Google AI Mode is now showing more ads than usual when searching for ecommerce queries. Previously, AI Mode would show organic product results within its answer, with ads appearing separately at the bottom. Today, when searching for headphones, we noticed that the products shown within the context were entirely ads not a single organic product appeared.

Here is a screenshot and a video of this.

Showing more ads within these results could reduce visibility for organic traffic, which isn’t great news for websites that rely purely on SEO rather than advertising. As an SEO myself, I’m not a fan of this shift.

The post Google AI Mode now showing more ads appeared first on Search Engine Watch.

Source:: searchenginewatch.com

Ubersuggest has had one of the stranger evolutions in SEO software.

Search Engine Watch wrote about it back in 2013, when Übersuggest was basically a free way to turn Google Suggest data into long-tail keyword ideas (kind of like Yoast Suggest, but a bit fancier). By 2018, it had become a more recognizable keyword research platform with search volume, CPC, and competition data.

But I don’t think another feature-by-feature tour is useful to anyone. There are plenty of other sites that do this.

I run TORO RANK, an SEO agency, and I work with lots of websites – from small ecommerce stores to large SaaS companies. My approach to judging software is slightly different. I don’t care if a platform can technically produce a backlink report, crawl a website, or generate keyword ideas. Most established SEO platforms can do all of those things.

What matters to me is whether the tool gets me to a useful decision faster.

Is it better to create five topics than to export 50,000 keywords? Am I able to identify enough of a competitor’s strategy to determine where to investigate next? Would someone with limited technical knowledge be able to use the data without spending hours learning the software? Equally importantly, at what point would I stop trusting Ubersuggest and reach for a more specialized tool?

You’ll find the answers below.

Ubersuggest starts at $29 per month, making comparisons with considerably more expensive platforms inevitable. However, after using the product, I don’t believe the most relevant question is whether Ubersuggest can replace Ahrefs or Semrush. Because it can not.

But there’s another question:

How much SEO software does a small business, freelancer, or agency actually need before more data stops improving the decision?

In my opinion, Ubersuggest comes close to crossing the line, but it also makes its limitations obvious once you advance in your career and gain more XP and knowledge. Only then it’s time for the big boys.

What Ubersuggest has become in 2026

Today, Ubersuggest is basically a budget SEO suite.

But don’t confuse it with a Semrush from Temu. Or a cheap Ahrefs alternative.

Ubersuggest does well at reducing the number of decisions a user has to make before receiving useful information. Instead of expecting you to understand a large SEO database first, it organizes information around relatively straightforward, albeit simpler questions.

1. What should I target?
2. What works for competitors?
3. What needs fixing?
4. Where are my rankings changing?
5. Where might I get links?
6. Is my brand appearing in AI search?

This design decision is more important than adding another checkbox to a feature comparison table.

For a small company doing its own SEO, this can be just enough. For an experienced SEO who wants raw exports, unusual filters, custom crawling, historical comparisons, and large datasets, well… if you were, you probably wouldn’t be reading this.

User-friendly dashboard

One test I use to evaluate SEO software is to ask myself if I could give it to a client, junior marketer, or non-technical colleague without first giving them a 45-minute tour.

Ubersuggest does pretty well here.

This becomes even more impressive when you consider everything that has been added to the product. It includes keyword research, competitor analysis, rank tracking, auditing, backlink research, and an entire AI layer. Usually, products become considerably harder to navigate as these types of capabilities accumulate.

Ubersuggest hasn’t completely avoided that problem, but it still keeps that task-oriented simplicity, so kudos to their team.

In practice, if you have a junior on your team, you can just tell them, “Find what this competitor ranks for,” or “Check why this site score dropped,” and they’ll find it relatively fast.

That sounds like a minor advantage. But in a growing agency, it isn’t.

Senior SEO time is expensive. If a tool requires an experienced specialist for every basic inquiry, the actual cost of the software exceeds its subscription price.

This is an area where Ubersuggest’s simplicity is valuable, not just beginner-friendly.

Keyword research is still the strongest reason to use Ubersuggest

Of all the new features Ubersuggest has added, I would be most comfortable incorporating keyword research into a regular workflow.

Not because its keyword database beats the competition.

