- TOC {:toc}
This, not so short, guide puts together years of one-to-one training and mentoring into a document. The purpose of this guide is to help writers do a great job, develop their career and allow us and other companies like us, to succeed in the challenging world of content marketing.
Imagine coming back from the shop with a new vacuum cleaner. You open the box and discover a collection of pipes and brushes. There’s also an accordion pamphlet with texts in a tiny font. Most of us will not bother with that pamphlet and figure out ourselves how to put together and operate our new appliance.
Both us (the client) and the vendor consider this as something of very little business value. It’s some necessary evil, including legal stuff that the vendor is required to disclose to protect from lawsuits, in case someone uses the vacuum cleaner in a wrong way and suffers damage.
But in a closer look, you’ll discover that this one piece of paper is the only opportunity for this vendor to connect with their client. The vendor can get quite a lot from positive interaction with their clients, including:
- A happy client who will recommend the brand to others.
- A happy client who will look for other products coming from this vendor.
- A happy client who will not go to the store to return the appliance, just because he didn’t understand how to use it correctly.
And the only “content” that’s available to that vendor is this pamphlet.
A great pamphlet will include:
- Clear setup instructions, in correct and concise language and a convenient font.
- Instructions for how to get help, if needed.
- Tips on efficient usage.
- Tips on smart maintenance.
- Instructions for how to sign-up to a mailing list for people like yourself.
Is this how most pamphlets that you’ve seen look like?
If not, it’s probably because the person in charge of this pamphlet didn’t consider it as a marketing opportunity, with significant business consequences. Many pamphlets look like they were written by a Chinese mechanical engineer, translated to English using Google Translate. The design appears to optimize the amount of paper, rather than the business outcome.
The same logic applies to any piece of content out there. If it has no business implications, it probably doesn’t need to exist.
When you write any piece of content, keep this story in mind. Look for what value you’re aiming to get out of someone reading your content.
In your first practice assignment, you'll identify the business value in content that most people dismiss as "technical" or "unimportant." This will train you to find the business angle before you start writing.
Readers decide fast. Research consistently shows that shorter content performs better:
1) Boomerang: 40M emails (2016)
Based on 40+ million emails, the “sweet spot” for email length is 50–125 words, with response rates above 50% in that range; response rates decline as emails get longer (e.g., ~44% around 500 words), and very short emails also underperform. This blog post is from 2016 and quotes email stats which you probably cannot replicate today, but the trend remains.
In a study using five versions of the same site (same info, different writing), the concise version (about half the word count) had 58% higher measured usability; “scannable” was +47%, and combining improvements was +124%.
Concise does not mean short.
It means:
“As short as possible, but not shorter than what the reader needs.”
Removing words that don’t change understanding is concision.
Removing words that carry meaning is oversimplification.
A simple test:
- If deleting a sentence doesn’t change the decision the reader will make, delete it.
- If it does, keep it — even if the text gets longer.
Edit with intent:
- One idea per paragraph.
- Delete explanations of the obvious.
- Replace phrases with single words.
- “In order to” → “to”
- “At this point in time” → “now”
- Remove warm-ups. Start where value starts.
- Cut 20–30% after writing. Then reread for clarity.
If clarity survives, the cut was correct.
Some topics are naturally complex. They require longer explanations to be understood correctly.
If you’re explaining:
- A complicated concept
- A non-obvious workflow
- A system with many moving parts
Then your job is not to shorten. It’s to explain well. A complete explanation is still concise. It’s long, because the topic requires a long explanation, but concisely.
However, there’s an important signal here:
If explaining your own product requires long, careful instructions, you’ve likely found a usability problem.
When that happens:
- Don’t “solve” it with more documentation.
- Find the right person (designer, product owner, developer).
- Explain where users get confused.
- Push to simplify the product.
Documentation should clarify reality, not compensate for complexity.
Long explanations are acceptable when complexity is unavoidable or as a temporary solution, while your company is simplifying a too-complicated product.
They are a warning sign for complexity that needs to be handled.
In your practice assignment, you'll take successful long-form content and cut it substantially while preserving its core value. This will develop your ability to identify what's essential versus what's padding.
We like to think that people read carefully, weigh facts, and make rational decisions.
They don’t.
Most decisions are made quickly, with incomplete information, and under emotional pressure. Prospect Theory (introduced in Thinking Fast and Slow), explains how people actually decide when there is risk, effort, or uncertainty involved.
Losses hurt more than gains feel good, at a ratio of about 2:1. We go out of our way to avoid losing what we already have and we don’t mind missing a potential gain.
| Bad example (nothing to lose) | Good example (loss is clear) |
|---|---|
| Use the Translation Dashboard for easier translation management. | Avoid wasting hours finding content with outdated translation by using the Translation Dashboard. |
| Renew to get updates for another year. | Don’t let WordPress updates break your site. Renew to get critical updates ahead of WordPress releases. |
When possible, describe what the reader avoids losing, not just what they gain.
