Andrew Frank

When Microsoft acquired aQuantive in 2007 for roughly $6 billion – an 85% premium – ostensibly to counter Google’s then-recent purchase of DoubleClick, some wondered whether competitive instincts were clouding judgment. AQuantive consisted of the interactive agency Razorfish, which Microsoft recently sold to Publicis for about $530 million, DrivePM, a performance ad network that Microsoft has now folded into the Microsoft Media Network, and the Atlas Suite of ad server products, which competes with DoubleClick’s DART system and was the principle motivation for the purchase.

In 2008 Microsoft doubled down on ad management tools with its purchase of Rapt, the leading pricing and yield management solution for publishers, which it quickly integrated into the Atlas Publishing Suite. Now sources tell me that Microsoft is no longer taking new customers for Rapt, and its biggest customer, Yahoo!, recently informed analysts that it will be discontinuing its use of the product in favor of internally-developed solutions. Meanwhile, ad operations providers such as Operative, Solbright, and FatTail have closed ranks on pricing and yield management in their ad product suites.

A few weeks ago, Microsoft announced a partnership with open source ad server provider OpenX, an Atlas competitor. The announcement describes an arrangement where OpenX provides a “distribution channel for Microsoft monetization products” (read: ads) and, in return, gains “access to a new base of potential customers – via referrals from Microsoft – for its enterprise advertising technology and services.” Although Microsoft’s director of advertising business development Peter McDonald, commented obliquely that “the OpenX ad server technology might be more appropriate [than Atlas] for certain types of Web publishers,” the partnership nonetheless suggests a migration path for Atlas users should Microsoft decide to exit the ad server business.

Another clue lies in the way Microsoft has evolved its description of PubCenter. In April, Microsoft described PubCenter as “the convergence into a single platform of the many systems we’ve developed and acquired…[including] technologies and tools provided by the former Atlas and Rapt solutions, as well as a self-serve offering.” Today, the PubCenter (beta) site offers “select web site owners the chance to place advertising from the Microsoft network on their web sites.” No word of convergence.

I’ve frequently advised publishers, vendors, and advertisers to be wary of the conflicts created by companies that offer both media campaign management tools and media under the same roof, so Microsoft’s decision to favor its media business over its ad server business might be a wise one. Still, $6 billion was a lot to pay for DoubleClick’s chief rival. And Google, for its part, seems to be well on the way to making its largest acquisition pay dividends.

This has been a month full of interactive advertising conferences in NYC, from OMMA AdNets, through Ad:Tech, and The Media and Money Conference, to the IAB Ad Operations Summit, to name a few. So many themes have been covered it’s hard to know where to start, but there’s one that stands out. In the ongoing quest for answers to the question of why it’s taking so long for brand advertisers to open their media war-chests to the Internet, a leading contender has emerged: it’s just not a safe environment for brands.  A recent rise in online advertising exploits provided an edgy backdrop to many event panels as they grappled with cautious optimism for a recovery.

The issue came to light  in October when a plague of phony insertion orders compromised major publisher sites including The New York Times, Foxnews.com, The Huffington Post, and Gawker, culminating in Starcom MediaVest Group’s request for a Federal investigation (see MediaPost or AdAge coverage). Fraudulent practices range from malware distribution, which appears to be on the rise (see Click Forensics’ alarming report, declaring that click fraud perpetrated by botnets, a result of malware distribution, has risen sharply) to the grey-hat technique of the month, “invisible advertising.” (Is it fraud? You decide. But it certainly isn’t the kind of thing that’s going to ease any brand-conscious minds.)

On the defensive side, AdSafe Media has set itself up to provide rating and filtering services for advertisers, which could help solve the website half of the problem (MediaPost), while sell-side optimizers such as The Rubicon Project, PubMatic, and AdMeld have been promoting their abilities to help publishers filter the advertising side. (PubMatic, incidentally, held an Ad Revenue conference of its own on October 8th which I attended and found quite good; see recap here.)

The bottom line is, major brands are going to continue to be skittish until these incidents calm down, but the incumbent leaders on the publishing and media agency sides should smell an opportunity here. Their common adversary, comprised of certain ad networks that are widely seen as depressing prices and arbitraging profits out of the system, is arguably also contributing to a general climate of low security by removing personal contact and active scrutiny from the marketplace. But the fact that premium players have also recently been successfully targeted suggests that they need to do more to distinguish themselves as safe – and thus worthy of premium pricing and greater spending allocations.

