Is your organic traffic eroding? - **Over the last few years, changes in Google's Search Console,** AI chatbots, and the use of social media as a search engine has caused major changes in organic traffic. Here's how to analyze, address, and increase
Is your organic traffic eroding?
-
Over the last few years, changes in Google's Search Console, AI chatbots, and the use of social media as a search engine has caused major changes in organic traffic. Here's how to analyze, address, and increase your organic traffic if it's suffering.
-
Email deliverability and outreach is becoming increasingly difficult for founders. This guide can help you get your emails seen, using a data-driven, personalized approach.
-
Founder and college student Yasser Elsaid hit $64,000 in monthly revenue with Chatbase, a tool that lets you build a custom ChatGPT interface for your website. Below, he shares how first mover advantage helped his growth skyrocket, and what concerns him most about building in the AI space.
Want to share something with nearly 75,000 indie hackers? Submit a section for us to include in a future newsletter. —Channing
📉 Is Your Organic Traffic Eroding?
by Kevin Indig
Over the last few years, increases in ads, search engine result page (SERP) features, and direct answers have pulled clicks away from organic results, but Google doesn't make it easy to understand these changes.
Also, as more users interact with AI chatbots and platforms like TikTok to find what they're looking for, Google has announced that it will make widespread changes to Search. Measuring the impact of SERP layout changes on organic traffic is complex. As a result, we often don't know the right way to turn negative traffic trends around.
Here's how you can get ahead of the curve!
What is SEO erosion?
SEO erosion occurs when organic traffic for keywords declines over time for a reason that's not related to an issue with the website. An erosion analysis focuses on understanding three patterns:
- Organic traffic moving to paid.
- Organic traffic declining due to non-paid reasons, like new or better SERP features.
- Organic traffic decreasing due to fewer searches (market pullback).
The example in the screenshot below shows how organic rank was stable over the last 12 months, but clickthrough rates (CTR) dropped:
The same view with impressions instead of rank (since we know it's flat) shows how impressions grow, but CTR remains unchanged. More people see the result, but don't click, because something else seems to deliver a better result:
Reasons for organic traffic erosion
- Top and Shopping ads.
- SERP features, like Featured Snippets, Instant Answers, Knowledge Panels, image carousels, video carousels, or local packs.
- Query refinement bubbles take users down the funnel faster and onto Google, instead of onto other sites.
The tricky thing about SEO erosion is that its impact is sneaky. You don’t notice it until your SEO revenue is down, and you wonder what happened.
Erosion has become especially difficult to catch over the last three years, because the pandemic completely disrupted user behavior. Many keywords have seen way lower search volumes over the past three years because the pandemic pulled demand forward.
How to run an erosion analysis
These red flags will indicate potential organic traffic erosion:
- If the organic position hasn't changed, but organic traffic has declined, it's not an SEO issue: It's either ads or SERP features.
- If organic traffic or conversions have declined, while paid traffic has increased, it's likely a shift from organic to paid.
- If organic traffic declined and paid traffic was steady or declined, it's likely an issue of SERP features.
We can use traffic data from Search Console and Google Analytics (or your web analytics tool of choice) to define and validate hypotheses for eroding organic traffic.
Search Console data itself is already helpful for understanding where you lost traffic without losing ranks:
But we can take it a step further. In the screenshot below, you see an example of an erosion analysis with changes in organic rank, traffic, conversions, and revenue, as compared to the previous year and paid traffic. It allows you to quickly spot patterns of pages that lost organic, but gained paid traffic, or lost organic traffic with stable ranks:
Building your analysis
The first step to finding out whether your organic traffic is eroding or exploding, and creating a custom click curve for the last full month. You'll be comparing it year over year. I've extensively described it in "The importance of custom click curves."
An erosion has occurred if your click curve shows lower click rates compared to the previous year. However, click curves can't tell you why, just that something has changed. You still need to figure out the driver, but be mindful that there might not be a single root cause.
To get closer to the main driver, you can use traffic data to narrow your hypothesis down. I'm using Google Analytics in combination with Search Console to describe the specific steps, because it's the most used web analytics tool out there. However, you can recreate the analysis with any tool.