When people are unsure what will happen, they delay or do nothing.
| Bad example (many unknowns) | Good example (predictable results) |
|---|---|
| Enable Translate Everything | Enable Translate Everything - your total cost is $123; existing translations don’t change. Disable to return to manual translation management. |
| Buy Now | Buy Now - pay $99 today. Auto-renews every year; Cancel renewals anytime. |
Before asking for action, remove uncertainty and define the escape hatch.
We have no way to evaluate things in absolute terms.
Is one second a lot of time? For trees in the forest, it feels like nothing. For CPUs, running at gigahertz, it feels like a lifetime.
Is $100 a lot of money? It is if you’re paying for a metro ride. It’s very cheap when you’re shopping for a house to buy.
We latch onto the first piece of information that we get and compare what comes next to it. Good anchors set the expectations that your actual offer will beat.
| Bad example (no anchor) | Good example (clear anchor) |
|---|---|
| [ Sign-up for $49/month ] | [ Sign-up for |
| Try our steak lunch for $99. | Chicken: $79 Steak: |
Understanding complex topics requires a lot of cognitive effort. We look for shortcuts.
Instead of describing all the details of a car, manufacturers work hard to associate their brands with what they want them to mean for us. For example:
- Mercedes → Quality, elegance, accuracy, prestige, expensive
- Porsche → Excitement, speed, expensive
- Toyota → Quality, efficiency, convenience, affordable
Instead of trying to explain a complex topic from scratch, look for the mental shortcuts that will allow your readers to understand it, in fewer words and with much less effort.
| Bad (all from scratch) | Good (uses known elements) |
|---|---|
| Our system performs continuous background synchronization between repositories, detects string-level changes, and automatically updates translation states. | Think of it like GitHub Actions for translations. |
| We offer translation quality that matches professional human translators while costing little and delivering translations quickly. | Human-quality translation at machine speed and cost. |
When using mental shortcuts through analogies or references (like "GitHub Actions"), consider whether your audience will actually understand the reference. If you're writing for WordPress developers who've never used GitHub Actions, the shortcut becomes an obstacle. The reference should simplify, not require additional explanation.
In your practice assignment, you'll apply these four behavioral psychology principles to improve actual content. You'll explain the reasoning behind each edit, which will develop your ability to diagnose why content isn't working and how to fix it.
Now that you know how people think, it’s time to identify the actual people that your content is targeting. Who are they? What’s on their mind? What do they want to get by reading your content? By answering these questions, you’ll know what to write so that your target audience wants to keep reading.
You should not invent these answers. You should derive them from observation, support tickets, sales calls, or direct experience. Different answers to these questions will lead to major changes in your content.
A single piece of content often serves multiple audiences with different needs. Let's take the vacuum cleaner example from the first chapter to illustrate this.
The target audience for the vacuum cleaner pamphlet includes:
- The person unboxing it and assembling the cleaner
- The person using the cleaner
These might be one, but very likely, they will be two different people. In many cases, a wife will be using the vacuum cleaner and the husband will be tasked with unboxing and assembling it. We’re making stereotypes here, for a good reason. You’ll need to make some generalizations and stereotypes when you define the target audience for your content.
Be careful about over-generalization. A definition of target audience such as “busy people, Asian, 35-45 years old” is not going to help you write effective content.
Here is a much more useful definition of a target audience:
“Python developer, familiar with object oriented programming, researching high-traffic servers, without the authority to make payments, speaking fluent English.”
The first definition (the busy Asian) of a target audience looks nice, superficially, but doesn’t really tell you how to explain anything to this audience. With the second definition (the Python developer), you’ll know what not to explain (because he surely knows it already), what to focus on, what to aim for (try a free tool) and what not to bother with (pay).
Going back to our vacuum cleaner pamphlet, here are two target-audience definitions that should work:
- Assembler: “Has experience assembling simple devices and to use a screwdriver; Follows instructions when they are clear; Wants to spend as little time as possible reading and assembling.”
- Cleaner: “Has experience cleaning floors and furniture; Understands the importance of using the right part for each surface; Will skip lengthy instructions.”
With these definitions in mind, you will avoid writing extensive introductions about why it’s important to follow the assembly instructions or why you should use the right extensions for floors and sofas. Instead, you’ll offer clear step-by-step assembly instructions with short texts and clear graphics. And you’ll include a detailed list of surfaces and the corresponding vacuum heads that work best for each. This is obvious when talking about the vacuum cleaner example, because we all know what that pamphlet should ideally include. Use the same thinking when writing much more complex content.