Publishers and agencies have a chance to take the upper hand on this issue, but they’ll have to move quickly. They need solid solutions of their own before someone like Google takes the reigns.

Apple’s recent patent application to incorporate advertising into an OS as a potential subsidy is creating quite a stir:

• “Apple Prepares to Rock the Market with Hardware Subsidizing Program”
• “Apple applies for patent on OS with embedded advertising”
• “Apple Patents Ad-Supported, Subsidized Mac OS”
• “Gimme a minute — just two more ads before it lets me use the Finder again”

As with the Google music rumors, or the latest round of Apple tablet speculation, there seems to be more smoke here than fire: Apple files lots of patent applications, and, despite some loose reporting, this is just an application, there’s no guarantee it will be granted, and given that the innovation appears to be the use of advertising to subsidize an OS (which, at the end of the day, is just software), one might expect examiners to scrutinize it carefully, and look closely at some of the prior art in Zune, among others.

What’s interesting about this is the indication that Apple is thinking hard about advertising revenue, which is something we’ve been discussing on the media team for some time. On the one hand, as the rise of Android puts Apple and Google in more competitive situations, the power of Google’s ad-based revenue engine to subsidize innovation can’t be lost on Apple. On the other hand, there’s nothing Apple seems to hold more sacrosanct than its branded user experience, and suggesting that this could be sold to advertisers will clearly strike some as sacrilege. Apple may imagine it can leverage its success corralling app developers through an approval process with advertisers as well, but the difference is, for advertising to work economically, it needs to operate at much higher reach and frequency than app development.

Bottom line: Apple’s likely to experiment with ads, and there are many ways for them to do so – especially if they open up a publishing channel in iTunes to deliver print content to an Apple eReader… but an ad-supported OS carrying the Apple brand? With ads appearing at predefined times, blocking activities? Running a preroll after that iconic startup chime? Hard to imagine.

Still, if the patent issues, perhaps there’s some upside in licensing the IP to Android….

I want to thank all of the social media monitoring and analytics vendors who responded to last week’s challenge.

First, the highlights.

This year’s challenge listed 61 company names, of which 15 responded, a nearly 25% response rate.  An additional 4 who were not mentioned also responded. Last year, 34 company names were listed, of which 23 responded, a 68% response rate. An additional 13 who were not mentioned responded last year. Of 36 who responded last year, 17 didn’t respond this year, even though they were all on the list.

Whatever the reason, some of the companies that didn’t respond remain on my list of top-tier players based on the work they’re doing for clients, which of course is the ultimate test.

Second, an observation. Even among those who responded quickly, I was surprised at how many neglected to take the step of mentioning their company’s name in the response. I’d suggest it’s wise to always keep SEO in mind and make it second nature when posting anything on behalf of your company’s brand. Search engines still matter, shankapotamus.

Contextual Bonus Awards (given to companies who were not on the list but responded anyway):

ImpactWatch (also noted for taking the opportunity to plug that they also monitor offline media)

Media Monitors (also noted for taking the opportunity to plug a newly released product: Mediaportal.com)

Waggener Edstrom (also noted for taking the opportunity to plug a newly released product: twendz)

VMS Info

Rapid Response Bonus Awards (given to companies who responded within fifteen minutes of the post)

InsideView  (special note as the respondent appears to be a client rather than a representative of the company)

Filtrbox (New to the list)

Responsiveness Awards (please note that some of these companies have taken pains to point out that they are not “watchdogs” or monitors in any limiting sense, but focus more broadly on strategic inferences from broad range of consumer conversations. However, their mention here at least attests to the fact that they are listening.)

BurrellesLuce

Cision

Converseon

filtrbox

InsideView

The McGraw-Hill Companies J.D. Power and Associates Web

MotiveQuest

Onalytica

Radian6 Technologies

Sentiment Metrics

Spiral16

Sports Media Challenge

Sysomos

Alterion Techrigy

Visible Technologies

Thanks again to all who participated and please don’t hesitate to let flow the feedback.

Final note: as MotiveQuest’s Tom O’Brien noted, Microsoft Advertising was in the news last week at Advertising Week in NYC with a social media monitoring product called LookingGlass (AdAge, ClickZ), which we’ll be evaluating in more detail shortly (though not necessarily on the blog).

