Truly Analytics Identifies Lessons from 2026 Digital Marketing

Truly Analytics revealed the real value of online directories and the overall reality of Digital Marketing in 2026

Directory submissions are often sold as a simple way to build backlinks, increase discovery and generate referral traffic. However, there’s effectively no quantification as to whether this works, if the listings are genuine or what the real traffic would be. Listings may go live, but indexing can vary, referral traffic may be limited and attribution often becomes unclear once multiple platforms claim influence over the same conversion.

This case study looks at how Truly Analytics was used to measure a directory listing campaign for the Truly Analytics product itself. A very meta approach to dogfooding our own product to answer our own questions.

To accommodate this we built First Party Attribution to understand which directories produced observable traffic, which referrals were linked to sessions and which claims could not be tied to direct activity. Further supported by Network Intelligence and Event Hooks we were able to both extract our traffic to an external long term attribution monitoring database as well as determine the quality of the traffic, extracting out bot activity and network origins.

This case study establishes how the campaign was measured, why standard analytics attribution was not sufficient on its own and what we observed and concluded from working at scale on this project. There’s a number of lessons digital marketing, GTM and other Founders can learn from the visibility provided by Truly Analytics’ featureset.

The Directory Campaign ¶

We purchased a directory listing service that submitted the Truly Analytics product to 140 free directories. The package also included some paid submissions within the listing provider’s own network. These paid placements were not made clear at the point of purchase and were only identified after reviewing each individual directory and network.

Instantly we have an immediate disconnect between what we were buying and what we got, plus what those networks actually are under the hood.

As the directory service is formulaic, we got all listings done and delivered back to us within a couple days complete with:

• Fresh gmail account used for registration.
• Screenshots showing the submission.
• Support to query and check the results.

Afterwards we then performed the following checking:

• Whether each directory listing launched.
• Whether each listing was indexed or discoverable through Google.
• Whether each listing was discoverable through Brave Search.
• Whether backlinks were present.
• Whether referral traffic arrived from the directory.
• Whether any referral session could be linked to meaningful product engagement.
• Whether attribution claims were supported by first-party session evidence.

We’d paid for the submission and outside a few problematic services we got what we paid for.

• Most directories were indeed directories. Some were not directories at all and had no hope of being relevant for backlinks.
• The service provider actually listed us on paid premium services for the directories that they directly owned. Strangely, this was not part of the service description, but welcome nevertheless as it allowed for higher chance of indexing.
• Some directories were completely non-functional, however we didn’t / couldn’t link this back to the service provider. These are just bad directories.
• Some directories rejected our application. We had paid for the submission work so this was fine. We dealt with those directly where applicable.

Tracking the Results ¶

Having listed on the directories and the submission provider having done their job, we can get to the meat of this report: what happened with traffic and backlinking.

Google Analytics is a useful reporting layer, however, it has huge problems with being blocked by users, complicating and dropping attribution and lacking further analytics reporting.

We built Truly Analytics because we knew that Google Analytics wasn’t giving us the data we needed. In our first case study Truly Analytics recovered 91.4% of the traffic for a cybersecurity firm. That’s a stunning figure that cannot be written off. Additionally, Google Analytics attribution runs tracking elements through the rest of the Google Advertising platform. For third party, non-advertising traffic such as a directory listings this is pointless, hence we added in First Party Attribution to give us a clear, one-to-one relationship between the attribution and the resulting session and any conversions.

As an engineering firm, we cannot accept total attribution exceeding 100%, multiple platforms all claiming credit and disregard for non-PPC attribution. These directory listings are exactly the attribution that would otherwise be disregarded or lost in Google’s platform and the perfect test for First Party Attribution.

Individual Directory Performance ¶

You can find our review of each directory including the individual attribution that we got from each directory, whether the backlink was discovered or indexed and our overall note & verdict at the below listing page.

• Each directory is detected for Domain Rating by AHREF’s API
• Each directory is checked weekly for indexing via Brave API and Google Search
• Any attribution that is detected by Truly Analytics is passed to our analytics service via Event Hooks, stored and then graphed as part of these pages

Key Meta-Lessons ¶

Outside of individual directory analysis we get overall meta-lessons that have vastly altered our marketing approach.

1. Google still has some muscle ¶

Google indexed new directory listings far faster than Brave, based on our checks across the same dataset. Seeing pages appear in Google within a few days was relatively common, while competing indexes were consistently further behind.

There is plenty to criticise Google for, but its search index remains extremely effective. In every like-for-like comparison we ran, Google discovered and indexed new listings first. They’re not out the game yet.

2. Chaos is the norm ¶

After examining hundreds of directories, it became clear how inconsistent - and occasionally deceptive - directories are. Some directories were near-identical copies of other directories had vastly different domain ratings. Others formed networks of almost interchangeable sites, all feeding demand into their own paid multi-directory submission service.

