conference. Along the way, I’ve figured out a rhythm that works for getting the most out of the conference and I thought it would be helpful for other folks.
If you’re coming, stop by the ClickEquations booth (#223) to say hi or shoot me a note ahead of time acohen @ clickequations . com
.
Before The Conference
The biggest value you’ll get out of the conference is the networking and the war stories you swap with other attendees. You don’t have to wait until the conference starts, though. Everyone gets an invite to SMX Connect. I recommend that you find attendees you’d like to meet and flag them as “want to meet”. It makes it easier to have a good lunch conversation.
Speaking of lunch, you should definitely log into SMX Connect early to sign up for the Birds of a Feather lunches. They’re tables organized for people who want to talk about specific topics, e.g. Paid Search or SEO.
Let’s be honest, if you care about, say, local search, then you probably don’t want to eat lunch with a bunch of Twitter nerds. The BOAF lunches are an easy way to make sure you get to chat with folks who are interested in what you’re there to learn about. You can only sign up for them on SMX Connect and they always book up in advance.
Craig and I are both hosting a BOAF lunch on Tuesday (Day 1)
Birds of a Feather Table 2: Paid Search – Craig Danuloff
Birds of a Feather Table 8: Search & The Complex Sale – Me (Alex Cohen)
.
Pre-Conference – Monday 3/7
Monday is when most people arrive. That night is the SMX Meet & Greet from 6 – 7:30 in the 2nd Floor Ballroom Foyer of the San Jose Marriott.
The event is casual drinks. It’s worth going in particular if you want to meet certain speakers and veterans. They’re not swamped after a panel and the event is more intimate. Plus, you never know if you’ll get invited out to a dinner or more drinks with a good group.
.
Day 1 – Tuesday – 3/8
This is the first day of the conference. The sessions change every show, so I can’t use past experience to recommend certain session.
What I can say is that the value of the session has almost everything to do with who is presenting. A speaker who can convey actionable tips in an easy to follow way is worth seeking out. Here are some I’ve personally seen present and enjoyed at past conferences:
Day 1 is also when Craig and I are hosting our Birds of a Feather lunches (see above).
This is also only 1 of 2 days that the expo hall is open (it closes on Thursday, which people sometimes forget). Check out the list of exhibitors. I’ll be working the ClickEquations booth, #223.
At 6:00, you can get free drinks at the Cocktail Reception in the expo hall (they give tickets to you at registration). It’s a good chance to unwind and see some of the vendors before heading off to dinner. See the section below on how to make the most of the expo hall.
After the Cocktail Receptions, people will generally splinter off for a quick bite before heading to the Internet Marketer’s Charity Party (Agenda Lounge, 99 South 1st Street). It’s $50 to get in, but the money goes to 2 good causes, it’s open bar and there are a bunch of great prizes that are given away. I recommend it.
.
How To Explore the Expo Hall
Most conference attendees will never work in a trade show booth, but I have quite a number of times. Let me share a few tips on how to get the most out of the conversations you have there:
The quietest times are, obviously, when the conference is in session. If there’s a time when there isn’t a session you want to attend (it happens), consider swinging through the Expo Hall then. You’ll get undivided attention.
Make sure that you attend some of the Theater Presentations. Theater presentations are 20 slots where vendors can do a presentation on whatever they like. They’re like a conference within a conference.
As you may suspect, some can be overly commercial. Good vendors (like us!) make them educational, because no one wants to hear us talk about our admittedly awesome paid search software forever. You can always leave if you don’t like it and there is often a giveaway.
Check out our Theater presentations on Tuesday at 2:00 – 10 Ways to Improve Paid Search Campaigns. Craig Danuloff, our Founder & President, is presenting.
A note on trade show giveaways (aka swag or tchotchkes). If you just want the free stuff and don’t have an intention of ever buying the product, just take it and enjoy. You don’t have to pretend to be interested. It’s better just to either not take it or politely ask for some and make way for folks who want to talk shop.
Also, be sensible! I once saw a woman take 15 stuffed monkeys from the shocked booth next to me. I felt sorry for that poor marketer’s budget.
Not every vendor in the expo hall is a fit for every attendee. For example, the first question I always ask folks is how much they spend on paid search advertising every month. If the answer is “$1,000 a month”, I’ll honestly say that it’s not a fit and give them a white paper with tips on how to help grow that spend profitably. You and the vendors have limited time. Focus on the conversations where it’s a fit.
Don’t forget your business cards! If for nothing else, you can enter raffles
.