Such a move clearly suggests there’s still play in the social media monitoring space, even as the table stakes go higher. Microsoft’s entry may tend to commoditize the brand monitoring function and shift value focus to either broader forms of analysis, characterized by more strategic service-oriented approaches. Will the incumbent agency space dominate these services, or can new approaches scale better working outside the agency legacy? Stay tuned…

About a year ago I issued a challenge to the social media monitoring and analysis vendor community to determine which of these vendors “ate their own dogfood.” In other words, who was able, using their own product, to spot mention of their own brand names in the blogosphere and rapidly reply. The results were impressive: 36 companies responded, mostly within 24 hours (and many within seconds), even though only 34 were originally mentioned. 13 of the respondents weren’t on the original list.

Much has happened in a year, but one thing that hasn’t happened is a thinning of the field through consolidation or collapse. While some companies have merged or shifted focus, and a clear leadership tier has emerged, the overall number in my database has nearly doubled to sixty-one, making for a pretty noisy category. The good news is demand seems to still be growing as well, so, while consolidation’s to be expected, there’s yet a prize in store for the best vendors, and plenty of choice and competition for users who are still waking up to the critical importance of analyzing social media, or refining their approach based on some experience.

The goal here is simple: to determine which of these companies can respond with a comment within 48 hours (when the results will be posted) of having their name appear on this blog.

Disclaimer: this is not a true product test, nor is it meant to imply that a company’s ability to detect its own brand is necessarily a significant factor in choosing the right vendor. It’s just a way to showcase the power of this dynamic category, and to discover if this list is missing anyone important. Some may find it a useful guide as links to the websites (which are not always obvious) are included.

It’s also a way to foreshadow that we’ll be publishing more research on this category soon.

I look forward to hearing from you.

1st2c
Aberdeen Group
Attensity
Attentio
backtype
Biz360
BrandIntel
Brandseye
Brandtology
Brandwatch
BurrellesLuce
Buzzgain
BuzzLogic
Nielsen Buzzmetrics
Channel Signal
CIC
Cision
Clarabridge
Collective Intellect
Converseon
Crawdad Technologies
Crimson Hexagon
CSC NameProtect
CustomScoop
Cyberalert
TNS Media Intelligence Cymfony
dna13
eCairn
Echo Research
Envisional
News Corp Dow Jones Factiva
filtrbox
Gnip
Imente
InsideView
The McGraw-Hill Companies J.D. Power and Associates Web Intelligence
Jive Software
Kaava
Lexalytics
Leximancer
Market Sentinel
Moreover
MotiveQuest
Networked Insights
New Media Strategies
Onalytica
PopularMedia
Radian6 Technologies
RelevantNoise
RepuMetrix
ScoutLabs
Sentiment Metrics
SentiMetrix
Spiral16
Sports Media Challenge
Sysomos
Alterion Techrigy
Trackur
Unbound Technologies
VibeMetrix
Visible Technologies

FCC Chairman Julius Genachowski’s announcement of the Commission’s intention to formalize and implement net neutrality regulations has elicited all of the predictable responses, from free-market political and telecom industry opposition to applause from advocacy groups and companies like Skype. (The cable industry alone straddled the fence and offered lukewarm support while warning against excessive regulation: for example, see Comcast’s blog post.)

The FCC also announced the “beta” launch of OpenInternet.gov, a site that will track the progress of all this.

In 2004, the former FCC chairman Michael Powell established four principles that formed the basis of net neutrality policy. Now, Genachowski has added two more:

The Principle of Non-Discrimination

The fifth principle is one of non-discrimination — stating that broadband providers cannot discriminate against particular Internet content or applications.

The Principle of Transparency

The sixth principle is a transparency principle — stating that providers of broadband Internet access must be transparent about their network management practices.

Now, the Principle of Non-Discrimination has been a hot-button issue in the debate, but it’s important to note that it doesn’t prohibit carriers from selling faster internet connections for more money; it just states that they can’t discriminate on the basis of content or applications. It also allows exceptions for "reasonable network management," such as throttling Internet traffic under times of heavy usage or to prevent spam. Its intent is to counter the possibility of carriers from favoring their own services on their networks, by, for example, making (Comcast’s) Fancast work better than Hulu.