Some directories accepted submissions but never published them. Others contained clearly fabricated product records, including supposed launch dates for products that did not exist at the stated time. The results ranged from genuinely useful services to completely egregious nonsense.

Know what you are buying. Don’t be swayed by large numbers, manufactured reach or impressive-sounding sales claims. A smaller, verifiable audience is worth more than a vast one that may not exist and is too large for you to verify.

3. Smaller, targeted distribution beats PPC ¶

Every paid listing we tested delivered a lower effective cost per visit than comparable PPC advertising in our market. Those listings also included benefits such as permanent backlinks, newsletter distribution and continued discoverability.

We have run PPC previously. It was enormously expensive and produced less lasting value than spending a few hundred dollars on carefully selected directory listings. For our purposes, targeted newsletters, sponsorships and relevant listings now make considerably more sense than buying temporary clicks.

Whilst PPC can reach someone at the moment of need, that’s still a costly expense against a continuous stream of value discoverable by humans but also bot accessible.

Do bots click adverts? May be something we look into in the future. As of this moment in time we’re bearish on PPC for a few reasons and would recommend the first few hundred bucks of your new marketing budget goes to direct placements in directories, newspapers, online editorials and more than ad-hoc programmatic placements.

4. Most directories are bot fodder ¶

Most directories exist to secure backlinks and search indexing. Their role is closer to algorithmic box-ticking than conventional marketing.

If bumping your domain rating is your objective, sure, use directories. If you expect human visitors, you’ll be disappointed.

Be realistic, if a directory doesn’t look like somewhere you would browse yourself - and it clearly doesn’t show that anyone else is using it - then it’s going to be a bot play. Use your instinct.

5. Human-centred launch and competition directories perform differently ¶

Adding a Product Hunt-style launch, voting or competition process completely changes both the character of a directory and the traffic it produces. These platforms give people a reason to visit, discover products, participate and return, rather than merely hosting static pages for search engines and AI systems.

Some directories genuinely foster this behaviour and generate measurable human traffic. They should be evaluated differently from passive catalogue sites: their value comes from an active audience and the opportunity to earn attention, not merely from securing another backlink.

If your audience is other founders or folks likely to be built into these sorts of directories, then go for it. It’s worth putting some time in here.

6. Email distribution beats website listings ¶

When we used 10words’ paid service, almost all attributable traffic came from newsletters sent on Monday and Friday. Truly Analytics was featured directly on the website’s homepage, yet effectively nobody browsed through from there. The importance of email is widely discussed, but this provided unusually clear, measurable evidence: the newsletter generated the audience; the website listing did not.

7. Even email traffic contains bots ¶

Email produced the most human visitors, but it wasn’t all human. Following the 10words newsletter, we recorded hits associated with Microsoft Azure data centres around the world with a direct one to one relationship across a number of data centers - every major Azure datacenter hit up our product page once and only once.

We can’t definitively determine whether these were distributed security scans, proxy activity or another automated process although it does look like that. What we can prove is that email campaigns generate automated traffic alongside human clicks. Just cause it’s email, doesn’t mean that it’s bot free.

8. Bots increasingly represent human activity ¶

In 2026, bot traffic should not automatically be removed or dismissed. Cloudflare stated that in 2026 bot traffic surpassed human traffic, Truly Analytics supports this. We see website and other traffic being majority bot.

However, some of those machines are acting directly on behalf of people.

With our Network Intelligence we directly see when an iPhone user appears through Apple infrastructure because of privacy relay technology. The network says bot, the person in front of us is just using their iPhone.

Likewise, AI systems such as ChatGPT and Claude can retrieve pages in response to a human question, with most ChatGPT requests coming through Cloudflare WARP and Claude coming from Anthropic’s network ranges. The request is from a bot, but that traffic is coming from a human sitting at the chat window.

9. Modern marketing now serves machines and humans ¶

Digital marketing increasingly needs two connected layers: machine-accessible distribution and human-directed conversation. Directory listings are primarily a machine-discovery play, ticking the box for search engine backlinks and providing a loose capture net for AI processing and search.

Human interactions cover case studies like this one, podcasts, real-world conversations and direct relationships.

The machine-facing layer should be structured, scalable and easy to process. Expensive time and personal effort should be concentrated where genuine human interaction occurs. Machines aggregate the information, but people still make the purchasing decision.

As the people are behind the bots, you need to talk to both. You want to be available for mass bot usage with every perspective of your product or service discoverable. After that though you need content that a human with their diminished attention span, painkiller requirements and more whimsical tastes can process.