Day 2 – Wednesday – 3/9
Assuming you didn’t go out after the IM Charity Party, there are plenty of good sessions in the morning. Here I’m going to give an admittedly biased plug for the two we’re on.
10:45 – Paid Search Testing and Reporting – Craig is on this panel, but it’s also moderated by Matt Van Wagner (who always does a great job of organizing) and Addie Connor, who is just plain awesome and brilliant.
1:30 – Search and The Complex Sale – I’m excited to present on this panel with a lot of folks I haven’t presented with before. Patricia Hursh is moderating and she’s a veteran of B2B panels. Sneak preview: I’ll be talking about how to use increase form registrations without losing contact information and how to use that effectively in managing your campaigns.
Note that the Expo Hall closes at 3:30 and they seriously mean it closes then. They literally start rolling up the carpets! It’s best to get there earlier in the day.
There are no dinner time events (though some vendors plan them), but the final networking event, SMX After Dark, is held at Motif (389 South 1st Street) starting at 9. It’s free, so most folks will be there enjoying the open bar.
.
Day 3 – Thursday, 3/10
A lot of people leave before the last day. Don’t. Most of the expense of going to a conference is in flight and hotel, so it’s best to get your money’s worth by hanging out for the third day.
The expo hall is closed and there are no events planned in the evening. I’ll be hanging around until Friday, so shoot me a note if you want to grab a bite and talk shop.
.
Bonus: Things I Often Forget to Pack!
I travel a decent amount for work and I’m kind of a forgetful guy. To keep my brain from failing, I’ve created a packing checklist to make sure I don’t miss anything critical. I listed some of the ones you might also forget.
In particular, remember to bring extra business cards. I’d say 1/2 don’t bring enough (50-100 is a good range). Why do you need chapstick, cough drops and gum? You’ll be talking to people in dry, air conditioned rooms and nobody likes to talk to the guy with bad breath.
Business cards
Cell phone charger
Camera/video camera
Sweater/hoodie (it gets cold in those conference rooms!)
The previous post (Bid Management Is Dead) probably should have titled “You’re Not Ready For Bid Management”. If you go back and read it again under that title it’s somewhat less controversial.
But it probably would have garnered a lot less tweets and comments, so looking back maybe it wasn’t a bad decision.
Bidding Isn’t That Important
The phrase from the post that got the most attention, aside from the title, was ‘bidding isn’t that important‘. Many of the thoughtful comments from some of my favorite PPC’ers were defending the honor and importance of bidding. Obviously they’re correct.
But ‘bidding isn’t that important’ was preceded by ‘For most paid search advertisers’ and I’ll stand behind the full sentence and even repeat it: For most paid search advertisers bidding isn’t that important.
The point of the post was intended to be the one made by the cute little skiing analogy – advanced tactics aren’t necessary until after you’ve mastered, or at least made substantial progress on, the basics.
Every month get to look inside dozens of new paid search accounts. From clients to prospects to folks I meet on airplanes who just want me to help them with their quality scores, I see exactly how all these accounts are configured in terms of keywords, match types, ad group organization, bids, and all the other factors.
It’s rarely a pretty sight.
Is The Keyword Bid-Ready?
The vast majority of paid search accounts that I see are terribly constructed and poorly managed. They’re filled with entirely too many broad match keywords. They have way too many keywords per ad group. Their ad copy is completely uninspired, and worse shows no signs of having been well tested. Their quality scores reflect these facts. And their bids do not seem well chosen.
This is who the last post was aimed at. The huge number of account managers – and sadly I don’t even think they know who they are – that need to work very hard on fundamentals before getting deeply into bidding strategies or tactics.
My advice in these cases is to fix these problems in the order I just complained about them. In such massively sub-optimal accounts, bidding is not a priority.
My worry, because I see it several times each week, is that these advertisers and even professional search managers hear so much about bid management as a core element of paid search that they’ve been missed the fact (not surprising because I’ve never heard it mentioned anywhere else) that there is a pre-requisite to bid management. It’s a 300-level course and you can’t take it until some other requirements are complete.
What The Pro’s Know
As several of the smart folks from RimmKaufman pointed out, paid search management is an iterative activity, and search managers can walk and chew gum at the same time. Ideally, you’ll make the best bid decisions possible at every stage of campaign development and tuning. If you’re playing the game at their level that is absolutely correct.