But it’s the Principle of Transparency that puts teeth in all this. In fact, the Principle of Transparency, combined with low enough switching costs and enough competition, might suffice on its own to align market forces with the intentions of the remaining regulations.

The Principle of Transparency is hard to oppose because it should generally apply to every industry, not just the Internet. And, as has been frequently demonstrated, when combined with the Internet it is far more expedient at exposing anti-consumer activities than any government agency could hope to be. That’s one of the great promises of the Internet.

Many web video platforms and content providers are already monitoring stream quality, but have little recourse but to optimize for lower bandwidth when bitrates drop inexplicably for their viewers. Transparency would empower them to get an explanation from the service providers involved, which in turn would increase competition and quality, while building confidence in the channel’s overall reliability.

If the telecom industry is to take up transparency at all, it might be to point out the double-standard of applying it to communication but not to content services, which are no less part of the public’s interest in the Internet. Why shouldn’t a company like Google be similarly required to reveal how it manages its search services? Similar issues of potential hidden discrimination are involved. Perhaps net neutrality is not just for service providers.

For a long time, web publishers and advertisers have been caught in the crossfire between web analytics and media metrics vendors producing starkly different numbers for their sites. Now, by joining forces, ComScore and Omniture have potentially eliminated this maddening issue…at least for their common customers. (To be precise, Omniture customers can opt-in to see benefits of the partnership without being Comscore customers, but the strongest benefits are likely to accrue to customers of both.) By fusing methodologies through a common SiteCatalyst tag (no retagging necessary), they’ve also dramatically changed the web measurement landscape, and thrown their competitors collectively off balance.

The issue is the long-standing disparity between measurement methodologies, which continues to produce gross discrepancies between the traffic measurements supplied by web analytics services (like Omniture) and the panel-based measurements typically supplied by media metrics providers (like ComScore). Since web analytics uses tracking pixels and server logs to generate traffic data, issues like accessing sites from multiple devices or locations, or cookie deletion, or spider misidentification, will tend to overcount the number of unique visitors to a site, while issues like panel size or universe limitations (i.e., at-work vs. at-home vs. mobile usage) will tend to undercount uniques. As a result, while the two camps bickered, web publishers and advertisers suffered from a lack of unified measurement that cost them  time, money, and credibility, belying the oft-repeated cannon that the web is our most measurable medium.

Since a unified measurement is so beneficial, the competitors of Omniture (such as Coremetrics or Webtrends) and ComScore Media Metrix (such as Nielsen/NetRatings or TNS Compete), now seem to have two choices:

1. They can join the dance and announce their own competitive alliances (potentially seeding major market share), or
2. They can finally get serious about addressing the remaining issue for customers, which is the lack of standards that continues to hamper interoperability, create switching barriers, and limit the scope of these types of alliances.

In a world governed by open standards, a web analytics customer would be able to opt-in to provision its server data to any or all panel-based metrics vendors for refinement, targeting, and advertiser arbitration, and we wouldn’t need alliances. In the absence of such standards, ComScore and Omniture have done the right thing by burying the hatchet and solving an acute problem for their customers, who are bound to be thankful. Now that they have, perhaps the new landscape will be more conducive to accelerating the slow pace of standards development.

One problem seems to be the dueling roles of the IAB (on the metrics side) and the Web Analytics Association (on the web analytics side)…perhaps they can follow Media Metrix Hybrid 360’s example and form a joint panel to resolve this. In the mean time, the publishers and advertisers should weigh in with their vendors on which course they’d prefer to see them take.

Google’s long-awaited relaunch of DoubleClick’s Ad Exchange (see Official Google blog) marks its most visible attempt to-date to leverage its DoubleClick acquisition to take it out of its lucrative search niche and into the sketchy world of display. (More on the DoubleClick blog, The New York Times and paidContent.)

Launched just before the Google acquisition, DoubleClick’s Ad Exchange struggled to gain share in a market characterized by strong network effects (where scale of liquidity matters most) and dominated by Yahoo’s Right Media Exchange. DoubleClick’s secret weapon was its integration of the Exchange with DART for Publishers (DFP), which enabled ad sellers to automatically offer ads on the exchange that would only be pre-empted if a buyer on the exchange could beat the currently booked price, a no-risk proposition for publishers known as “dynamic allocation.”

