|ASA: What are theTop 5 Publisher Requests?|
| 10:37 pm on Nov 19, 2008 (gmt 0)|
From another thread here: [webmasterworld.com...]
|I'll make you a deal: if you all, the members of the AdSense WebmasterWorld forum, can agree on your top 5 asks in order of priority, I will personally email the Director of Product Management for all AdSense products with your list. |
So lets get it started.
| 12:40 am on Dec 6, 2008 (gmt 0)|
|I’m talking about the stuff you’re saying with the people who you would want to hear it. That’s the best I can do. |
And these are the same people who have been in place for months, years? This dialog will become more frank and earnest or we will all disappear. We are going to see a lot of 'creative destruction' this coming year as marginal business evaporates, the pie shrinks and our ranks thin. So, I'm going to put a little more effort into communicating on this forum, at least for now.
Having bailed out of GOOG at $600 seems just a distant sweet memory, but now I still can't afford to get back in. <sigh>
| 3:39 am on Dec 6, 2008 (gmt 0)|
|We know that “more channels” is a request that we hear a lot. The people who build the product know that it’s a request we hear a lot. What would be useful to know is this: would you rather have more channels than more custom filtering? Would you rather have more channels than Google Analytics integration? |
I'm curious why you're asking this - does increasing the number of channels per account take so much effort that other things will be pushed back for a significant period of time? Seriously, it almost sounds like a "you can only have one" statement even if that's not how you meant it.
As for a direct answer, I think you'll find some people who want the channels more and some who want the filters more. So I guess AdSense is going to have to decide which is best to work on first if these are mutually exclusive. (It seems to me that more channels would be much easier for you guys to implement that a filtering system, but what do I know?)
Personally, not knowing exactly what custom filtering wound entail, I pick more channels first, then custom filtering and Google Analytics somewhere behind in the distance.
| 5:04 am on Dec 6, 2008 (gmt 0)|
Thank you AdSenseAdvisor for your post.
It was nice to read that this was not a corporate response but someone who is now taking an active interest with the longest post any ASA has made.
I would like to make things easier for you and the forum.
You asked for the top 5 suggestions. To all two are very obvious and they are more channels and minimum EPC. The question WE need to know is how easy would this be for Google to implement?
If adding more channels is a problem...say so...however I have been offered these and I do NOT need them! Give them to someone else who's screaming for them.
Why give them to me when I only track via URL and not by page etc? Surely your algos understand that? If not, tell me how much you're prepared to pay me to sort that out:)
AdSenseAdvisor, you've been thrown into the new ex-BBBG world (Blair/Brown/Bush/Greenspan) with the remit of trying to pick up the pieces and ensuring the future of Google.
What I would like to see are G ideas BEFORE they are trialled without ANY notice! I bet there are some discarded ideas that would work but have never seen the light of day.
If, IF!...I ever ignored my customers' requests/demands/neccessities then I would be out of business, and for 170 years we have listened:-)
| 6:19 am on Dec 6, 2008 (gmt 0)|
|We know that "more channels" is a request that we hear a lot. The people who build the product know that it's a request we hear a lot. What would be useful to know is this: would you rather have more channels than more custom filtering? Would you rather have more channels than Google Analytics integration? Would you rather have more channels than any of the other things you want? |
Disclaimer: This is not meant as offensive post. But you keep asking questions that do not make sense to anyone who has a true marketing background.
Don't you guys have product managers?
Don't you guys have market researchers?
Don't you guys have focus groups?
Don't you guys know how to manage your products?
Well, apparently not. Your product teams apparently have not lurked in this forum for new ideas and comments on the existing program for years. They did not listen to the previous ASAs. Or they have done this and now can not prioritize. Ouch.
But fear not! Here's a handy product marketing crash course for glueless cooglers. (Please feel free to print this message and pass it to your CxO level. After all, this is a genuine management problem, nothing else.)
You should acknowledge that it can not be the =responsibility= of this board to come up with a final feature list, or the prioritization of feature requests mentioned here. We have discussed this at lengths here. The reason is simple: we are not Google's product managers. We can not comment on so many things (Strategy? Goals? Short-term vs. long-term? Technology?), because we don't know and often we don't care. We care about our own sites and users and business models and revenues.