But with respect I’d suggest that the RimmKaufman’s (and their lucky clients) and Zaharias’s and Summerhill’s and Fergie’s are all PPC PHD’s (and beyond). These folks and anyone like them obviously wisely spends considerable time and resources on bid intelligence.
What Many Others Don’t
Unfortunately, it isn’t easy for most paid search advertisers or managers to know if they or their account or keywords are bid-ready. Ironically in an industry with as much shared knowledge and free resources as any there is neither an automated account grader nor a simple and widely accepted set of rules-of-thumb to simplify self-assessment.
If I had to compose a test, I’d start with these questions:
What percentage of your revenue is driven by Broad Match keywords?
How many keywords are in your average ad group?
What is the largest number of keywords in any ad group?
How many keywords attract over 25 unique search queries?
How many text ads were testing in each ad group before you arrived at the ones now running?
What percentage of your keywords, on an impression weighted basis, have quality scores below 7?
Questions like these and perhaps a half a dozen others, can give a sense of how well the basics of PPC have been implemented and executed.
The Truth About Bid Management (Part 1 of 100)
Yes bids need to get set along the way. The ‘perfect’ answer to all of the above questions is not reached easily or quickly. Yes bidding in an iterative process that ideally would be done and redone as this road to perfection is achieved. And the farther one gets the greater percentage of time and level of effort bidding deserves.
My argument ‘against’ bidding is really just the suggestion that there is a threshold – a minimum level of campaign quality – before fine tuning bidding should much of a priority.
I believe this both because until that time non-bidding activities have more impact, and because the data generated by an ‘under-developed’ campaign is dirty and a poor basis for any bid algorithm or strategy anyway. That last bit will be the subject of more posts in the near future.
So my esteemed colleagues are right. Bid management is not dead and can in fact be quite important. But I think that the point of the previous post was completely valid. For many advertisers, depending on the state of their account and their skills, bidding is overemphasized and over-rated. Bidding is also hard to get right, easy to get wrong, and I believe many elements of it are dramatically misunderstood.
Regular readers of this blog know I have a lot of pet peeves related to paid search, but perhaps the largest is the fact that paid search management – the process and the software that supports it – is often referred to as ‘bid management’.
The use of the term bid management to summarize paid search management suggest a profound lack of understanding. Of course, many people genuinely think that’s what the discipline is called. They don’t really mean anything by it – they’re just repeating the phrase they heard.
My problem with the phrase is that it perpetuates a level of importance for bidding that is undeserved. For most paid search advertisers, bidding isn’t that important.
Putting Bids In Perspective
If you’re a good skier, sharp edges matter a lot. When you’re learning to ski, they hardly matter at all.
First you have to get comfortable just standing on skis, and then standing on them while sliding down the hill and riding over all kinds of terain. They you have to figure out what to do with the various parts of your body – ankles, knees, hips, arms, head – while all this sliding is going on. You have to learn to initiate and complete turns, and at some point you start subtly shifting your weight through the process.
Until you reach a certain point, it hardly matters if your skis are slabs of wood from a garden fence. As long as they slide downhill, they’re fine. But eventually your skills reach the point that their length, flex pattern, and ultimately fine tuning points like sharpness are both noticable and impactful. But only after a long list of much more fundamental issues have been resolved.
Paid search bids are like that. Initially, any bid that gets your keyword shown in the top 2/3 of the first page is fine. After that, it’s not worth worrying about it again until a lot of other more important skills are mastered, tasks are completed, and data is gathered. When they are, and only then, is it worth allocating serious time to bidding, and even attempting an intense and sophisticated approach to bidding.
Spending a lot of time on bidding when the other attributes of each keyword aren’t in very good shape yet is like sharpening your skis when you still fall down a lot. It’s not the place you need work.
First Things First
This ‘sharp edges’ analogy actually doesn’t even do justice to the problem with premature bidding.
Generally speaking, sharpening edges actually improves ski performance and have a very small chance of actually causing problems for even a beginning skier.
But bids actually cannot be reasonably be calculated for keywords that are not haven’t already been properly tuned in terms of organization and negatives and match types and ad copy. The data surrounding these keywords is garbage data – putting that into even a very clever bid algorithm or calculation results in a garbage bid suggestion.
So attempting to bid prematurely results in inappropriate bids which will *cause* poor performance and potentially skew the data that will be used to drive future decisions.
Bidding on a weak foundation isn’t just wasteful, it can be harmful. It produces bad numbers based on incorrect assumptions that serve to drive you further away from optimal future results.