The unboxer and assembler is looking for easy-to-follow instructions, so that (probably) he can get done with it as soon as possible. If I was responsible for the assembly, I would be looking for visual instructions, clearly labeled, allowing me to get back to whatever I was doing quickly, and not being called back for tweaks.
The vacuumer is going to use the machine many more times and for longer. She (or maybe he), will benefit from understanding the differences between the different parts, to make the actual cleaning process easier and more efficient. The person doing the cleaning will probably be in charge of maintenance. Using a poorly maintained machine is worse than learning how to clean it properly. If I’m vacuuming and maintaining it, I’m looking for a detailed list of what each brush is good for (pictures would be wonderful) and very clear instructions on how to take it apart, clean the parts and even put it back together. From past experience, I lose more time fighting with the parts than reading instructions. So, detailed instructions would be great.
In the case of the vacuum cleaner, it’s straightforward. We want to know how to use the machine.
Other pieces of content are harder to figure out “why is someone going to read this?”. We might be looking to satisfy our curiosity, we might be looking for entertainment, we might be looking for a solution to a specific problem. Think about it and you’ll find why someone might want to read your content. Don’t start writing before you’ve figured it out.
In your practice assignment, you'll compare two pieces of content - one that addresses its target audience effectively and one that treats readers generically. This will sharpen your ability to recognize when audience understanding is driving content decisions versus when content is written from the company's perspective rather than the reader's.
Good content always accomplishes a goal. But content doesn’t accomplish its goal if the right people don’t know of its existence; or don’t choose to read it; or start reading it and abandon (bounce); or don’t trust what it says; or don’t take the action that you intend. If any of these happens, something along the funnel is broken and the content isn’t working.
Before you start writing, you need to understand the specific funnel, or funnels, for your content. It’s fine that content will have multiple funnels (like traffic coming from Search and traffic coming from a link on the site’s homepage).
In the case of our vacuum cleaner pamphlet, the right people are very likely to notice the content. But will she choose to read it? Will she read enough to understand how to assemble, use and maintain the machine? Will she trust that the instructions are for her? Will she decide to take action and assemble it, or open the support QR before returning the product to the store?
And, most funnels are much more complicated than the vacuum cleaner pamphlet.
Every funnel has predictable stages where you can lose readers:
- Discovery: Your content appears when the right people are looking
- Selection: People choose your content over alternatives
- Retention: Readers stay engaged rather than bounce
- Trust: Readers believe what you're telling them
- Action: Readers take the step you want them to take
Problems at any stage break the funnel. A common example: someone searches for "how to translate WooCommerce product descriptions," clicks your result, sees a page about general multilingual setup, and leaves immediately. The funnel broke at stage 3 (retention) because your content didn't match what the search preview promised.
Different metrics tell you how each part of the funnel performs:
- Bounce rate: The percentage of visitors who leave immediately without any interaction. High bounce usually means your content doesn't match what people expected when they clicked.
- Time on page: How long readers stay. Very short time with high bounce means the content didn't match expectations. Moderate time with low conversion might mean the content is unclear or doesn't lead to action.
- Click-through rate (CTR): The percentage of people who see your content in search results and click it. Low CTR means your preview doesn't appeal to searchers.
There's no universal "good" or "bad" for metrics. Context determines whether a number indicates success or failure.
For example, a 5% click-through rate (CTR) from search results would be excellent for a general topic like "best summer trips" ranking in position #9. But a 30% CTR for your brand name search when your homepage ranks #1 would be terrible. It means that 70% of people searching specifically for you aren't clicking your result, which signals a serious problem with your title and preview text.
Similarly, 40% bounce rate is good for a troubleshooting guide (people found their answer and left) but bad for a landing page (people left without taking action).
You'll learn to access and interpret these metrics through your practice assignments and feedback. For now, understand that metrics don't tell you what's wrong. They tell you what to measure.
Below are several examples of content funnels. There are many others that will apply to different projects that you’ll be working on. Use them as examples. Each case deserves its own analysis, but the basic questions to answer are similar.
When we think of Content Marketing, search-engine traffic is what comes to mind. People search for something, they see different results (possible answers), choose one that looks the most promising, read some, and “if you’re lucky” take action. Of course, we already know that luck has little to do with the performance of your content.
In Google (which accounts for the vast majority of searches), your content will include a subject and a preview text. That’s it. According to these two pieces of information, along with the position of your page in the search results, the people searching will decide if they click into your content or somewhere else.