And Google, btw, wouldn't want to have it that way either, because there is a huge risk of getting biased input from a group that is just very vocal. Which you want to avoid at all cost.
Here's how I would do it.
1. Compile a list of the feature request mentioned here (and elsewhere), and make a note on the total number of requests for each feature. It's a long list. The more the merrier. Do not forget to include the ideas your team comes up with.
2. For each feature: think about business implications (Is it in line with corporate strategy? What would be the cost to the company? What would be the benefit? When could it be launched?). Inject this into your business plan. You know your priorities, so assign ratings for each area to each feature (e.g. cost to make, profits gained, risk of losing customers, etc. etc.).
3. Throw out all ideas that are totally off, and concentrate on the top-20/top-30 feature requests. THIS IS WHERE ONLY YOU GUYS CAN MAKE THE DECISION - ONLY YOU KNOW YOUR PRIORITIES. WE CAN'T HELP YOU HERE.
4. Now think about useful descriptions and possible implementations for these features. Add screen mockups or even alpha product screens where possible. Goal is to discuss such features with a wider audience (under NDA) - if all this takes too long, why not ask here and elsewhere how features might be implemented? You will find a lot people who would love to work with you on this.
5. Take all this into focus groups (again, under NDA) and hear what selected people from the industry will tell you about this, in a controlled environment. Ensure to catch opinions of your target groups, not just "everyone" or "those who are loudest". Listen to different markets, different publisher types, different users - but most importantly - to those people you TARGET. (Hint: "All" is not a very good target.)
6. Re-rank and throw out those ideas that are totally off, and concentrate on the top-10 feature requests (or how many features you actually CAN implement given your limited resources).
7. Implement the features and roll them out to a closed beta group and see what happens. Finetune until your closed user group is really happy with the product.
8. Then, and only then, release the product to the public. Do not forget to make as many features as possible "optional", so that even more publishers are happy with your product.
9. Then go back to step 1 for the next release. (Note: It's a good idea to re-use existing feature lists. No point in collecting the same stuff all over again.)
But I wonder - this is all basic product marketing know-how, really. Marketing 101 if you want. It has been practiced in countless companies, real world or virtual, around the globe for ages. No rocket science here.
How comes that Google wakes up only now? What is going on in MV? Is it really that you can only throw unfinished products at people and see what happens? Don't you have product managers that actually know how to manage a product?
Should this be really true, investors should be scared. Very scared.
| 7:50 am on Dec 6, 2008 (gmt 0)|
|Don't you guys have market researchers? |
Don't you guys have focus groups?
Actually, they do.
| 8:43 am on Dec 6, 2008 (gmt 0)|
Actually, I know they do. It was a rhetoric question. I know they have product managers, too. Well, maybe they have people who call themselves "product managers". ;-)
Google apparently just don't know how to utilize the results from their market research and their focus groups. Otherwise they would not seek help here in a way that unveils their helplessness. Read this again:
|The people who build the product know that it's a request we hear a lot. What would be useful to know is this: would you rather have more channels than more custom filtering? Would you rather have more channels than Google Analytics integration? Would you rather have more channels than any of the other things you want? |
"The people who build the product" don't have a clue where to begin. They don't know how to prioritize. Probably due to a lack of guidance from their management. These questions are a CLEAR sign for the lack of product management skills. THEY DON'T KNOW. THEY ARE CLUELESS.
The more I think about this, the more I come to the conclusion that Google's success has been built upon pure luck. They were lucky to come out with a good concept (search) at the right time, and they built their company around this concept and monetized it using a way that worked (text ads). They became leaders in the market because others were totally, absolutely clueless at that time. Google's success was NOT the razor-sharp execution of a plan. Not at all.
Today they still neither have a plan nor the skills to actively manage their products. I don't know what they are spending their money on, but certainly it's not on hiring top-notch product managers that clean up the rubble. Or their product managers don't run the show (but the engineers are! Scary!). It must be quite frustrating to work in Google product management, realizing that the company has zero plan at all. They don't know how to use the tools they have (market research, focus groups), and they are afraid of/incapable of actively managing their products. Hence the fear of failure ("if we do this, then it might happen that [insert major fear here]... Uhhh. No we can't do that."). Hence the cry for help here. Hence the countless product launches that do not make sense. They are desperately poking in the haystack for another needle.