The Fairy Tale
The idea that paid search success is driven by keywords and bids hasn’t been true for many years, but it remains the dominant narrative of our industry. Don’t fall for it.
In upcoming posts we’ll examine why you should bid no keyword before its time, the steps you can take to prepare keywords for bidding, the tests that will tell you when a keyword is bid-ready, and then finally we’ll take a long hard look at how to bid once it is actually time to do so.
It’s a well covered topic and we on the panel all agreed that an “either/or” argument was too cliché. Let me suggest there are 3 fundamental things that form the basis of any PPC + SEO discussion:
Everyone should be building search engine friendly websites and, as their resources allow, doing SEO in its proper form (content development, research, link building, etc.)
Most businesses will find paid search to be a profitable way to acquire customers, because of the high degree of intent from searchers and the ability to target, value and satisfy that intent on a granular basis. A simple pilot to measure incremental profit from the channel will tell you.
Having an organic listing when you have a paid list is probably a good thing. Having a paid listing when you have an organic listing may be a good thing. Your mileage may vary and there are ways to test and measure if you don’t trust people who have done it before.
My focus is on one trend that is changing the nature of the search engine results pages (SERPs): the shift in the balance between paid and unpaid listings.
Consider this search for “anti virus software” as an example*:
Paid search ads dominate the search results. In fact, the only organic listings you can see are AVG and avast! (I’m counting the Shopping results as ads).
The total number of possible ads hasn’t changed. Given adequate advertiser interest, bids, ad quality and the commercial intent of the search, ads could always fill the top 3 slots (aka premium position) and the right column.
What has changed is the space taken up by, and visual dominance of, the ads in several ways:
Most notably, there are now display ads in the SERPs. Google’s Product Listing Ads show product images on the top 3 positions in the right column.
As of last week, ads are now showing longer titles that pull in the first description line, mimicking the title tag of organic listings
Most ads in the premium position will show site links, effectively adding another line and 4 more calls-to-action to the ads
Retailers with an sufficiently high number of positive ratings will get a call out with stars and the number of reviews, distinguishing the ads again and contributing yet another line to the text ad
The background of the ads has switched to pink. Personally, I find this harder to distinguish vs. white than the previous yellow and blue incarnations.
This is the most dramatic example, but we see it in other searches as well:
“kitchenaid mixer”
“yoga bag”
I have always thought of Google as a publisher and each search results page simply as content they need to monetize. These changes are similar to a news site stacking on more ad slots or increasing the size of their ads.
But, where a traditional publisher’s ads don’t really compete with the content, Google’s ads do. Paid search ads may now draw attention away from organic listings and push natural results further down the page.
In essence, paid listings may be getting clicks at the expense of organic listings at an increasing rate.
At the very least, it puts more pressure to be among the top organic listings as they lose prominence in the results. It also adds important context for your SEO analysis, essentially a new set of competitors with new tactics.
On the flip side, paid search ads now offer an increasing number of ways to advertise in the SERPs. There are more levers to pull, formats to try and offers and messages to test. If you’re weak in the organic listings, these new controls may give you an edge.
One final note: not every industry will be affected by these changes equally. In some cases, there are few, if any, paid competitors. For example, media and entertainment searches such as “justin bieber” or “somewhere movie” are dominated by video, image, news, and real time results among the usual organic listings.
*Screenshots are from my Dell D620 laptop. Your experience may look different based on screen size and personalization of search results.
2011 will be a pivotal year for paid search advertising. Trends set into motion last year – new ad formats, an emphasis on display, a local ppc push – are on a collision course with the new complexity in advertising and the maturation of online marketing.
This is the year that will fundamentally change how we think about and buy access to prospects, namely keywords. 2011 is the dawn of paid search without keywords.
.
Changes from 2010: Universal Paid Search
For years, Google’s search results were dominated by the “10 blue links” – simple headlines, descriptions and URLs to entice and satisfy searchers. Until it wasn’t. Universal search wove in images, video and real-time updates.
For most of its history, too, AdWords been presented in a text format even as the search results morphed into a multimedia experience. The result is that attention was pulled towards organic results at the expense of ads.