It doesn’t.

I’d still prefer using ahrefs for kw research.

But I like how easy it is to enter a topic and get a shortlist worth investigating.

In agency work, acquiring an additional 30,000 keyword rows isn’t a challenge. Deciding which ten deserve someone’s time is. It’s like… good for smaller projects, very intuitive, and you get all the basic data you need to start a new website.

The classic Keyword Ideas report is still useful

This is probably the least exciting screenshot in this review, but it’s one of the Ubersuggest features I actually use from time to time.

The report provides the familiar data on volume, intent, cost per click (CPC), and difficulty, as well as different ways to expand a seed topic.

Nothing here is WOW.

But it just works.

For some websites, all you need are several plausible and verified opportunities to help you decide what deserves a SERP analysis.

Ubersuggest does that well.

I still wouldn’t approve a content brief just because a keyword has an attractive difficulty score. First, I would inspect the SERP, understand the intent, and look at the sites already ranking. Then, I would compare that with the website I’m working on.

But Ubersuggest can make the first filtering stage considerably faster.

AI Keyword Overview is where the product gets more interesting

I’m losing interest in SEO platforms that add generic AI writers or chatbots.

I already have access to much better AI tools.

What I have implemented in my workflows, and Ubersuggest does as well, is connecting the LLM of your choice to the actual data.

That’s the direction that makes sense.

A language model can generate 100 plausible keywords in seconds. However, it cannot reliably determine which keywords have meaningful search demand, what currently ranks, how difficult the SERP looks, or how competitors are performing.

That makes AI Keyword Overview much more interesting to me than another “Generate Content” button.

This will matter much more over the next few years. The SEO platforms that benefit most from AI won’t necessarily be the ones with the best chatbot. Instead, they’ll be the ones with useful proprietary or licensed data that AI can query.

Ubersuggest appears to be heading in that direction.

Competitor analysis is not really competitive intelligence

Right off the bat, don’t use Ubersuggest for competitive intelligence. It can be okay for a cold lead report, but that’s about that.

But you could absolutely use it for the first 20 minutes of proper research.

When I start working on a website in an unfamiliar niche, I usually have some fairly simple questions.

• Who is winning?
• Which sections of their sites attract search traffic?
• Which topics repeatedly appear across competitors?
• Where are the obvious gaps?
• Which pages deserve manual investigation?

Ubersuggest helps you quickly get through the initial orientation.

For smaller companies, this may be the only competitive research they need.

For my agency, it is sometimes used in the briefing.

I also treat all competitor traffic numbers as directional. Neither Ubersuggest nor the other third-party SEO platforms have access to the actual data.

If one competitor appears dramatically stronger than another, it’s reason enough to investigate the difference. I wouldn’t include that information in a client report as audited traffic, though.

Backlink analysis is okay-ish

Backlinks are where the slogan “you get what you pay for” applies.

It is fine for routine checks.

But for serious link research, I need a more robust dataset.

The only think I kinda like is the backlink intersection. It looks and feels much easier than ahrefs‘.

Also, Ubersuggest is decent enough for generating a prospecting shortlist, identifying obvious gaps, and supporting lightweight competitor research.

However, I’d never make it my primary link-intelligence platform if link building is crucial or a substantial part of an account.

Site Audit – incredibly basic

I wasn’t even going to write this section, but here it goes.

Yes, Ubersuggest has a Site Audit.

Is it good?

No.

Is it okay for beginners?

Arguably.

However, if you are a beginner or a small freelancer with 1-2 websites, it has its benefits.

That said, when investigating crawl behavior, JavaScript rendering, canonicals across hundreds of thousands of URLs, unusual directives, structured extraction, or complex e-commerce architecture, I need reliable crawling software like Screaming Frog and/or Sitebulb.

Ubersuggest can’t give you that complete picture.

BUT!

Small business owners usually don’t want 40 columns of crawl data.

They want to know:

“Is something seriously wrong with my website?”

“What should I fix first?”

“Did new problems appear this week?”

For that job, Ubersuggest is okay, but don’t expect too much from it.

The crawl limits reinforce this. While they are reasonable for many small businesses and brochure websites, they are clearly not designed for enormous e-commerce, publishing, or programmatic SEO estates.