People searching are already looking for something. Good search results will provide the answer for what the person is searching for. Bad search preview will ignore the person searching and focus on what the business wants to say. Google will compose the search preview from your content. Make this easy for Google and increase the reader’s trust by opening the content with exactly what you’d like to appear in these short search results.
The Bounce Rate for content is the percentage of people visiting your content and immediately leaving without taking any action. Google goes deeper. It also checks the amount of time readers are on your page. When people bounce quickly, the reader says that “this is not actually what I was looking for”. This should be a signal for you, that you’re either targeting the wrong audience or your content doesn’t look like what people are searching for. Google wants to deliver the right options to people searching, so when it sees high bounce rates for your content, it’s going to show your content to much fewer people.
We get a lot of stuff in our inboxes. Actually, much of the email heading our way doesn’t even reach our inbox. Our email provider will decide for us that it’s SPAM or promotion or something else that we normally don’t care to read.
When you’re writing content where the “discovery” is via an email, you need to make sure that:
- It actually appears in the inbox
- The subject line (alone) makes people want to read the email
- The email appears to be digestible (most of us don’t care for long stories in our inbox)
- The email is relevant to the reader
- The email asks for a clear and reasonable (to the reader) action
The entire content might be in the email. The content might be part in the email and part online (like a landing page). The logic is the same. Every piece is a step in a funnel, where you’ll lose the reader if done wrongly.
Usually, we’re not sitting and waiting to consume advertisements. Advertisements interrupt us from whatever else we were doing and try to draw our attention to something that the advertiser wants to promote. Content that appears in ads needs to fight uphill for the reader’s attention, because that reader was engaged in something else when the ad showed up.
All the principles that we explained until now are valid for any type of content, but they’re especially critical for advertisements, because the reader’s attention and patience is much thinner.
Writers cannot be experts in everything. They need to learn before they write. We prefer first-hand knowledge, which should come from actually using (as an expert) what you’re writing about. We don’t compile material from other sites to produce our content.
Your purpose, when learning a product is to really know it, inside and out. Start by learning what the product is intended for. Then, set up a real project, where you’ll use the product to accomplish something similar to what real users will accomplish. Of course, it’s going to be a lot of work and require a lot of background.
For example, if you’re going to write about a financial calculator app, you’ll need to know the topic before you learn about the product. If you’re not able or willing to learn about compounding interest, don’t bother learning about the financial calculator and don’t agree to write marketing or technical documentation for this project.
Remember that most products that you can document don’t cost millions of dollars, so clients expect to learn how to use them in a matter of minutes, maybe hours. If it takes you weeks to learn the topic, maybe you’re stepping into a project that it’s right for you.
Interviewing clients is a great (and critical) method to learn how clients feel about the product(s) that you’re writing about. By interviewing clients, you can understand how clients recall using the product. Keep in mind that clients are not maintaining an accurate and complete log of how they really felt when using the product. Clients will recall anecdotes. Typically, clients will recall one highlight (good or bad) and the overall result of using the product (they succeeded or failed). This means a single interview tells you how that one person remembers their experience, not how the product actually works or what most users need.
To get a comprehensive (not complete) picture of what clients feel about the product you’re writing about, talk with different people at different stages of using the product. Aim to talk with people very shortly after using the product. Take organized notes of everything that you hear so that you don’t need to rely on your memory. Always take notes of the context (who the client is, what the client is trying to achieve, the client’s experience, etc.). With comprehensive notes from many clients, you’ll be able to notice patterns.
Want to know how many interviews are enough?
You'll know you've interviewed enough clients when you notice two things:
- The issues repeat with new learning becoming very rare.
- You’ll notice clusters of clients with similar issues and highlights (for example, agency clients often need different things than one-off clients).
You will see that there isn’t a “typical client”. There are clusters of clients with similar backgrounds and needs. You’ll notice what clients value most in the product. What they’re missing. What confuses clients. With this information along with your own knowledge of the product, you can create effective technical documentation and marketing materials.
And, finally, remember that clients have their own agendas. Many clients will want to advise you, and help improve the documentation and the product itself. Keep notes of advice but understand them for what they are. You’re not receiving organized research material from someone who methodically went through the whole product and took organized notes. Often, clients provide advice meant to improve the product for their use case. Often, what’s good for one client is great for many others, but it’s not always the case. An expert client will advise you to omit “background information”, which is great for him, but will confuse beginners. By understanding that everyone has an agenda, you’ll be able to write material that helps many clients and not just the ones you’ve interviewed.
LLMs can summarize what others have written about a topic. This is useful for:
- Understanding technical terminology before hands-on testing
- Identifying what competitors emphasize
- Checking if your approach aligns with industry standards
Of course, this research is useful to orient yourself, quickly learn about industry terminology and get the “general idea”.