As it turned out, this addressed the wrong issue. The display market had (and has) no shortage of supply of low-cost advertising inventory; its problems are on the demand side, and DoubleClick had much more trouble attracting ad buyers than sellers to its exchange. Google’s approach to this problem is apparently to leverage its huge customer base of search ads buyers by making it easy for them to migrate to display. Eric Schmidt has indicated on numerous occasions that he sees display as the next growth phase for Google, and has acknowledged that search must soon plateau in its growth. So the stakes are high as Google seeks to demonstrate to the market that it’s not a “one trick pony.”

There are two related challenges in this strategy. The first is the question of whether the bulk of Google’s AdWords customers – mostly SMBs for whom the channel is more about sales than advertising – will embrace the more indirect, brand-oriented medium of display, which requires, at minimum, some graphical execution. To address this, Google has launched a Display Ad Builder “to help businesses easily set up and run display ad campaigns.” Google goes on to say, “80% of advertisers who use that product have never run a display ad campaign before,” a claim that withers on double-take.

The track record of do-it-yourself ad creative tools is not encouraging. I recently spoke with Anupam Gupta, president and CEO of Mixpo, which had attempted a DIY video model for local merchants and marketers and had concluded unequivocally that bringing agency-produced local spot ads online was a much more promising model. Of course, Google is not putting all of its eggs in the DIY display basket, they’re also wooing ad networks and agency buyers who are more likely to invest in display media and creative services. For example, AdAge quotes Curt Hecht, president of the VivaKi Nerve Center, part of Publicis and a customer of Google’s platform: "Our view is display looks more and more like search every day. AdExchange makes it easier to buy without making a lot of calls to publishers or working with a lot of ad networks."

Which brings us to the second problem: agency buyers and ad networks, which often find themselves in “co-opetition,” are highly motivated and empowered to put considerable price pressure on additional intermediaries, especially if their name is Google. In its AdSense search business, Google has the advantage of opaque pricing and weak competition; in display exchanges, transparency reigns, and Google has plenty of competition. Agency buyers and ad networks, already feeling squeezed, have in common buying power and experience, which will drive exchange margins to a minimum. On the other side, publishers, especially those who consider themselves “premium,” are feeling even more squeezed, and are talking with agencies about private exchanges for their own networks that could minimize price friction in the value chain. Such developments will drive Google out toward the long tails of sites and advertisers, where display tends to lack the performance qualities of search.

Google is well aware of these challenges, and many have lost betting against Google. But the balancing act of pricing, advertiser support, and alliances among networks, agencies, and publishers in low-trust climate, with Microsoft and Yahoo! as competitors, will be very tough to pull off.

“I warn you against believing that advertising is science.” So said Bill Bernbach, one of advertising’s great legends who holds the number one rank in Advertising Age’s honor roll of most influential people. So what would Bill Bernbach think of Adobe’s idea, as expressed in its $1.8B acquisition of Omniture, that analytics (e.g., SiteCatalyst) can and should play a deeper role in the design process (e.g., Flash)? As the patron saint of creatives and the champion of the “big idea” model of advertising, he’d probably despise it.

“Logic and over-analysis can immobilize and sterilize an idea,” Bernbach said, “It’s like love – the more you analyze it, the faster it disappears.” Such notions seem quaint today, in the age of web analysis, as data and analytics guide marketing and publishing budgets and increasingly become the currency of media. While some lament that digital over-analysis is indeed taking a toll on the creative quality of commercial communications, creative developers continue to take solace in the fact that they still hold sway in the design of Flash experiences, whose effects will be evaluated later – for now. And many have dutifully been instrumenting their designs with tracking pixels for years to make them more transparent – with mixed results.

But what happens when they’re introduced to their new cube-mate, an analyst trained in Omniture products who will look over their shoulder and suggest where to put the data collection tags and how to incorporate design elements aimed at real-time optimization? Such a collaboration might produce breakthrough results. Or it might not.