10. AI systems appear to use different indexes ¶

This is supported elsewhere and our evidence corresponds, strongly suggesting that AI labs are not relying exclusively on the search indexes available through Google, Brave or DuckDuckGo. In some cases they go out and find the data directly making their behaviour quite different to a search engine. With the traditional search engines you’re either indexed or not. With AI, who knows what’s happening.

11. AI retrieval changes the required scale of online presence ¶

AI systems can inspect far more pages and sources than a person conducting a conventional search. That adds to the job of website content: your goal isn’t to deliver a single 2 line elevator pitch. You need to deliver that pitch and cover the many different angles, variations and aspects of that pitch across a broad base of content.

As an AI agent can hit up hundreds of pages based on one query, the elevator pitch approach is only relevant for the human crowd. A human will give 30 seconds to a handful of pages at most. An AI can cover thousands of lines covering the same product or solution in seconds.

Hand in hand with this, directory distribution increases the chances of discovery across different indexes, while the company still needs to hit that elevator pitch. This requires more scalable content, repetition of important claims across appropriate channels, and far broader topic coverage than the landing pages of yesteryear, whilst you still need to provide those landing pages to accommodate the odd human visitor.

12. Brave’s slower index still matters ¶

Brave indexed new directory listings significantly more slowly than Google. Whilst we will quite happily dunk on Google its search index remains remarkably fast. However, indexing speed isn’t everything.

We included Brave’s Search because they provided an API which made it trivial for our systems to check the index. Meanwhile Google has withdrawn its search APIs and leaves automation dependent on scraping or browser tooling.

If humans are moving to agents then they’ll use AI Lab search and APIs like Brave’s Search API. Google has rendered smaller search engines relevant by making their service harder to integrate with.

13. Distribution is pay-to-play ¶

We are an engineering company first. Building solid, reliable, beautiful, operational systems is our heritage. Sales & marketing were always valuable but with large scale experiments like this the difference between engineering and sales is stark.

You can have excellent products for free. You will almost never have excellent digital marketing for free.

Every free directory without any barrier was tapped out years ago. Brownbook is filled with millions of listings and new listings every minute. There is no chance that backlinks get indexed.

Other directories placed free products into queues of months to years. The demand blows away any chance of value. It’s a real time tragedy of the commons.

The best performers are the pay to play options. This could be cash, or it could be something more involved such as Product Hunt which requires a wait period and involved launch to get something out of it.

The good news is that a small cost can get you quite far, so whilst it’s better to have some budget than no budget, some budget can still perform well.

14. A structured free tier can create enormous value ¶

Free listings were a clear differentiator between highly successful directories and weaker competitors. Anything purely free was a mess, however, something that was free with requirements managed to made the barter system work for them.

For example, some directories exchanged listings for reciprocal backlinks at scale, building domain ratings that rival or exceed those of major international publications. Their free tier was a scalable acquisition and authority-building system, a payment that didn’t require monetary exchange.

The reality is that charging money immediately and dramatically reduces the available market, especially among users in countries with lower purchasing power or limited access to international payment systems. A free tier can extract value from that larger audience potentially without compromising the underlying business. Some users cannot pay but can they provide a backlink, review, testimonial, testing, referral or case-study participation?

This is a question for every product owner - what can you get from your users that isn’t cash? We’ve seen directories build substantial assets around a free tier of users.

Where We're Going

With the above data we’ve adjusted our strategy. Here’s the new playbook:

We market to both bots and humans

• Targeted human interaction and outreach
• Large scale machine accessible content covering every perspective of our product that we can

We prepare continuous budget expenditure for marketing

• As new opportunities come up we’re going to need budget for the paid option
• We ignore the free, focus on the paid yet competitively priced
• We accept cost-free marketing but budget for time, effort or advisory. There’s no free lunch

We look to utilise a cost-free userbase

• We have a valuable product and service, not everyone can pay
• Devise a strategy that allows for meeting our non-monetary goals in exchange for serving that audience that can’t or won’t pay

Email is real

• We don’t have an email list, we should probably get one
• Aside from the common reasons, they’re just clearly better engagement. We don’t need to build the list with the end goal of a big list, it’s just going to be better connection with those users immediately

Hit up multiple things

• Who knows what works. Some of these directories are no different to others yet the difference in domain rating, engagement and pricing is all over the place
• We don’t have to be the best, or even good. We just need to be there. Some of these directories are awful, yet they exist and they presumably make money

Determine your future with Truly Analytics ¶

Truly Analytics is built for teams that need clearer analytics visibility, privacy-preserving measurement and better attribution evidence.

It can run alongside an existing Google Analytics setup and includes a free 3-day self-service trial. No credit card, payment details or sales call are required.

Everything in this case study was only possible with Truly Analytics.