In Universal Google Analytics:
- Go to Behavior → Site content → Landing pages.
- Add segments for organic and paid traffic.
- Select the main conversion goals as "Event." This requires you to tag and define your main conversion, and give it a value.
- Export.
Next, get organic ranks for each page from Search Console. This step works best with the Search Analytics for Sheets integration because you get more data, but you need to run separate pulls for each year, since the extension has no comparison mode. Make sure to select the same time period as in Google Analytics.
In the last step, use VLOOKUP or INDEX MATCH to match Google Analytics and Search Console data by URL, and calculate the delta between the last month and the same month last year. You might have to remove the protocol (https://) and root domain from the Search Console data to match Google Analytics data.
Sort the sheet by negative organic revenue or conversions, and look at the columns for rank and traffic to define your hypothesis.
To validate your hypothesis, isolate pages that dropped in organic traffic, and look at SERP screenshots for queries that lost the most clicks, CTR, or impressions. You can get this in Semrush under "Position Tracking" or "Keyword Overview," and in Ahrefs under "Keyword Explorer" (compare SERPs between two dates).
The more you narrow your view down, the sharper it becomes. There are several ways to segment your data:
- Compare one full month with the same month of the previous year.
- Narrow your view down to a subdirectory, or even a landing page.
- Look at data from only one country and device.
- Keep in mind that Search Console filters traffic out, especially if you have many URLs. Make sure to set up separate Search Console properties for subdirectories.
- Filter out brand queries.
Prepare for more organic traffic erosion
AI chatbots and the fragmentation of Search put a lot of pressure on Google. I expect more organic traffic erosion due to new SERP features, more people using AI chatbots to get answers, and more users leaning on vertical search engines like Amazon, Pinterest, or Booking to find better results.
To keep track of the changes, I recommend running erosion analyses at least every six months, storing rank and traffic data in a data lake, and visualizing changes over time.
Here are more resources about the impact of SERP features on organic traffic:
Have you checked your organic traffic erosion? Let's chat below!
Discuss this story.
📰 In the News
from the Growth Trends newsletter by Darko
💬 You can now edit your WhatsApp messages.
✅ LinkedIn is rolling out new verification and anti-scam features.
💻 Prepare to test Privacy Sandbox APIs with Google's ads platforms.
🏛 Facebook has been slapped with a record $1.3B fine over EU-US data transfers.
🦸♀️ Check out these six marketing superheroes, and what you can learn from them.
Check out Growth Trends for more curated news items focused on user acquisition and new product ideas.
📧 Get Your Emails Seen With This Data-Driven Approach
from the Hustle Newsletter by Susie Ippolito
The Signal: Tyler Cook is a founder who earned his chops as a young, ambitious door-to-door salesman. This helped him become a data-focused email deliverability expert who never loses sight of the human recipient.
At a recent Trends event, Tyler shared how he gets emails into inboxes, a task that is a growing concern for founders.
*Six-month rolling average. Source: Google Trends
Here are six key takeaways!
1. Technology for the health of it
It may sound strange to consider the "health" of your email, but just like your arteries, you want to keep things clean and flowing smoothly.
Tyler's recommended tech stack consists of tools to keep your domain, email list, and email content in tiptop health so that they land in the inbox.
2. You've got a rep to protect
Much of the energy you put into email management should be spent building, maintaining, or rehabbing your domain's reputation.
For the best shot at email deliverability, Tyler's advice is clear:
Your goal is to maximize the amount of positive signals tied to your sending domain.
Positive signals include:
- Domain health.
- Sender email health.
- Audience engagement.
*Review your checklist at regular intervals to ensure the health of your domain. Source: Tyler Cook
3. Let's get personal
Random views on TikTok will not earn your brand the kind of loyalty you can get from an engaged email audience.
Tyler describes email as a brand-owned channel because:
...people willingly give you a way to address them directly.
This comes with a brand responsibility to nurture those relationships differently and more intentionally than we do on social media.
Tyler suggests using engagement signals to segment the audience. This provides a personalized experience that reflects the customer's unique journey.
