vgeek

Haven't there been numerous recent leaks detailing how Google has let revenue team take over from the search quality team over the past few years? [1] They can adjust auction parameters for owned properties to juice CPCs, blend ads to drive CTR and disable functionality of ad-blockers in tranches to squeeze as much growth as needed. Search quality has seemed awful for quite some time [2], but all that really matters is growth for ad ePM-- driven by CTR and CPC. The leaked documents from last year also had comments about correlations between lower search quality and higher ad click rates, so there is that, too.

Youtube also increases ad load. More ads per video, combined with organic growth, higher engagement and wider distribution (smart TVs probably have insane metrics-- plus no real ad blockers there, either) mean they'll keep having display/video ads doing extremely well, too.

[1] https://www.techspot.com/news/102765-who-prabhakar-raghavan-... [2] https://methodshop.com/why-google-search-sucks/

Flock (YC17)

They are growing and showing up everywhere, but you never hear about them in terms of real innovation, just helping build the surveillance state. Combine this with cloud based doorbell cameras and it seems like someone is always watching.

This is definitely a neat place to visit. No digital cameras are allowed past a specific point, even. I have snapshots from walking the main loop with my dog from a few years back posted here:

https://vgeek.net/gallery/index.php/category/103-green_bank_...

Who?

I think of your suggested idea of objective reviews and classifications quite often, but the problem ends up being discoverability. There have been a handful of sites that were highly useful by aggregating price and consumer sentiment/recommendations from Reddit. They get initial visibility and search traffic, then get penalized by Google. Froogle/Google shopping gets priority and is the only aggregator that Google really wants to include.

Look at defunct sites like Nextag that were moderately useful in the space-- they had free and paid placements. They were steadily growing search visibility until Google started pushing their own product (Froogle, free product listings in 2014ish) and Nextag suddenly "violated Google's policies" and lost 90%+ of their traffic rather quickly (probably 1MM daily visits to under 10k basically overnight). Google shopping technically offers "free product listing placement", hidden well below the ads-- likely as a defense to anti-trust on monopolization of that specific space.

Brickseek and CamelCamelCamel are the two most successful/long lived tools in the space-- and they grew their visibility from the now defunct but once huge deal site FatWallet and SlickDeals. Walled garden subreddits frequently disallow posting of specific tools, so it makes organic growth super challenging-- given that pretty much all commercial queries start with Google or Amazon.

The same thing happens every 3-5 years. HowStuffWorks, About.com (now like 10 different domains), many IAC acquired properties, RedVenture sites, even random sites like LiveStrong.com will be wildly prominent when the domains historically aren't relevant or authoritative for a given niche.

Even recently, sites like CNN were using subdomains with affiliate offers managed by third parties(1). These sites weren't being de-ranked algorithmically-- someone at Google would have to apply a manual action to remove them from the SERPs. What incentive would there be to do so if a prior agreement was in place?

Google doesn't really care about discoverability for smaller domains that may have good content. They are either being risk averse (avoiding potential spammers, junk AI content) by favoring trusted domains, favoring brands who are likely to spend on display or search ads, or maybe a combination of these.

1) https://searchengineland.com/google-begins-enforcement-of-si...

Flock (YC17) is the primary facilitator here, along with their ALPRs. This is the first, and surely not the last, where the data they collect for law enforcement, HOAs and others, ends up in the hands of industry.

Anyone who has managed ads with performance in mind will be familiar with their tactics. (My post history on HN includes plenty of complaints like neutering reporting, organic search quality and how they essentially extort brands via brand trademark bidding.)

The 3 components that make up quality score: ad relevance, landing page quality and clickthrough rate are a black box. You can have ads and landing pages that match specific keywords (such as in a single keyword ad group strategy) and still have 1/10 scores on ad relevance and landing page and an insane CTR (think 40%+). You can look at search competition reports, but in some niches with few participants (think very specialized manufacturing equipment) you end up with the same <5 participants for years, yet the effective ad costs increase every quarter/year. Google adjusts bid floors for auctions, even if there isn't really upward pressure resulting from an increase in number of advertisers or participant bids.

Related to this-- their goal is always to increase the number of participants in every auction. The ad fill rate is easily improved by loosening match types. It isn't just the SEO space that Google has been redefining words in-- they do the same with keywords and match types. For example, "cheap textbook", "cheap textbooks" and "cheapest textbooks" are 3 distinct queries with widely varying CTRs and conversion rates, and hence values to advertisers-- but even when using exact match, Google may give you any one of the examples when you specify which one you prefer as an exact match.

Most infuriatingly is how you don't 100% know what you are paying for-- even when you explicitly specify what you want. Anyone who has looked at the search queries report can attest that sometimes 50%+ (emphasis on the insanity! I have observed this on $x,xxx,xxx/yr spend accounts all the way to my own personal campaigns) of ad spend for search ads, where you are explicitly saying I want to show for these specific keywords, is bucketed in the "Total: Other search terms" row. *It did not used to behave this way.* Negative match keywords are less aggressive in what is excluded relative to their exact/(phrase)/broad modified equivalents. This entirely ignores concepts like Google Shopping Ads requiring constant negative matching of longtail keywords (including for terms that you can't even really discover) that are all super small amounts individually, but in aggregate can be ludicrous amounts of spend that is essentially impossible to block for less sophisticated advertisers.