That said, the audit is OK for a recurring website health monitoring for smaller websites. Don’t expect to perform any in-depth technical SEO forensics.

Rank Tracking is useful for small websites

I don’t think the biggest question with rank tracking is whether a platform lets you monitor 125 keywords, 1,000 keywords, or 100,000.

The question is whether anyone takes action when those rankings change.

If you deliberately track queries connected to important products, services, or content areas, Ubersuggest’s limits can be enough for a small site.

However, for an agency maintaining extensive keyword universes for dozens of clients, the limits + everything I said above, make Ubersuggest unusable.

AI Search Visibility is basically a warning system

I like that Ubersuggest has added AI visibility monitoring.

However, I would advise caution when converting the resulting percentage into an additional vanity KPI.

A limited set of prompts cannot provide a universal understanding of your “AI visibility.” Different queries, models, locations, phrasings, and refresh cycles can produce different answers.

What the feature can do is give smaller brands a practical early-warning system.

Are competitors appearing where we aren’t?

Which sources keep getting cited?

Did our brand disappear from a prompt where it previously appeared?

Those are important questions.

For just $29 per month, you can get a solid package that includes traditional SEO monitoring and a small AI-search dataset, all in one account.

I wouldn’t pretend that the number is more comprehensive than it actually is.

The ChatGPT app and MCP server may be Ubersuggest’s smartest 2026 additions

This, in my personal opinion will become more important than the other reports in the Ubersuggest dashboard. People are getting more and more used to the chat interface, and most of them are going in that direction.

That said, Ubersuggest’s ChatGPT integration lets SEO questions begin conversationally while Ubersuggest supplies the underlying keyword data. Which is nice.

When I ask an AI system for “20 low-competition accounting keywords with reasonable US search demand,” it can generate convincing suggestions. However, the part that the model cannot safely invent is what I actually need from SEO software: volume, difficulty, SERP information, and real search data.

MCP could be even more important

The MCP integration takes the same idea further.

Rather than manually entering data from Ubersuggest into the next application, your agent can query Ubersuggest as part of a larger workflow.

That’s potentially useful in an agency.

• A content workflow could investigate competitors. It could collect keyword opportunities. It could turn the findings into a brief.
• For example, an analyst could request a list of a site’s major audit problems and then use an AI assistant to categorize them.
• Researchers can now combine domain, keyword, and backlink questions without having to manually navigate through several reports.

Of course, I wouldn’t automate the judgment itself. However, automating the repetitive movement of data between tools is nice-to-have.

In that sense, Ubersuggest’s future may be less about creating the best SEO dashboard and more about becoming an affordable SEO data layer for other interfaces to use.

AI Writer is useful, but I wouldn’t buy Ubersuggest because of it

The AI writer feature is exactly the opposite of how I feel about the ChatGPT and MCP integrations.

It works.

It can generate ideas, titles, outlines, metadata and drafts.

But in 2026, that’s commodity content, to say the least.

Most marketers already have ChatGPT, Claude, Gemini, or another AI writing workflow in place.

So I wouldn’t give Ubersuggest extra points because it can generate an article draft.

What they could do better to keep the feature is supply it with the data they already have. Just saying. Right now, it’s just a blank free-chatgpt-like writer.

While there’s plenty of AI-generated text, reliable data and useful context are scarce.

Ubersuggest is the short answer, not the conversation

After spending years on the platform, I think that most of Ubersuggest’s weaknesses come from the same design choice as its strengths.

It wants to get you to an answer quickly. It doesn’t want to give you reasons to analyze and go deeper. Simply because it can’t support such a workflow. But hey, 29$/months is not like $99.99/month.

These issues become much more noticeable when it comes to project limits, crawl limits, tracked keywords, export ceilings, and the lack of a conventional API.

For this reason, I believe that Ubersuggest should not be judged by the number of features it reproduces from Semrush or Ahrefs.

It’s suited for a totally different type of workflows and audiences.

Who I would actually recommend Ubersuggest to

I would most easily recommend it to a business that manages its own website, a solo publisher, a freelancer with a small client roster, or a content marketer who needs enough SEO information to make daily decisions without relying on SEO software.