*Segment your audience by engagement to provide a personalized experience. Source: Tyler Cook
4. It's all in the flow
A warm, intentional welcome flow isn't just friendly, it also optimizes email deliverability. Tyler says:
[This] is the best time to capitalize on the responsiveness of new leads, and prime for future engagement.
The flow below will signal to email servers that you are genuinely interested in your audience, and that they are equally interested in you.
*Source: Tyler Cook
5. Rehabilitate an unhealthy domain using positive signals
It's not enough for your emails to appeal to your subscribers. You must also please their email servers by keeping your domain healthy.
If your emails are not landing in inboxes, it's time to rehab your domain. Tyler suggests using the following segments to generate the most positive signals in the shortest amount of time:
- Segment one: Anyone who has purchased something in the last 30 days, and engaged in the last eight days, are your super signals!
- Segment two: Anyone who has responded to an email in the last nine to 30 days.
- Segment three: Anyone who has engaged in an email (such as clicking a link) in the last eight days.
Pro tip: The key to success is to send each segment email 3-4 days apart to avoid ruffling the feathers of sensitive email servers.
6. Keep it clean
A clean email list is a productive one. Don't just delete inactive emails; stay close to the data so you know when to reengage stale audience members.
In Tyler's experience, the reengagement triggers should go off around the 45 day mark.
But don't go writing up a reengagement flow if they're already ignoring your emails. Connect with them in other places, like SMS, social media, or even via direct mail to recapture their attention.
As a door-to-door salesman, Tyler tracked every "no" he got, along with the reason for the objection, and how to overcome it in the future.
He plotted his success by tracking data, and continues to do so today, to reach millions of people with just a few knocks.
Will you implement these tips into your email strategy? Share below!
Subscribe to the Hustle Newsletter for more.
🧠 Harry's Growth Tip
from the Marketing Examples newsletter by Harry Dry
If your product is unique, explain it as simply as possible.
Go here for more short, sweet, practical marketing tips.
Subscribe to Marketing Examples for more.
🚀 Yasser Elsaid Hit $64K MRR in Six Months
by Channing Allen
Just a few months ago, Yasser Elsaid was in his fourth year of college, interning at FAANG companies like Meta and Tesla. When he placed his side hustle in front of the tidal wave that is OpenAI's ChatGPT, the timing just worked out.
His bootstrapped chatbot company, Chatbase, is already doing a mind-blowing $64K MRR, and traffic hasn't dipped since the initial launch. He just hired two employees to help with the overwhelming demand. And no, he didn't pass all of his university classes...nor does he care.
I reached out to Yasser to get the story behind this whirlwind growth!
Starting Chatbase
Chatbase is a tool that lets you build a custom ChatGPT interface for your website.
I'm originally from Egypt, and I moved to Canada in 2019 for undergrad. Back when I started university, I had no idea what I wanted to do. I was surrounded by many students who had the goal of working at a FAANG company, so naturally, I adopted the same goal and started working towards it.
I was able to get internships at Tesla and Meta during my third and fourth years of school. At the start of both internships, everything was super exciting. Free food, massage chairs, smart people...I couldn't find anything to dislike. But as time went on, I saw myself having the exact same day every day. It wasn't a bad day, but it wasn't exciting. I realized that the lifestyle at these big companies is too safe and too manicured for me. Those internships taught me a lot of things, the most important being that I wanted to become an indie hacker.
I always worked on side projects. None of them made any real money, but listening to success stories on the Indie Hackers podcast was enough proof that I could make something work. As my time in university went on, I started to focus less on optimizing for FAANG (leetcode), and more on building side projects.
I realized that I would be working hard no matter which path I chose, and the solo founder path sounded the most exciting. I was inspired by seeing folks like Pieter Levels build cool AI products, and I started playing around with the OpenAI API. I saw a lot of potential!
The idea for Chatbase wasn't super hard to come up with. I didn't do any validation or market research since this space was just being born, and it was obvious to me that "ChatGPT for your data" would become a thing that many people would find useful.
Product dev and tech stack for Chatbase
Chatbase actually started as a "ChatGPT for your PDFs" tool. It was the easiest and most obvious use case. You just uploaded a PDF, then you could chat with its content. The MVP for this version took me about two months to build. I tweeted it to my 16 Twitter followers on February 2, 2023.