Every incremental change by Google seems to be to make the quality worse for the end user-- whether they are the searcher getting worse results or* the advertiser being goaded into wasting money on things Google isn't accurately presenting.* Specialty publications like Search Engine Land covered large things like the changes in second price auctions, display going to first price and match types-- but they typically aren't super critical, because a handful of Google's organic search "liaisons" are hires from industry publications.

I also follow SERoundtable (I have worked as SEO/digital marketer/developer for roughly 20 years), but tend to discount many of the comments due to the assumption that many of the people complaining in broken English may not actually have the quality of site that they believe they do, but there are tons of good sites getting caught up in updates-- not just now, but in every update. The past ~2-3 years have had entire types of sites (e.g., useful blogs, data driven sites, useful/non-spammy aggregator sites) get wholesale demoted/deranked/deindexed.

In ~2016 Google started shifting towards optimizing for financial objectives more aggressively than user experience. Timing updates to coincide with beginning/end of fiscal quarters, blending ads, features solely created to drive incremental searches (People Also Ask/Related Searches), various misaligned defaults within GAds interface, branded search extortion, stance against header bidding, etc.. Essentially when they stopped promoting the "Don't be evil." slogan, they had legitimate reason to do so.

If I could give anyone advice with regards to establishing a website that is reliant on Google for traffic-- it would be to be extremely careful. I have one site now that is super high utility for end users, great UX, super fast, high repeat user rate, no ads/tracking/popup spam, great feedback from users and it is -60% in Google traffic from the March 2024 core update. There is 0 support from anyone at Google to identify why a site suddenly loses traffic. There are search liaisons who give snarky replies, but good luck getting any constructive feedback.

Even relying on paid traffic is just as dangerous-- given the black box that is Quality Score (it ties mostly to Click Through Rate, but has adjustable floor to increase effective costs) and Google's consistent drive away from measurable performance that helped them destroy traditional marketing channels so successfully.

All that I can think is that there is absolute panic at Google right now. When >50% of product searches start directly on Amazon (https://searchengineland.com/50-of-product-searches-start-on...), Google can't siphon anything off. With Meta adding things like Llama 3 to FB Messenger, there is going to be another huge hit to Google query volume-- albeit most likely low commercial intent queries (see: not as monetizable by Google), at least initially, but it will help increase user familiarity with chatbots and observed data will probably help improve Meta ad targeting ability in ways that may rival search query intent.

High value categories like home services, banking and finance are among Google's last relatively safe bastions of profit-- but eventually advertisers in these spaces have to reach a level of sophistication to realize they're giving too much of their margin to Google, leading to push-back. Highly fragmented, lower margin spaces like restaurants (or other "near me" driven niches) that have success on GMaps seem relatively safe for Google at this point. If Meta handles the chatbot transition (if it actually happens) well, they stand to gain a lot of ground there, too, given that they do already have a decent amount of small businesses who use FB pages as their sole internet presence, along with associated meta-data like hours/location/menus/reviews.

"Here are pre-packaged PR points why to not pursue further antitrust action against the 21st century's largest monopoly who also pretty aggressively avoids paying their fair share in taxes."

A herd of corgis.

This seems tangentially similar to Times Beach-- where dioxins were used for dust control.

https://en.wikipedia.org/wiki/Times_Beach,_Missouri

Yes, you are stating the obvious assumptions that altruism is absent and there is an adversarial relationship between capital and labor with frequently diametric objectives. If companies are constantly complaining about "nobody wants to work" then they could move further up the labor supply curve by offering better wages-- but would it actually improve employer profits (via higher revenue) if they give up margin? If a McDonalds could serve 300 customers instead of 200 over a lunch rush, wouldn't it be beneficial to the employer?

From the employee side, they would get paid more, could pick up shifts at varying businesses based on their schedules. Then maybe at scale, such a platform would be incredibly useful for organizing labor. This obviously would terrify employers.

I'm not really debating, more of just sharing random ideas and complaining online, because this is basic game theory as to why employers wouldn't want such a tool (potentially higher labor costs short/long term, increased labor mobility, unionization risk, other issues like training/liability/insurance/taxes/etc.), but employees/labor could be greatly served by such a platform.

Why is it always surge pricing for consumers, but not employee compensation? (in a non-exploitative manner like the gig companies employee) Yes, there have been cash-flow improvements like daily payments, and maybe other improvements to scheduling apps-- why not a solution for surge pricing for employees? E.g., Wendy's is busy for the Tuesday afternoon shift, wages +30% for workers who want to come in for X hours (if order throughput is +300% from 11am-1pm, why wouldn't they want extra hands, even if the marginal cost is much higher, albeit more transient). Or Amazon/DHL/UPS sees an increase over a weekend and offers $X bonus per shift for standby/off shift employees, increasing the amount until enough people volunteer.