Google countered that trend with their big push for universal paid search in 2010. It was, perhaps, the most radical evolution to the paid search results since the introduction of Quality Score. Consider the changes:
New Ad Formats – Text is no longer the exclusive medium for advertising on Google. No format exemplifies that more than Product List Ads (and their cousin, Product Extensions). There is no headline, copy or display URL. Instead, it’s just a product image, name, price and vendor slotted in the highest positions on the right side. What’s more, you don’t choose keywords. We also saw display creep into image search results with Image Search Ads and traditional display ads
New Calls-to-Action – The way you satisfy your search with advertising on Google has evolved as well. Most notably, through the introduction of click-to-call as an option for mobile search ads (as well as the limited release AdWords call metrics). Similarly, more and more of the site experience is being pulled into the search results. The beta Comparison Ads creates a marketplace for loan and credit card comparison all on Google. The call to action is comparison and filtering, not just clicking on an ad.
New Buying/Monetization Models – Cost-per-click (cpc) and cost-per-thousand-impressions (CPM) are no longer the only ways you can buy. Comparison Ads are sold on a cost-per-lead basis. Product listing ads are sold on a cost-per-acquisition (CPA) basis for some advertisers (CPC for most).
New Display Targeting Options – Remarketing (aka retargeting) brought highly focused display buys to the AdWords interface. Specifically, the ability to only show display ads to segments of people who visit your site, in many cases after clicking on a text ad.
New Advertising Automation – In a move that radically simplifies advertising for small businesses, Google began testing Google Boost. It involves no keyword research and no bidding. If you have a Google Places page, you can even do it without a website. It’s virtually hands-off advertising for SMBs.
Of those changes, Google Product Listing Ads and Google Boost offer the best glimpse into the future of paid search without keywords. They’re notable for dramatic departures in every step of how you advertise on Google:
Targeting – automated targeting towards certain audiences as determined by Google vs. keywords chosen by the advertiser
Ads – Product listing ads bring a product search like result in the top position in the right column and Boost promotes a map-like result in a preferred position above organic results.
Pricing – CPA and monthly budget caps replace daily budgets and CPC bids.
These changes underlie some of the key trends that will shape paid search in 2011.
.
Trends that Will Shape Paid Search in 2011
In order for Google to continue their pace of growth, they need two things.
First, they need to another line of business to complement AdWords and display advertising is it. They’ve pushed even more aggressively into the channel, most notably with the acquisition of Invite Media, a demand side platform.
Second, they need to remove obstacles to profit and incremental growth within AdWords. These barriers are primarily how wide advertisers target and how much they pay for the people they reach (see: Why Google Wants to Eliminate Bidding In Exchange for Your Profits).
As ad formats and advertising choices multiply, so does complexity. The third barrier to incremental growth is bringing new advertisers into the market and generating add-on revenue from existing advertisers. That requires reducing barriers to entry and scale.
His vision of search in 5 years is already here today and my prediction is that paid search without keywords will become even more prominent in 2011.
.
The Dawn of Paid Search Without Keywords
As search marketers, we’re used to buying based on intent as expressed in the search query. We do that primarily through keywords paired with match types and bids.
In paid search without keywords, intent is just one of 5 ways we’ll target prospects:
By Intent – Search behavior, as expressed through search queries.
By Audience – Demographics and psychographics, similar to traditional media and display ad buys
By Business Type – Certain verticals (personal finance) and markets (small businesses) will have special targeting
By Product – Online retail is one of the biggest spenders online and Google’s Merchant Center can bridge products and ad space
By Behavior – Prior behavior and interests can make some audiences more valuable than others
We’re seeing this already today in various forms, such as Product Listing ads and buying on the Display Network. Kevin Lee offers a compelling vision of how real-time bidding could further expand the way we buy paid search.
.
How We’ll Advertise on Google
The difference between how we buy paid search today and buying without keywords is fundamentally about subtraction instead of addition.
Think of a paid search account today. We target a niche and then gradually add other targets that define your total potential universe. We’re adding pieces to make a whole.
To consider the other side, let’s take the example of Product Listing Ads. We don’t choose when our ads show up; Google does. But we can set limits of what we’re willing to pay, effective preventing products from potentially showing up. We’re starting with a whole and subtracting out pieces.
Remarketing offers a model for how this process will work.
We can set the target audience you want to reach, say people who put an item into their shopping cart.
Then, we can establish exclusions to remove some undesirable part of the audience, say those who complete a purchase (cart recovery being the desired goal in this example).
Finally, we set preferences for your buy – frequency caps, etc.
In paid search without keywords, we’ll start wider and have to be vigilant about segmenting and excluding. We’ll also have to start dealing with a very sophisticated competitor: Google.
.