I guess Dilbert works in MV. (Investors: be VERY afraid!)
| 12:44 pm on Dec 6, 2008 (gmt 0)|
If your AdSense/AdWords developers are busy so much that they cannot implement e.g. ad filtering by keyword within a week, then... employ or hire 5 more developers? Or 50. If you need more data centers, buy them, build them, or rent them. You aren't a poor company, or are you?
If I owned the most successful ad network on the internet, I'd invest a lot more money in it than Google seems to do (especially if it was practically my only source of income).
[edited by: true_INFP at 12:58 pm (utc) on Dec. 6, 2008]
| 1:25 pm on Dec 6, 2008 (gmt 0)|
Extra channels is a matter of infrastructure.
Filtering options more of a matter of impact on network income.
The only place where they could intersect at at the research team in the beginning, neither is labor intensive to develop. If it's a matter of priorities my choice is the filer.
My advice is to get the research team working on both immediately and come back to us with a quick verdict so we can move on to discussing many other important features that can impact our earnings and Google's.
Doing good so far ASA, just make sure all those discussions materialize into something solid, anything, within 2008 before you guys take off for Christmas.
| 3:53 pm on Dec 6, 2008 (gmt 0)|
It's possible that "What are the top 5 publisher requests?" is the wrong question. A better question might be "What are the top 5,000 publishers doing that the bottom 5,000 publishers aren't (and vice versa)?" Or, to put it another way, "What differences in behavior can be seen in the two groups?"
- Is there any difference in the use of channels by successful and unsuccessful publishers? (This could have an impact on whether an increase in the number of channels is desirable.)
- Is there any difference in the use of the competitive ad filter by the two groups? (This could have an impact on whether an unlimited filter is desirable.)
Of course, it's possible that Google has already done this kind of study, and that former ASAs weren't speaking off the top of their heads when, for example, they warned that using the competitive ad filter too agressively or that opting out of CPM ads could result in lower earnings.
In any case, identifying the differences in "AdSense behavior" between successful and unsuccessful publishers might be a useful exercise, because the existence of tools and features may be less significant than how existing tools and features are being used.
| 3:55 pm on Dec 6, 2008 (gmt 0)|
Whooops! This is what happens when you put a little peep-hole in the black box. Before I get too cynical I'd better just agree with Hobbs and encourage ASA to work hard on these suggestions.
I find it particularly amusing that signor_john took zett's rhetorical questions as actual queries. zett, IMHO has done a good analysis considering what he has to work with. If our dialog here with ASA is any real indication of what's going on there I think there does need to be some serious shaking up in MV before the economic earthquake hits. And it will. My AdSense revenue is holding up well because advertisers are tossing everything they have at saving Christmas. A lot of advertisers are going to go belly-up in January, Ch. 11 and Ch 7 and there's going to be a monster claw back.
Think about that over Chistmas dinner.
| 4:26 pm on Dec 6, 2008 (gmt 0)|
|If it's a matter of priorities my choice is the filer. |
I would agree with that except for one thing.
More channels is more channels - it's pretty straightforward. About the only way that could be messed up is if they increased the number, but by a ridiculously low number - like increasing it from 200 to 220 for example.
Filtering is another thing. We've all experienced new features that seemed to have a poor implementation and flopped (CPA for example). Who knows what we would get with filtering?
So the way I look at it, give us the channels, that seems to be something solid and relatively foolproof. And it seems very possible as some people have already reported being offered an increase to 500 channels.
Then give us the filtering. At least if the filtering is a flop, we'll at least have the additional channels to work with.
| 4:31 pm on Dec 6, 2008 (gmt 0)|
Ah, and just in case my vote is counted somehow, somewhere, by someone:
My choice is a better filter. Give us full control (fool control?) over the ads that show up.
| 4:42 pm on Dec 6, 2008 (gmt 0)|
|Of course, it's possible that Google has already done this kind of study, and that former ASAs weren't speaking off the top of their heads when, for example, they warned that using the competitive ad filter too agressively or that opting out of CPM ads could result in lower earnings. |
That could be true.
On the other hand, I recall those monthly optimization tips, repeated month after month, warning me that using the competitive ad filter too agressively might be suppressing my earnings, even though my filter was empty for months. That sort of thing makes it easy to be cautious about some of the adviced dished out.