Given that labor is viewed as fungible by most employers, why not enable more promiscuity by employees to help with wages and/or security? Things like background checks seem fairly transitive. Maybe do some sort of skill set certifications like food handling, warehousing, basic labor, etc.?

Matomo has per-user tracking (visits log) across sessions. Attribution won't be perfect, but you can make it tolerable with UTM tags. It can be GDPR compliant without much effort. You can also query the raw data via SQL if you want.

PostHog is pretty popular for product companies now, but who knows how long they will continue with a free offering.

Nvidia gpu? If so, you can disable Nvidia specific suspend/hibernate functions and it may work fine.

Having managed branded search spend down, it is 100%. Extortion would likely be a better word, the "x" makes it seem cooler while being more representative of the "pay us or we show your competitors" situation.

Yep, they sold DunnhumbyUSA to Kroger and rebranded to 84.51. They used to have a super visible building right by Fort Washington Way (downtown interstate) and moved closer to Kroger HQ maybe 8-10 years ago after the sale. They still did other shopper card analytics type work, but they are pretty much only Kroger brands (and associated suppliers-- think typical CPG companies) at this point.

Blue Ash is a more educated/affluent suburb with significant P&G and Kroger facilities. Probably preferable to a 1hr each way (typical for many suburbs to downtown) commute and paying for parking. Downtown Cincinnati is kind of boring-- there have been a few new places open, but there are better choices for food and entertainment in various other neighborhoods.

I had an interview at Kroger HQ out of college and it was probably the most depressing office building I had ever been in. That combined with the type of work encountered at legacy industry bigco and it has to be hard to recruit anyone decent. (This coming from someone who is wildly inept, too :D)

Their analytics company 84.51/dunnhumby was always the flashier option that could recruit at a national level.

As useless trivia-- one of Sanders' relatives founded Lee's Famous Recipe once Sanders essentially exited KFC. Subjectively, it is better than modern day KFC.

The Corbin KFC/museum draws tons of tourists-- especially from China (I think they have more locations there than in the US.) If anyone happens to go there, Corbin also has a pretty good pinball arcade with ~60 tables and a redemption game section.

Focus on performance marketers who have verifiable performance. You want someone who speaks in numbers (e.g., ask vlookup vs index/match type questions) and is driven to test-- so look for someone who can explain different strategies (e.g. alpha/beta, how they segregate kws/match types, thoughts on long tail, negative management, manually managing bids vs automated systems, campaign structures, even basic things like constrained budget allocation) in a manner you feel comfortable with. Given your niches and scale, maybe place more emphasis on testing novel channels to help establish both CACs and LTVs, because much of automotive search will be likely weirdly high funnel because it is a highly considered purchase category.

Marketers are going to be good self-promoters (it is their job, after all), and the good ones are worth a lot because they can save multiples of their comp in wasted spend, but most are unfortunately flunkies who parrot Google's talking points, always use automated bidding and constantly change KPIs.

Fark.com

https://www.aliexpress.us/item/3256805789824743.html is neat, but has limited configuration options.

Maybe Nixie tube clocks or e-ink displays?

Any un-paywalled source?

You're forgetting the additional eye-poke joys of minor plugin updates causing WSODs, how slow it is without caching, how painful styling certain things can be, how difficult the DB schema is to wrangle with regards to making _any_ changes and dozens of others that are worth forgetting because of various PTSD.

It _is_ flexible for people wanting to be able to create types/taxonomies within the GUI, if you ignore the chaos behind the scenes. It also has good support for returning said content via JSON if you choose to do so, but if you're doing that, just define your own back-end and avoid the suffering.

Have you had Crabby Jack's on Jefferson Highway?

A similar Intercept article was posted on HN a while back, and got limited discussion: https://news.ycombinator.com/item?id=31127965

As much as HN hates advertising, so many seem ignorant of how much of the ecosystem works. I guess lots of people simply work for companies involved in these schemes, so they simply turn a blind eye or feign ignorance.

You can visit a gaseous diffusion site outside of Oak Ridge, TN at a new, albeit small, museum: https://amse.org/k-25-history-center/ Much of the infrastructure has been leveled, but there are a few remnants. Site selection factors included obscurity, lots of water from nearby rivers and electricity availability-- they needed MW in a time where KW was a lot of power. There is also the AMSE itself and tours of supercomputers at ORNL.

Due to the availability of hydro generation in the early 20th century, Alcoa, TN (Aluminum Company of America) also established operations in the same area due to the energy required for smelting. Their other large operation hub? Niagara Falls.

Their discoverability was awful.

What about how AliExpress' search works? The first few listings will be prior viewed items, typically wildly unrelated to the current search-- just remarketing ads based on products visited. Then the things labeled as ads are even more outlandish and inappropriate (serums, pills, pseudo-scams)-- but I guess the ROI is there, else they wouldn't be running them. Similar to the old days of FB where any scammy rebill products were allowed, then once an advertiser base was built, they canned all of the exceptionally unscrupulous advertisers.