Google as Competitor
There is something of a quirk in how new keyword-less features are being added, that presents a new dynamic in the auction. In some cases, these new tools offer a choice between managing the advertising, or at least the bidding, ourselves or giving full control to Google, as is the case with CPA pricing for Product Listing Ads.
That creates a hybrid auction where some advertisers are controlling bids, while Google changes the bids of others while deciding whether both are eligible/relevant to appear. In effect, you are placing your bidding wits against Google’s.
It remains unclear what controls these new ad formats will include, what degree of reporting and the impact to traditional text ads, which will also compete for space and attention.
.
What’s Coming Next in Paid Search without Keywords?
It wouldn’t be a good predictions article unless I put my neck on the line with some solid ideas about what 2011 will hold for paid search marketers.
I think 2 things are especially likely:
Impression Based Retargeting – Retargeting was a huge step forward for Google in display last year. But, they’re leaving the fast majority of the opportunity on the table. If you’re average clickthrough rate is 5% (which would be generous for some), then 95% percent of you’re the people who were served your ad aren’t even eligible to be retargeted to. But, what if you could target someone who searches, say, “organic dog food” with a banner ad on the Display Network? It opens ups a world of possibilities and budgets.
Automated Inventory Based Product Ads – I always think it’s interesting that Google created an umbrella category for Product Listing Ads and Product Extensions called Product Ads. With retail being a significant vertical in paid search with large, constantly change inventory, I expect Google will continue their push to use Google Merchant Center. For example, we could easily see automated retargeting or regular banner ads filled with products you viewed or searched for, just like Criteo is doing today with Zappos.
.
What This Means for Advertisers and Agencies
No one but Google knows what’s coming. But, it’s clear that everyone needs to get comfortable with audience based buying, starting first with retargeting. The skills and tools to understand, target, and bid on various audiences exist today. Now there’s an increased urgency to dig into them.
On a final note, I specifically resisting the linkbait urge to title this post “the death of keywords”, because they’re not going anywhere. AdWords is too critical to Google’s bottom line to disrupt that revenue until new ad formats and monetization models are proven and well adopted.
One thing is certain, though: the future of paid search without keywords is coming in a big way in 2011.
Did you ever notice that every time you have first hand knowledge of anything written about in the press it’s clear that the largely or completely misrepresent the facts or misunderstand the issue?
Session-based broad match is an AdWords ‘feature’ that considers the past search queries of any users when deciding which ads to serve for them. So if someone does a search for ‘rental palace in Monaco’ and then does a search for ‘cheap dog toothbrush’ Google may decide to show an add your ran against the keyword ‘luxury Monaco rentals’ in reply to their toothbrush query.
The theory, which is reasonable, is that Google knows that user was very recently interested in those rentals. Why not show them the ad a few minutes later? That’s still relevant.
Missing The Point
In The Journal article they find advertisers who are not happy about paying for clicks unrelated to the users current query. They also find folks (like PPC RockStar himself David Szetela) who don’t mind and have had good experiences with the feature.
They then ramble on and back and forth about if what Google is doing is cool or uncool.
The 500lb gorilla never makes an appearance.
Why isn’t ‘Session-based Broad Match’ a user controlled option?
The article doesn’t even introduce the idea that the ‘solution’ to the grand question of ‘is it good’ or ‘should people pay for it’ is ‘let them decide’.
Perhaps in the Murdock tradition the WSJ now operates with the goal of only exploiting problems and not wasting any breadth (or ink) on solving them.
Make It An Option
At minimum Session-Based Broad Match should be an opt-in campaign-level feature. Or better yet there should be an option to bid differently for session-based impressions.
The problem isn’t the feature. The problem is that as an AdWords advertiser you don’t get to choose whether or not you use it. Google decides to show your ad and bill you for the click in a way most people didn’t intend, don’t understand, and may have valid opinions or business reasons to want or not want.
Bundling session-based broad match without offering any control reduces advertiser control and transparency – session based clicks are reported as such in one report but are generally hard to detect so many don’t know when they’ve happened.
In my experience the vast majority of advertisers are surprised when they first hear this ‘feature’ even exists. Few know they they’re paying for it, most likely in very small amounts but on a regular basis. That’s no way to treat your customers.
AdWords added many great and complex features this past year, and extended advertiser control with things like Modified Broad Match. They have the resources and capability to make Session-Based Broad Match an option.
I remember the conversations we used to have in the mid 90′s when trying to figure out what to call the industry that would let people buy thing over the internet.