Bernbach is also credited with introducing the idea of the creative team consisting of a copywriter and an art director, which transformed how advertising was designed. Before Bernbach, we’re told, advertising concepts were thought up by copywriters who handed them off to art directors to illustrate. The team concept energized and accelerated the creative process. (Developers may wish to draw analogy to the “pair programming” technique associated with agile methodologies.) Could teaming up creative developers with analysts using a common Flash platform produce similar benefits? Well…

Creative developers tend to be holistic, “right-brain” thinkers who may naturally resist decomposing concepts into measurable units. Analysts tend to be reductionists, inclined to embrace such decomposition and measurement. Top-down organizational initiatives are unlikely to address the issues impeding their successful collaboration.

So, while most commercial web development interests appreciate the need for increased transparency and accountability in their efforts, the question of how best to integrate analytics with web design as a practical matter remains thorny, and the impact of Adobe’s integration efforts will depend not just on Adobe’s execution, but on industry-wide attempts to evolve the web development process in line with Adobe’s vision.

On July 29, 2009, Yahoo! and Microsoft announced that “Yahoo! will become the exclusive worldwide relationship sales force for both companies’ premium search advertisers.” They also announced that “Self-serve advertising for both companies will be fulfilled by Microsoft’s AdCenter platform, and prices for all search ads will continue to be set by AdCenter’s automated auction process” This led me, and others, to buck the tide of negativity around the deal, and Yahoo!’s role in particular, and suggest that this division of markets was strategic for both companies, and Yahoo!’s control of premium search advertising on Bing might just turn out be to the bigger windfall. More on this in a second.

On August 10, 2009, Publicis Groupe SA announced that it had agreed to acquire Razorfish from Microsoft for about $530 million in cash and stock.

(Razorfish, in case you haven’t been following, is a top interactive agency that Microsoft picked up as part of its acquisition of aQuantive in 2007 that gave it the Atlas family of ad servers and an online ad network. Razorfish is doing some of the best work in the business and they will be an asset to Publicis’ VivaKi “digital holding company” which includes Digitas and Performics (recently acquired from Google) and many other leading agencies, putting Publicis at the front of the pack of ad holding companies (WPP, Omnicom, Interpublic, Havas, and Dentsu) racing to incorporate digital services as a centerpiece of their offerings.)

But also included in the deal, according to Publicis CEO Maurice Lévy, was “a five-year media-buying relationship. In return for buying a certain amount of display and search advertising on Microsoft properties, Publicis will receive better ad rates.” (see WSJ coverage).

What’s wrong with this picture?

Razorfish clients include Ford, Best Buy, McDonald’s, and Microsoft, and Publicis as a whole represents many more “premium” brands. Clearly, Microsoft can’t sell search advertising to these advertisers and still claim that Yahoo! is “exclusively” empowered to sell to this market without some clarification of terms. Microsoft declined requests for such clarification, but a Microsoft spokesperson was quoted in Forbes as saying, “the Yahoo! and Microsoft proposal had not yet been approved as an official partnership, and that he was ‘fairly certain’ that no deal would have been struck with Publicis if it ran counter to the Yahoo! partnership.”

Retaining the ability to sell premium search is, in my view, essential for Yahoo!’s future, for two reasons.

First, as Yahoo! and Google have both repeated frequently, the future of online advertising is “Search-Display Convergence,” which means both that search data will become an increasingly essential component of targeting display ads, and search ads themselves will come to resemble display ads in graphical and interactive richness. “Premium” advertisers, in particular, will want integrated campaigns that combine the targeting of search with the branding capabilities of display (and video, which is increasingly being incorporated into display).

Second, the next wave of growth in online advertising will have to be on the premium side. The new market that Google discovered of SMB advertisers buying keywords on self-service auction sites has shown phenomenal growth, but by now most of the customers who could benefit from this form of marketing are already involved. From here, it will be a technology-driven battle for market share which Microsoft is eager to wage (and Google, with Caffeine, appears equally up for).

On the other hand, among premium advertisers, the shift of media spending from traditional channels to online is still at an early stage, especially among CPG manufacturers which continue to spend many times more on TV and print but are compelled to follow consumers who have shifted their time to online activities far faster than the they’ve shifted their budgets. These buyers are looking for a partner on the media side that can provide massive reach and handle integrated, rich campaigns that are free from the perils they see in social media and user-generated content and sketchy long-tail pages that result from buying traffic blindly on networks. That need defines a potentially lucrative role for Yahoo!.

But not if their partner Microsoft is competing with them for this market by offering bulk discounts to major media buyers. Which is why I think the Microsoft-Yahoo alliance plans just got more complicated.