But never write content based solely on secondary research. If you haven't used the product or talked to clients, you're just rehashing what others said. That's not the content we want.
Remember our vacuum cleaner example from the first chapter? You wouldn’t want to read a “how to use this vacuum” pamphlet by an author who never assembled or used it, right? If the author only read and watched YouTube videos, but didn’t assemble the machine or never used it, you’d be wasting your time reading his pamphlet. Same for your writing, about much more complicated products.
But you're one person, and you won't encounter every issue that confuses others. So, before you’d write that how-to-vacuum pamphlet, you surely want to speak with different people who assembled and cleaned with that machine, to get the complete picture of what your pamphlet needs to include and highlight.
Writers must understand how readers actually interact with their content. On-page metrics reveal this interaction from day one, long before you can measure business outcomes like sales or subscriptions.
On-page metrics measure immediate reader behavior:
Click-through rate (CTR) from search shows the percentage of people who see your content in search results and click it. Low CTR means your title and preview don't appeal to searchers.
Bounce rate shows the percentage of visitors who leave without any interaction. High bounce means readers didn't find what they expected.
Time on page measures how long readers stay. Very short time with high bounce indicates a mismatch between what readers expected and what they found. Moderate time with low conversion to your desired action suggests the content doesn't guide readers effectively.
Scroll depth reveals how far down the page readers scroll. If most readers stop at 30%, your opening failed to engage them or they found their answer immediately.
Click-through rate (CTR) to your objective shows the percentage of people who continued from your content to the “business objective”. For example, if your content is the company’s homepage and the objective is to sign-up to a free trial, this CTR is the percentage of visitors who went to the sign-up page. Most content has such a “desired next action”, but some content (like an errata page) doesn’t necessarily have it.
On-page metrics appear as soon as readers visit your content. You don't wait months to see if content drives sales. You see immediately whether readers engage with what you wrote.
A page with high traffic produces meaningful metrics within hours. A page with sparse traffic needs days or weeks to accumulate enough data to distinguish patterns from random variation.
Check metrics as soon as you have enough visits to see patterns. For high-traffic pages, this might be the day after publication. For niche topics, wait until you have at least 100-200 visitors before drawing conclusions.
Google Analytics organizes data by page and date range. You select your page, specify the time period, and view the metrics listed above. GA4 groups metrics under "Engagement" for bounce rate and time on page, and under "Acquisition" for how readers found your content.
Matomo presents similar metrics with different terminology. "Bounce rate" appears as "single page visits." Time on page appears as "average time on page." The logic remains identical.
No universal threshold separates "good" from "bad" metrics. Context determines whether numbers indicate success or failure.
A troubleshooting guide with 70% bounce rate and 45-second average time might be excellent. Readers found their answer quickly and left to implement it. The same metrics for your product overview page would be terrible. That page should introduce readers to multiple features and guide them toward trying the product.
A pricing page with 3-minute average time suggests confusion. Readers should understand pricing quickly. A tutorial on translation memory with 8-minute average time might be too short. Complex technical concepts require time to absorb.
Remember the chapter on funnels? The source of traffic matters when you set performance goals. If it’s your own clients, clicking on a link in your newsletter, to read about new features in your product, you’d set different objectives than for search traffic.
Publish your content, then monitor these metrics within days or weeks depending on traffic volume. When metrics indicate problems, update the content.
High bounce rate with short time on page means your opening doesn't match what readers expected from search results. Rewrite your opening to immediately address what readers searched for. Update your title and meta description to set accurate expectations.
Low CTR in search results means your preview doesn't compel clicks. Rewrite your title to include the specific problem readers want to solve. Adjust your opening paragraph because Google uses it to generate search previews.
Long time on page with low conversion to your desired action suggests unclear guidance. Add explicit calls to action. Simplify your explanation. Break complex topics into smaller, more digestible sections.
Continue this cycle. Content is never "finished". Readers’ taste changes. Competitors publish better content. You get fresh ideas. Your job includes maintaining performance and pushing it up, not just achieving it once.
On-page metrics reveal reader behavior but not business impact. Strong engagement doesn't guarantee sales. A low bounce rate doesn't prove readers trust your content enough to act on it.
You need these metrics because they appear immediately and guide optimization. But you also need business metrics like trial signups, purchases, and customer retention. On-page metrics are your early warning system. Business metrics are your proof of success.
The relationship between on-page metrics and business outcomes is complex. A page with mediocre engagement metrics might drive significant revenue if it reaches the exact right audience. A page with excellent engagement might generate zero business value if it attracts the wrong readers.
Track both. Optimize for on-page metrics first because you can improve them quickly through content edits. Then track business metrics over longer timeframes to verify that improved engagement translates to business results.