Couple that with the fact that prior ASA's didn't seem to be very high on the totem pole and it makes you wonder if those prior ASA's were telling us things based on knowledge or just repeating something they heard in a meeting.
And then there are those publishers who have decided that even if a study showed having a near-empty filter resulted in more AdSense income in the short-term, the undesirable ads would harm their site's reputation to the point it would lower overall income in the long-run.
| 5:02 pm on Dec 6, 2008 (gmt 0)|
|On the other hand, I recall those monthly optimization tips, repeated month after month, warning me that using the competitive ad filter too agressively might be suppressing my earnings, even though my filter was empty for months. That sort of thing makes it easy to be cautious about some of the adviced dished out. |
The optimization tips were boilerplate. They weren't personalized. But I suspect most of us knew that. :-)
|And then there are those publishers who have decided that even if a study showed having a near-empty filter resulted in more AdSense income in the short-term, the undesirable ads would harm their site's reputation to the point it would lower overall income in the long-run. |
Or maybe some of those publishers haven't been willing to admit that AdSense isn't a good fit for their sites. If a site is getting mostly junk ads because there aren't enough good advertisers in the niche (or because advertisers may have blocked the publisher's domain), playing whack-a-mole with unwanted ads isn't going to solve the problem.
New AdSense features aren't likely to be a panacea for having a hard-to-monetize topic or audience.
| 5:28 pm on Dec 6, 2008 (gmt 0)|
OK, I neither have a full filter nor use all my allotted channels.
I was hoping to get those out of the way so we can move on to other juicy stuff, the kind that can boost earnings & productivity.
signor_john, please let the requests be, and have yours along with the AdWords team tips he started on in a separate thread.
| 5:36 pm on Dec 6, 2008 (gmt 0)|
|...I suspect most of us knew that... |
I don't know who the "us" club is that you speak for, but I haven't seen anything to convince me that comments by prior ASA's was anything other than boilerplate language.
|...Or maybe some of those publishers haven't been willing to admit... |
Must you attempt to belittle everyone who doesn't agree with your opinion? I didn't say anything about "mostly junk ads" or playing whack-a-mole.
There are plenty of good publishers here where AdSense is a good fit yet they still experience undesirable ads sometimes.
| 5:36 pm on Dec 6, 2008 (gmt 0)|
Duplicate post removed by FarmBoy
| 5:48 pm on Dec 6, 2008 (gmt 0)|
|I haven't seen anything to convince me that comments by prior ASA's was anything other than boilerplate language. |
I haven't, either. Heck, I said they were boilerplate.
|Must you attempt to belittle everyone who doesn't agree with your opinion? |
Nobody is "belittling" publishers who don't do well with AdSense. I took AdSense off one of my own sites because the ads were junk, but that didn't mean I belittled myself or the site.
Unfortunately, a lot of publishers (at least on this forum) do believe that AdSense features and tools are the solution to their problems with AdSense. Features and tools may be nice, but if if AdSense is a pig's ear for a publisher, new features and tool won't turn it into a silk purse.
| 6:08 pm on Dec 6, 2008 (gmt 0)|
|Features and tools may be nice, but if if AdSense is a pig's ear for a publisher, new features and tool won't turn it into a silk purse. |
I'll have to agree. Like Hobbs I use my filters and channels but neither is maxed out.
Channels are very helpful to me when I'm developing new areas, telling me what works and what doesn't but once I learn that I don't need more channels, I can retire those that have served their purpose and apply them to new tasks. Maybe some instruction on how to use existing channels more effectively would be better and use less bandwidth than adding more channels.
But this is about requests, not necessarily just features. My request would be more information about how things (especially smart pricing) work.
[edited by: OnlyToday at 6:09 pm (utc) on Dec. 6, 2008]
| 6:12 pm on Dec 6, 2008 (gmt 0)|
|a lot of publishers (at least on this forum) do believe that AdSense features and tools are the solution to their problems with AdSense |
Apparently Google thinks that it might be a good idea to improve the Adsense program. Otherwise ASA would not ask us to name and prioritize our feature requests, right?
Before I hand over a visitor for fractions of a cent to an avertiser who runs, say, a dieting site (where my sites have nothing to do with dieting at all), I would like to keep this visitor on my site. Simple. Alternatively, IF the diet site wants my visitor, they need to pay at least X cents. Simple. If Google has not got suitable ads, please display my house ad. What is wrong with that?