The lifetime plan is especially compelling for that audience.

If you’ve tested Ubersuggest and know that your website fits comfortably within its limits, and if you expect to continue using the core research tools, the economics are hard to beat.

For an agency, though, my answer is different.

There are plenty of reasons to keep Ubersuggest in the toolkit.

However, I couldn’t imagine it to be the only tool.

Final verdict: Ubersuggest is a tool you’ll outgrow fast, if you are serious about SEO and GEO

In my opinion, Ubersuggest is worth the money in 2026.

But it’s not because it delivers a professional SEO platform worth $200 for just $29.

Because it doesn’t.

It covers the first layer of a lot of SEO decisions at a price that is easy for any serious website to justify.

Keyword research is useful. Competitor analysis helps you get oriented quickly. Backlinks can help you find prospects. Site Audit can identify obvious problems. Rank tracking is useful for smaller campaigns. AI Search Visibility provides smaller brands with an affordable way to monitor generative search.

For many website owners, that may be enough, and I don’t judge.

Ubersuggest’s limitations become apparent when SEO moves beyond a series of targeted questions and becomes a large-scale analytical workflow.

I want larger datasets, deeper link intelligence, specialized crawling, richer exports, more projects, greater automation, and conventional API access.

As an agency owner, I wouldn’t replace my professional SEO tools with Ubersuggest.

If I were running one or a few websites and paying for sophisticated SEO software primarily for keyword research, basic competitive analysis, and rank tracking, I would seriously question why I was spending four or five times more every month.

And I think that’s the strongest sales point for Ubersuggest.

The post Ubersuggest review: what I still use it for at my SEO agency appeared first on Search Engine Watch.

Source:: searchenginewatch.com

Google is giving businesses a new way to reach people while they decide where to go, with an advertising setting that prioritizes potential customers using Maps, Waze and local Search.

Called Local Customer Optimization, the feature sits inside Performance Max campaigns for store goals. Google Ads liaison Ginny Marvin announced the update alongside an easier way for eligible advertisers to feed purchases from physical stores into Google Ads.

Together, the changes show where Google is taking its local advertising pitch: reaching someone around the moment they choose a destination, then connecting advertising activity to money spent inside the business.

According to Google’s documentation, the new setting prioritizes people close to, or interested in, a business’s area who appear ready to act immediately. That includes people planning trips, navigating or looking for a local service. The goals can include visits, calls, directions and store sales.

Maps and Waze already form part of Performance Max’s store advertising options. On Waze, for example, a business can appear as a branded pin along a driving route, with location details and directions available when someone taps it. The new setting adds a specific budget priority for immediate local demand within that existing system.

For a restaurant or retailer, the potential attraction is timing: reaching someone who is already considering a nearby stop.

There are restrictions on which campaigns can use it. Local Customer Optimization requires offline goals and is incompatible with campaigns advertising products from a linked Merchant Center account. A retailer using that setup would need a separate eligible campaign to try the feature. Adding an online conversion goal also requires switching the setting off. Google explains the campaign restrictions here.

The second update addresses what happens after someone walks through the door.

Google says Store Sales in Data Manager will let eligible advertisers connect a CRM or Google Sheets directly to their advertising account, reducing the work needed to supply transaction data. Marvin’s announcement describes the local optimization rollout as underway globally, with the Data Manager integration due at the end of September.

Access to that measurement remains a significant qualification. Google’s published Store Sales requirements include account approval and, for sales counts and default value reporting, approximately 300,000 store visits and more than 500,000 account-level ad interactions over the preceding 90 days.

Manual reporting of transaction-based sales values carries additional requirements, including at least 30,000 uploaded transactions within 90 days and regular uploads of more than 10,000 new transactions from the past 30 days.

Those are existing measurement requirements; Google has not announced their removal alongside the simpler connection process. They apply to Store Sales reporting, separately from the Local Customer Optimization setting.

Photo credit: Mika Baumeister / Unsplash.

The post Google wants Maps and Waze to send more customers into stores appeared first on Search Engine Watch.

Source:: searchenginewatch.com

Amazon is opening a new route for its advertisers to reach people inside ChatGPT, putting its advertising business closer to the conversations where users decide what to buy.