Once you’ve defined the correct target values for on-page performance and you’ve optimized your content to achieve it, lagging indicators will follow. Your content will generate business and appear in high positions in search results (see the next section on SEO). Don’t wait for “no sales” or “poor SEO performance” to evaluate your content. Use the on-page performance to be the first to notice problems with your work, and correct it.
Great writers don’t need to seek praise for their content. Their content’s performance does it for them.
Search Engine Optimization (SEO) means making your content discoverable and appealing to people searching for what you're writing about. If you're writing content that no one can find, you're wasting your time.
SEO remains critical for two reasons:
1. Humans searching: People use search engines to find solutions. If your content doesn't appear when they search, they'll find your competitors instead.
2. LLMs researching: Large Language Models like ChatGPT and Claude search the web for current information. When someone asks an LLM about your topic, it will cite whoever ranks well in search results, not necessarily who has the best product.
Poor SEO means invisibility for both audiences.
Before optimizing anything, answer: Who is searching? What do they want to accomplish?
A developer searching "WordPress translation plugin" has different intent than someone searching "how to translate WordPress site." The first wants to evaluate options. The second needs information and guidance. Your content must match the specific intent behind the search.
Discover the exact words and phrases your target audience uses. Don't guess. Research.
Use Google's autocomplete: Type your topic and see what Google suggests. These suggestions come from real searches.
Check "People also ask" and "Related searches" at the bottom of search results. These reveal what else people search for on this topic.
Look through forums, like Reddit, to see the terminology that people use when searching for what you’re offering.
Look at what competitors rank for using tools like Ahrefs or Moz (covered below). If three top-ranking pages all target "translate WooCommerce products," that phrase matters to your audience.
Beware about using keyword-research tools without context. The fact that a keyword is popular doesn’t mean that this is your correct goal. It’s only your goal if this is the language that people use when they’re searching for what you’re offering.
Write for humans first, search engines second.
Your title and opening text must match search intent. People scan search results to identify what to click on. When searching, the only thing that they’ll see from your content from competitors’ content is the search results. These include the title and “search preview”. Help your audience by including the keyword they are searching for in the title. Your title needs to say “here is what you’re looking for”. Usually, Google will use your page’s title “as is” in search results. If Google’s algorithm considers your opening text as a great summary for your content, it will use it “as is”. You should aim for your opening text to be what Google shows in search results. For this, write an opening text that briefly summarizes your content and also set it as the page’s “SEO description” text. If Google needs to compose the search preview from the content of your page (disregarding your opening paragraph), you should rethink the opening paragraph.
Use keywords naturally throughout. Don't repeat the same phrase mechanically. Both search engines and LLM understand natural language. Stuffing your content with keywords will make your content unappealing for people and hurt your search performance.
Google still uses incoming links as a voting mechanism. The more incoming links from reputable and relevant content, the better. Votes (links) from sites that sell links never worked and still don’t work. Incoming links from irrelevant sites, using irrelevant keywords is a strong signal for Google that someone is trying to game its algorithm. It also doesn’t work.
We'll cover link building methodology in detail in a later chapter. For now, understand that links remain a critical ranking factor.
Google Search Console (GSC) provides complete, accurate data about how your site appears in Google search results.
GSC shows:
- Which queries trigger your content in search results
- How many people see your content (impressions)
- How many click through (clicks)
- Your average position for each query
- Which pages on your site get the most search traffic
Use GSC as your primary SEO measurement tool. It's free, accurate, and comes directly from Google.
Limitation: GSC only shows data for your own site. You cannot see your competitors' performance.
GA and Matomo track what happens after people click through from search results. They show on-page metrics covered in the previous chapter: bounce rate, time on page, conversion to your desired action.
Use these tools to understand whether search traffic engages with your content or leaves immediately. High search traffic with a high bounce rate means your content doesn't match what searchers expected.
Limitation: These tools show behavior but not search position or keyword data. Combine them with GSC for complete understanding.
Third-party SEO tools estimate your competitors' performance. They crawl the web to approximate:
- Which keywords competitors rank for
- How much search traffic competitors receive
- Which sites link to competitor content
- Historical ranking changes
These tools help identify content gaps. If a competitor ranks for "WordPress multilingual ecommerce" and you don't have content targeting that phrase, you've found an opportunity.
Critical limitation: Third-party tools provide estimates, not exact data. Their traffic estimates can be wrong by 50% or more. Use them to identify patterns and opportunities, not as precise measurements.
Run searches yourself. Type in the keywords you're targeting and see where your content appears. More importantly, see who appears above you.
Examine the top-ranking content:
- What intent does it satisfy?
- How is it structured?
- What makes it more relevant than your content?
- What does it include that you're missing?