The availability of better filters =does= solve problems that some forum members experience. q.e.d.
| 6:39 pm on Dec 6, 2008 (gmt 0)|
There's been some great discussion here, but I get depressed when things digress into a publisher telling google they need to hire more people, or that their managers aren't good enough.
This kind of thing makes publishers look very bad, and maks me wish I could disassociate myself from y'all. Not to mention some of the posts are offensive and insulting to google (but some fool is going to tell me the deserve it).
Back to topic...
Extra features won't help the core problems.
Google needs to enforce its existing rules properly.
Communication is great but not at the expense of providing publishers with a quality product/service that works, does not insult US, and provides a good user experience.
However, IF google refuses to enforce its own rules, then yes, give us proper filters so we can remain in the program and police things ourselves.
To put this in perspective my sites cater to executives and leaders, managers, HR people,etc who are making six figure incomes, and really don't want to buy dead popes. Police the ads properly.
| 7:00 pm on Dec 6, 2008 (gmt 0)|
|To put this in perspective my sites cater to executives and leaders, managers, HR people,etc who are making six figure incomes, and really don't want to buy dead popes. |
Who knows? There are only so many dead popes available for sale, and religious relics, like art, can be a good long-term investment. :-)
| 7:04 pm on Dec 6, 2008 (gmt 0)|
I'll try to keep this short.
|1. Policy enforcement is not easy. We have a lot of people working on it, including many of the smartest people I’ve ever met. |
I understand this. But please understand that it is RESULTS that count. It's your reputation (and ours) that is at stake, and I suggest to you that the lowered CTR's are at least, in part, are a consequence of your not achieving enough on the Adwords side (you can't separate the two sides to solve these problems).
2. Scalability is a factor.
I understand that too, and google's strategy in that respect, probably more than most people here, since I have some familiarity with algorithms. The deal is that there are markers you can establish as being associated with both problem advertisers and problem publishers, and that you can automatically use these to flag sites, or put them on hold until a human can get to them. There's been discussion (this is just an example) that no domains be admitted to the program until six months after they have been bought or transfered on the adsense side. There have actually been a number of ideas presented here over the last years that are scalable and oriented towards quality.
So, if we can figure out, conceptually how to do it, you can figure out the details. If you can't be honest, and change your terms of service, give up on quality. Because without quality you are just adbrite. (I don't mean to pick on them except to use them as an example of a company that appears to allow anyone in and anyone to advertise (adult sites, torrent sites, etc).
3. We’re not in the business of judging the quality of websites. We create and update a list of criteria and things that are not allowed, and we go from there.
With respect you ARE in that business. You accept and reject adsense sites based on what? You claim to accept and reject ads based on what?
I think dibbern2 has a good point (though I would think of the AdSense team as more of a friendly ‘little g’). Apart from policy violations, should it be up to us to determine the value of a site to advertisers or to internet users? Would you trust me to do that? I sure wouldn’t.
Ackkk. Has it occurred to you that as a google publisher I TRUST you (I have no choice) with the choices as to what ads are displayed on my site? I HAVE TO TRUST YOU, since you don't give me the tools (and I don't have the time) to do YOUR quality assurance.
I could write a small book responding to the above. Most of what you have said in this thread seems sensible to me. The above both depresses me and infuriates me.
Let me speak as an advertiser. The power of adwords and adsense (and its marketability, I might add) is that I can advertise without killing myself with the details of which sites, and how much, etc. If I have to hire a fulltime person to handle adwords, and to decide which specific sites in the program I should be on, what I'll do is hire a full time person to sell ads direct and compete with you.
And I'll make a bleepload of money more (at least after the recession).
As an advertiser I want you to make choices on my behalf.
You already decide WHERE my ads show, at what cost (I don't determine my real cost, only my max), how often, and on and on.
As a publisher you ALREADY DECIDE what is shown on my site, what is relevant, etc.
Now I yell.
GOOGLE DECIDES ALMOST EVERYTHING ABOUT BOTH ADSENSE AND ADWORDS ON OUR BEHALF.
We MUST trust you enough to allow you to do that. Despite your black box.
Why? Because we believe that google is smart enough (actually, for the first time since google emerged, I don't know if i believe this) to realize that the better google does its job and puts money in advertisers and publishers pockets, the more google will make.