The pilot launched on September 10 for select U.S. advertisers, with Delta Vacations among the first participants, according to Digiday. It allows participating businesses using Amazon Ads to buy placements inside the ChatGPT app.

Delta Vacations president Katrin Koenig said the collaboration would help the company reach travelers with relevant vacation options.

Amazon’s advertising operation already extends well beyond its online store. Its demand-side platform, or DSP, is software for buying ads across different websites and apps.

ChatGPT adds a setting where a potential customer can explain their needs in detail. Someone planning a holiday, for example, might discuss a budget, travel dates and destinations before choosing a provider. A relevant ad could reach that person while they are still comparing their options.

OpenAI says its ad system considers the context and intent of the current conversation, alongside information about the advertisement and other eligible signals. That makes research and planning conversations a potential opportunity for advertisers. OpenAI’s advertising documentation explains how those placements are selected.

There are some early indications that ChatGPT ads can introduce brands to new audiences. In an August 24 update, advertising technology company Criteo said more than 2,000 brands were using its ChatGPT integration. It reported that more than 80% of the resulting ad traffic came from new customers, with click-through rates two to three times higher than comparable formats in other environments.

Those are Criteo’s own figures from a separate integration. The update does not identify the comparison formats in detail or establish how many purchases would have happened without the ads. It provides an encouraging signal, but no benchmark for what Amazon’s advertisers should expect.

The Digiday report also points to continuing limitations in ChatGPT’s measurement tools, technology and ad formats. It does not spell out which Amazon targeting and reporting features the pilot supports.

For users, the placements remain subject to ChatGPT’s advertising rules. OpenAI says sponsored ads appear separately below responses, and advertisers cannot pay to change the answers themselves. It also says advertisers do not receive users‘ private conversations or chat histories. Ads may appear on the Free and Go plans; Plus, Pro, Business, Enterprise and Edu remain ad-free.

The post Amazon is bringing its ad empire to ChatGPT appeared first on Search Engine Watch.

Source:: searchenginewatch.com

Google has released Meridian GeoX globally, giving advertisers a way to test whether their spending generates extra sales or gets credit for purchases that would have happened anyway.

The open-source tool helps businesses run advertising experiments across different geographic areas. It can measure campaigns across advertising platforms, including those outside Google, and assess both online and offline sales. Google announced the global release after introducing the product earlier this year.

Google used the launch to criticize how advertising performance is often reported.

“Click-based attribution tools consistently double-count conversions,” the company wrote, warning that businesses can end up with “green dashboards, but flat bottom lines.”

Consider a shopper who clicks a social media ad, later clicks a search ad and then buys a product. Depending on how the reporting is configured, both platforms might claim the purchase. The business still made one sale. And even counting that sale correctly does not establish whether the customer needed either ad to make the purchase.

GeoX is designed to investigate that second problem. Marketers call it incrementality: the extra business an advertisement actually causes.

For example, a retailer could pause a campaign in selected regions while keeping it running in others. GeoX uses historical data and results from the experiment to estimate the effect of that change. The aim is to work out how sales differed from what would have been expected without the intervention. Google’s experiment guide describes the process, from collecting regional data to evaluating whether the measured effect is statistically significant.

Advertisers can also increase spending in selected areas to test whether a larger budget brings in additional business. The software supports testing several changes against a shared comparison group, which Google says can reduce the cost and time needed to run experiments. Google’s product documentation outlines the available designs.

Results can be used independently or fed into Meridian, Google’s existing marketing measurement software, to refine estimates of advertising returns and help businesses decide where to allocate their budgets. Because GeoX is open source, teams can inspect its code and methods.

Using it requires preparation. Advertisers need historical daily sales or other performance data broken down by geography, plus spending data for experiments that change existing campaign budgets. They also have to implement the campaign changes and collect results during the test.

Google’s technical paper identifies differences between regions and changing patterns over time as obstacles to reliable measurement. The researchers say GeoX’s methods improve accuracy and reduce experiment costs in their benchmarks.

The post Google launches tool to check whether ads actually increase sales appeared first on Search Engine Watch.

Source:: searchenginewatch.com