Manual searching reveals information no tool provides. You see exactly what searchers see. You understand why competitors outrank you. It shows you what content people actually want to see (Google’s algorithm discovers this). See how your competitors appear in the search results and what they write about. Many people have voted by clicking on these competitors’ sites and staying on their content (not bouncing). This is your best intelligence tool for what your target audience is actually looking for.
Do this regularly. Search results change. Competitors publish new content. Your position fluctuates.
Remember that Google personalizes the search results. When doing SEO research, always use a browser in incognito mode (so that you don’t see results tailored for you) and use a VPN to check what people in your target audience see.
Diagnosis: Search engines haven't discovered your content, or they consider it irrelevant to any query.
Solution:
- Make sure that your content has some links, at least from other pages on site(s) that your company maintains. Search engines discover content by following links. Keep these links very relevant to your audience.
- Give it a bit of time. A few days is fine to start seeing organic traffic. A few weeks almost certainly means that you have a problem.
Diagnosis: Act quickly (within days). Google is giving your content a chance (usually happens with new content) and measuring how relevant it is for visitors. Right now, it’s not relevant and Google will soon stop showing it.
Solution:
- Repeat your research for how people search for what you’re offering. You might be targeting the wrong keywords.
- Rewrite your title and opening paragraph to more compellingly address what searchers want.
Diagnosis: Act quickly (within days). Visitors are voting with their feet (mouse) against your content. Google measures bounce rate and will soon understand that your content is not what people expected (click bait, intentional or not). Google’s algorithm will soon stop offering your content in search results.
Solution:
- Make sure that the content is really about what your opening promises.
- Make sure that your content gets to the point quickly.
Diagnosis: Search engines understand your content but don't consider it among the most relevant results.
Solution:
- Make sure that your “on page” performance is great.
- Strengthen your title and opening to more directly address search intent.
- Build more incoming links from relevant sites.
- Make sure that you’re aiming for a realistic goal. The first page of search results includes just 10 spots. Not all the sites on the Internet can appear in these 10 first spots.
Remember the previous chapter on on-page performance. SEO results lag by weeks or months. It takes time for search engines to discover updated content, re-evaluate its quality, and adjust rankings.
On-page metrics appear within days. High bounce rate, short time on page, low conversion—these signals warn you about problems immediately.
Fix on-page performance first. When readers engage with your content, search engines notice. They track how long people stay, whether they click through to other pages, whether they return to search results looking for something better.
Strong on-page performance leads to improved search rankings. Weak on-page performance guarantees poor SEO results, no matter how well you optimize keywords.
Monitor both. Act on on-page metrics quickly. Use SEO metrics to verify that your improvements translate to better search visibility over time.
In the previous chapter, on SEO, we explain how important links are for SEO. The “easy” links that you can generate are from other pages on the sites that you maintain. Good links help readers. To do this, they appear where clarification is needed and the target of the link is obvious to the reader.
Adding links often feels like the magic solution for every problem, so instead of talking about where to add links, we’ll talk about frequent mistakes that writers make when adding internal links. Try to avoid these problems 🙂
The link’s anchor text (what people see before clicking on the link) needs to exactly say what the page is about. If the anchor text is very different from the page’s title, it’s probably wrong.
| Bad link | Good link |
|---|---|
| Click here to learn more. | Need help replacing the filter? Read about vacuum maintenance. |
The objective of linking to other pages is to help readers understand the material. You can’t include everything on the one page that you’re writing right now. In our vacuum example, you’ll probably want to have a page that explains how to assemble it, how to use it and how to maintain it. When you talk about using the vacuum cleaner and you describe troubleshooting problems, you’ll probably want to help readers reach the page about maintenance (where you teach how to clean or replace the filter). A link to the vacuum maintenance does the trick.
But think about pages that include links every 3 works (like Wikipedia content). These links have just turned from being useful to being a nuisance. Having too many links on a page causes a few problems:
- The content is visually difficult to read, with so many words having “link styling”.
- The reader’s attention keeps wandering off. Every link, even if not clicked, causes our mind to consider - what’s there, should I click on it?
- It’s harder to notice and use the few links to actually useful resources.
As a rule of thumb, only link to what many readers will find useful often. If someone might, one day, maybe, need that information, don’t link.
Many pages have a goal - an action that you want readers to take. Think about the checkout page for an e-commerce website. You want visitors to enter their payment information and pay. Obviously, any link that will send visitors to any other page will reduce sales. The checkout page is an extreme example, but links easily distract from any other objective. A homepage with dozens of links will make it harder for a visitor to notice your “sign-up” button, even if you’ve styled it in bold red in the center of the page.