I dunno what you are doing now, I guess we'll see the numbers in your quarterlies, but no doubt it will be chalked up to bad economic conditions, and not on bad decisions.
In any event, not to be snarky here, but if you want more of my input, I'm a business consultant, and you can hire me at a ridiculous rate. I haven't the time to spend on this anymore. I suspect if you really don't understand your relationships with advertisers and publishers to the point where you ask whether we want you to make decisions for us, you are going to need some help somewhere.
Just enforce your existing policies. I don't care how. This is business. Let's see some results, and let's see quality.
You could be Lycos soon.
| 7:39 pm on Dec 6, 2008 (gmt 0)|
|Not to mention some of the posts are offensive and insulting to google (but some fool is going to tell me the deserve it). |
I do get the sickening feeling that Google or some parts of it have gone off the rails. This wouldn't be serious if the whole economy were not also simultaneously going off the rails. Yes, I think we may be in for some creative destruction (after the mindless destruction). I hate what I'm saying here and I wish I could be more positive but it really is hard to reconcile this thread with a positive outcome.
| 7:45 pm on Dec 6, 2008 (gmt 0)|
|Google needs to enforce its existing rules properly. |
I get depressed when things digress into a publisher telling Google that they need to enforce its existing rules properly. This kind of thing makes publishers look very bad, and makes me wish I could disassociate myself from y'all.
| 7:56 pm on Dec 6, 2008 (gmt 0)|
Probably not, to judge from Google's recent history and market share--not to mention a report the other day from MediaPost that said Google has 54.9 percent more active advertisers than it did in the 4th quarter of 2007.
Back to requests. Here's mine:
Introduce (and publicly acknowledge) a quality-scoring mechanism for publishers that, among other things, draws on TrustRank and authority data from Google Search. Use the publisher quality score to influence payout percentage, how ads are allocated, or both.
This step would accomplish two things:
1) It would reward publishers who produce content of intrinsic value, while providing a negative incentive to "AdSense opportunists."
2) Over time, it would create a better environment for advertisers, thereby encouraging more advertisers to use the content network.
| 8:11 pm on Dec 6, 2008 (gmt 0)|
|...a report the other day from MediaPost that said Google has 54.9 percent more active advertisers than it did in the 4th quarter of 2007. |
The fact that contextual advertising is more accountable than traditional kinds may be the saving grace that Google needs because make no mistake that the whole pie is shrinking rapidly. Actually, being Lycos may not be so bad as being a daily newspaper or a tv station. Broadcast outlets have fixed costs in the form of contracts with talent, I do believe Google's DC's are paid for.
|Introduce (and publicly acknowledge) a quality-scoring mechanism for publishers that, among other things, draws on TrustRank and authority data from Google Search. |
I like this and believe it would have the effect you say.
If this pronouncement by ASA is not true:
|We’re not in the business of judging the quality of websites. |
| 9:53 pm on Dec 6, 2008 (gmt 0)|
Well, they're in the business of judging the quality of advertisers' landing pages, so why not apply a similar mechanism to publishers sites? They wouldn't even have to call it a quality score--they could use a term like "Smart Rating" or "Smart Score." :-)
| 9:56 pm on Dec 6, 2008 (gmt 0)|
|among other things, draws on TrustRank and authority data from Google Search. |
There be dragons on this path.
The mechanisms at play in TrustRank and the even more mysterious authority values are just partially understood by webmasters. And that's by design, of course, to keep the bad guys guessing. Its not going to get any easier, unless you are a PR9 dot edu or equivalent.
Consider also the roller coaster status of the search index, with daily churn punctuated by disaster events like Bourbon, Florida, and on and on.
Would you really want your ad serving and payout rate to go south because its hitched to the serps wagon in some way?
These are just the top thoughts. There are enough other dragons to make a very long, off-topic post.
| 11:52 pm on Dec 6, 2008 (gmt 0)|
OK, antispam data, then.
My point is that Google does invest a lot of effort and money on the search side to distinguish wheat from chaff, and there might be value in having AdSense tap into what Google Search knows at some basic level.
| 12:00 am on Dec 7, 2008 (gmt 0)|
Sorry for drifting off topic again, but I have believed that there is a trust factor employed already on the AdSense side.
No proof, just a hunch from looking at some obscure AS sections.