At the time AOL was a major force, and was known as an ‘online service’ while trying to pretend that the internet didn’t exist. For that reason I never liked the name ‘online commerce’ and steered towards ‘electronic commerce’ which ultimately turned into ecommerce.
Eventually AOL pretended to embrace the internet, the stigma fell from ‘online’ and the terms became interchangable. More important than the name, was the idea that the goal was to get the sale, or capture the lead, via the web site or page. That online conversion was the focus.
We’ve moved through probably five generation of ecommerce in the last 15 or 16 years. Today the role of the site or page extends far beyond seeking an online conversion. Often it’s just a step on the way to some other larger deferred event.
What starts online now ends in call centers, in leads that are sold or valued days after the site visit, or in all forms of offline and other transactions and value events.
External Conversions in ClickEquations
Last week we upgraded ClickEquations to allow all types of external conversions to be easily imported. This allows you to see complete reporting on all the positive results from paid search, and to have all of them influence bid decisions for your keywords.
Businesses which have conversions take place in call-centers can use this to finally connect the revenue captured on the phone to the specific clicks that drove people to the site where they got the phone number.
Lead-gen companies who don’t know the final value of leads until days or weeks after the site visit, can now connect those ultimate values to the specific costs of acquiring each unique lead.
And it’s not just initial offline or deferred revenue. This new capability can be used to bring subsequent purchases into the system – so you can see and consider the lifetime revenue impact of each keyword and text ad in your campaigns.
Better Reporting and Control Over Multiple Conversion Types
To support the new external conversion tracking in ClickEquations a number of other enhancements have been added. One that offers a lot of flexibility is that you can now customize the display names for all online and offline conversion events, and choose which events to roll-up into summary conversion and revenue metrics. For example, a web retailer may want to combine online sales and call center sales into their revenue numbers, but also track lifetime values as a separate column in their reports.
Full report customization is supported through the platform. Choose any individual online or external conversions or roll-ups in any report, in the management module, or in our Excel plug-in. However your business collects, defines, and tracks conversions and revenue you’ll be able to report on it exactly the way you want in ClickEquations.
Considering Multiple Revenue Streams When Calculating Bids
The most important reason why it’s important to bring every form of revenue into your paid search platform is so you can accurately match costs with revenue when deciding on future bids.
If until now 25% of your revenue was captured offline, but you or your system were calculating bids without considering that revenue – or without being able to allocate it across keywords accurately – then there has likely been a lot of room for improvement in your bidding practices.
ClickEquations now provides the ultimate in customization, allowing you to choose which combination of revenue events to consider when calculating each unique bid rule within your account. This makes it possible, for example, to only consider online revenue when calculating bids on some keywords (brand keywords, for example) while considering both online and offline revenue when calculating bids for others.
Standards Change
There was a time when clicks were the thing to count but then conversions took over. There was a time when ROAS was a useful metric but eventually it became clear that ROI was far more accurate and important. There was a time when attributing all the revenue from a sale to the last click was the best we could do but then a variety of more realistic attribution models became available.
Each of those transitions moved paid search reporting and management forward with better clarity and a more realistic measure of the business. Including sales and revenues that happen after online clicks, but as a result of them, is another huge leap forward.
We’re excited to be continuing to move forward with ClickEquations.
Testing is the core activity of good paid search management. We’re trying to find the keywords, match types, bids, ad copy, and landing pages that produce optimal results. The only way to figure that out is to test.
But testing in paid search is very difficult, for a number of different reasons.
There are a lot of interacting variables – it’s a complex system.
Critical data is hidden from us – we can’t see what’s happened.
Our tools don’t facilitate testing – which is just crazy.
On top of these we have a statistics problem. Two of them actually.
The math we need isn’t trivial
The data we usually have is sparse and dirty
All of these issues come together every time we look at something in our account – be it a keyword or text-ad or ad group – and based on the data we see decide to make, or not make, some kind of a change.
Of course, we’re just acting on the information we have using the tools at our disposal. But the fact is that we’re making a boat load of assumptions and accepting a lot of averages and approximations. But the biggest risk we’re often taking is in assuming that we’ve got enough data to decide.
Very few of our keywords or text ads – out of the thousands upon thousands in our accounts – get hundreds or even dozens of clicks or conversions. And when there are double or triple digit numbers, the truth is the data isn’t pure at all – it’s a roll up of many different queries and geographies and days and times.
If a keyword converts 4.5% of the time based on 100 clicks that turn out to be from 39 different queries, 76 different cities, 13 times of day, and all on Sunday – does that tell you anything about what’s likely to happen with the next 100 clicks that originate from different queries, cities, times, and days? Maybe. Maybe not.