My tweet went viral. I didn't even have a pricing page yet. Anyone could upload as many documents and send as many messages as they wanted. As soon as I saw the traction, I stopped doing anything university-related, and focused 100% of my time on Chatbase. I knew I couldn't mess up this opportunity.
I ended up failing two classes, but it was worth it!
I'm using React, Next.js, and Supabase to build the web app. I'm also using OpenAI's API, Langchain, and Pinecone for the AI part of the app. And finally, I'm using Stripe to handle payments.
This stack has served me well so far, but I think I might need to change a few things in the future to save on costs. I might experiment with different Vector databases or hosting solutions.
Marketing and user growth
Chatbase was probably the first "ChatGPT for your data" SaaS tool. This was a huge advantage, since it wasn't that hard to make it go viral. It had the "wow" factor of being able to upload your own personal documents and chat with them, plus it took advantage of the overall AI and ChatGPT hype.
I also had a few tweets that went viral, and AI influencers started posting about it.
This viral marketing was crucial in getting my first batch of customers, but the problem is that it's not a good long-term strategy. Now there are so many tools doing what Chatbase does, so AI influencers are not likely to post more of the same tools.
Here are other methods I've used to keep getting customers:
- Launching on Product Hunt.
- Posting to relevant subreddits: A story works better than "here is a tool I made."
- Being active on Indie Hackers.
- Submitting Chatbase to AI directories.
My traffic hasn't dipped:
Business model and revenue growth
Chatbase is a subscription service with four different plans. Most subscribers are on the lowest plan (mostly to test the chatbots), but they quickly move to the higher plans or churn. Most of the revenue is generated by the $399 monthly plan.
Chatbase uses Stripe to handle all the subscriptions, and it's working well so far. Here's my revenue growth from the past 30 days:
This crazy MRR growth is due to being really early in the market, plus taking advantage of all the hype around ChatGPT. Not sure how long this will last, since now there are similar tools, and bigger companies are also interested in the space.
One piece of advice for other founders is to find a space with some competition. If there is no competition, you are either a genius, or no one needs what you're building. If there is a lot of competition, it will be a race to the bottom.
The future
Chatbase will continue to offer "ChatGPT for your data" as a service. I'll continue adding more features based on customer feedback. I see it being the easiest way to create a custom AI chatbot for your data that you can interact with through the website or the API.
I won't be looking for a new grad job any time soon! I just added two employees to help with coding new features and customer support, and I imagine the team can grow to five or six people.
At this point, my main worries are the increase in competition, and not knowing what the future will look like in this space. Since all of this is very new, something could happen one day that may change the trajectory of all these AI tools.
The biggest challenge was convincing myself to work on side projects with no immediate return. Once I started doing that, it became very hard to lose. Even if Chatbase doesn't succeed, I'll have learned a lot, had fun, made money, and will always have a cool story!
I learned that making $10K a month from a side project is not a crazy goal to have. People achieve it every day. Keeping up with new technologies and trends will help you find opportunities faster than anyone else. If you can take advantage of these opportunities, one of them will be a success!
Closing advice
- Launch as soon as possible. Skip creating a waitlist, and just create a very basic MVP. Then, launch it!
- Launch multiple times, if possible, especially when you make significant changes to your app.
- Build in public. There is no good reason not to build in public. Every new post you make has a chance of going viral and changing your life. Plus, you get to meet cool, like-minded people. You don't have to share revenue numbers. Just share what you're working on today. Twitter is still a good place to do this, but no one is doing this on TikTok now, so there is a big opportunity there.
- Engage with the community. Follow successful indie hackers, start creating content, share your learnings, and meet other indie hackers near you.
Discuss this story.
🐦 The Tweetmaster's Pick
by Tweetmaster Flex
I post the tweets indie hackers share the most. Here's today's pick:
🏁 Enjoy This Newsletter?
Forward it to a friend, and let them know they can subscribe here.
Also, you can submit a section for us to include in a future newsletter.
Special thanks to Jay Avery for editing this issue, to Gabriella Federico for the illustrations, and to Kevin Indig, Darko, Susie Ippolito, and Harry Dry for contributing posts. —Channing