The risk of turning off a winning text ad, due to not waiting for significant data based on a narrow and known set of variables, is real. In many cases the changes made in paid search accounts, due to all the limitations listed above, are essentially coin flips masquarading as educated guesses pretending to be informed decisions.
And now it turns out we’re even farther from data we can trust that we knew. Even real scientists practicing controlled experiments with what is, by comparison, perfect data, aren’t getting the right answers because even their ‘perfect data’ isn’t enough to tell the truth.
In this amazing article called The Truth Wears Off is worth the time to read fully. In it they explain how scientists are finding that their carefully controlled and ultimately professionally vetted and published results are, over time, proving to be far less certain than they originally appeared.
Here’s one example:
“In the late nineteen-nineties, John Crabbe, a neuroscientist at the Oregon Health and Science University, conducted an experiment that showed how unknowable chance events can skew tests of replicability. He performed a series of experiments on mouse behavior in three different science labs: in Albany, New York; Edmonton, Alberta; and Portland, Oregon. Before he conducted the experiments, he tried to standardize every variable he could think of. The same strains of mice were used in each lab, shipped on the same day from the same supplier. The animals were raised in the same kind of enclosure, with the same brand of sawdust bedding. They had been exposed to the same amount of incandescent light, were living with the same number of littermates, and were fed the exact same type of chow pellets. When the mice were handled, it was with the same kind of surgical glove, and when they were tested it was on the same equipment, at the same time in the morning.
The premise of this test of replicability, of course, is that each of the labs should have generated the same pattern of results. “If any set of experiments should have passed the test, it should have been ours,” Crabbe says. “But that’s not the way it turned out.” In one experiment, Crabbe injected a particular strain of mouse with cocaine. In Portland the mice given the drug moved, on average, six hundred centimetres more than they normally did; in Albany they moved seven hundred and one additional centimetres. But in the Edmonton lab they moved more than five thousand additional centimetres. Similar deviations were observed in a test of anxiety. Furthermore, these inconsistencies didn’t follow any detectable pattern. In Portland one strain of mouse proved most anxious, while in Albany another strain won that distinction.
The disturbing implication of the Crabbe study is that a lot of extraordinary scientific data are nothing but noise. The hyperactivity of those coked-up Edmonton mice wasn’t an interesting new fact—it was a meaningless outlier, a by-product of invisible variables we don’t understand.”
The article, nor the scientific community, doesn’t seem to have a conclusion about why strictly followed scientific procedure is turning out inaccurate or at least inconclusive. The options are a publishing bias, a world that has no stability, or the need to vastly expand the minimum definition of required data for a valid test.
In the meanwhile, be careful cutting bids or or killing keywords based on a few dozen clicks over a couple of days…
Coda
The issues of testing, getting good data, and doing a better job managing the statistics that most of us (and our tools) ignore is one I hope to cover a lot in the coming year. If we want to genuinely optimize our spend and our revenues, the status quo is clearly insufficient.
We’re sponsoring a webinar tomorrow (Tuesday) at 2 p.m. ET / 11 a.m. PT called the “Top 10 Mistakes to Avoid When Selecting Bid Management Sfotware” (hint: one of them is looking for bid management software).
The phrase from the post that got the most attention, aside from the title, was ‘
As several of the 
First you have to get comfortable just standing on skis, and then standing on them while sliding down the hill and riding over all kinds of terain. They you have to figure out what to do with the various parts of your body – ankles, knees, hips, arms, head – while all this sliding is going on. You have to learn to initiate and complete turns, and at some point you start subtly shifting your weight through the process.
But bids actually cannot be reasonably be calculated for keywords that are not haven’t already been properly tuned in terms of organization and negatives and match types and ad copy. The data surrounding these keywords is garbage data – putting that into even a very clever bid algorithm or calculation results in a garbage bid suggestion.
The idea that paid search success is driven by keywords and bids hasn’t been true for many years, but it remains the dominant narrative of our industry. Don’t fall for it.
If a keyword converts 4.5% of the time based on 100 clicks that turn out to be from 39 different queries, 76 different cities, 13 times of day, and all on Sunday – does that tell you anything about what’s likely to happen with the next 100 clicks that originate from different queries, cities, times, and days? Maybe. Maybe not.
Facebook
Twitter
YouTube
Blog
Comment or Question? Leave It Here!