Your content should link to other content when these links truly help the reader understand the material. We’ll include a few examples:
| A good link | Example | Why it’s useful |
|---|---|---|
| A link to the “terms and conditions” page from the checkout page. | [ ] I’ve read and accepted the terms and conditions. | Legally, your checkout must disclose the conditions under which clients are buying from you. Many times, these conditions take up a lot of space. So, linking to the terms and conditions from the checkout page is the best solution, which almost every e-commerce site uses. |
| A link to the full list of languages your tool translates between, appearing in the “features” page of your translation tool. | We translate between 52 languages. | Not everyone, but enough people will want to see if their language is supported. |
| A major feature, from your homepage. | WPML’s Automatic Translation helps you translate at human quality and machine speed and cost. | The homepage mentions a number of major features. It can’t provide all the details about each of them, so it links to the “marketing” page for each major feature. |
Internal links accomplish two things for SEO:
- They allow search engines to find (and thus index and include) the pages you’re linking to.
- They signal the search engines which pages you consider to be the authoritative on your topics. If many of the pages on your site link to page X about a certain topic, it signals to Google that you consider page X to be the most relevant page for that topic.
If your site has multiple pages talking about the same topic, instead of excessive internal linking, consider unifying some of that content and dropping outdated content. You don’t need to work hard to signal to Google what’s the best page on your own site. Ideally, it will be just one page that addresses each topic.
The same rules that apply to internal links apply when building links from other sites. Links need to help readers and provide the right information, at the right time, in a clear way. There’s just one difference - now you need someone else (a human, not a website) to agree to link to your content. This one difference is very significant.
Pages link to content on other sites because someone, a human, decided to include that link. So, when you’re “building links” you’re not getting another site to link to you. You’re getting the person responsible for that other site to link from their content to your.
The reasons remain mostly the same, like when you’re deciding to link from your site (say, your homepage) to some other page that you wrote. You should add that link if it genuinely helps readers.
If you’ve just finished writing another page on a topic that’s been covered hundreds of times and you’re “building links” to it, there’s no reason for an authoritative site to link to you. But, if the content that you wrote is essential for the success of the readers of the “other site”, there’s a good chance that you’ll get that link even without asking for it.
For example, if you’re running a site for a service that helps with immigration to Denmark, nobody needs to pitch you the idea of linking to the right page, in the government website on immigration to Denmark, which lists the valid reason for getting a visa. You’ll link to it because your readers need this information and it helps your credibility, not because you’re doing the Danish authorities a favor.
So, when you’re “building links”, think about:
- What types of pages, on what types of websites will benefit from linking to what I just created?
- What are the actual sites and pages that fit this description?
- How do I reach the person who maintains the content on each of these sites?
- How do I explain it to this person, so that it’s about their readers and not about me promoting my content (for SEO, obviously)?
No.
Almost always “no”. It’s a bad sign for someone considering linking to you for the wrong reasons, which will usually result in exactly the opposite of what you’re aiming for. A small exception is when another person simply wants to get compensated for their time. But almost always “no”.
Yes, for the same reasons above. You link to content that will best help your readers. If that content is on your site, that’s great. If it’s on someone else’s website, go ahead and link to it. Different companies have different policies about outgoing links. These policies can be very different, often conflicting and still be correct, because circumstances are different.
Below are two policies about external links, which are almost opposite and both are valid. They’re opposite because the conditions are vastly different. Understanding that different circumstances will produce different linking policies will help you when you’re trying to get links from other sites to your content.
Example 1 - wpml.org (website translation, very established)
Since WPML’s website talks about a very well known brand and product (in WPML’s tiny ecosystem), it doesn’t mention competitors or link out to them. This policy comes from the idea of “there’s no bad publicity”. WPML’s website links freely to material written by partners, but only when this material is useful to readers often.
Example 2 - PTC.wpml.org (software translation, new arrival)
PTC is a relatively new name for software translation, so there’s no fear of it advertising competitors. PTC’s website can freely link to competitors when comparing what PTC offers versus what these well-established and known competitors have. Old and established competitors don’t need PTC to get people to know them. PTC can benefit from the anchoring effect of competitors to demonstrate its relative benefits.
What works best and requires the least effort (in link building) is to produce original and helpful content about topics that many people need and is not available elsewhere. By becoming the authority on popular topics, links “just happen”. It’s the same logic that dictates your entire content strategy, not just “incoming links”. Once you’ve posted that valuable, unique and original content, it’s usually a matter of communicating it to the right people.
Link building isn't a separate activity from content creation. If you're spending significant effort 'building links,' it usually means your content isn't valuable enough for people to naturally want to reference it. It’s OK. Not all content needs to be unique, have thousands of incoming links and appear on the top page of Google searches. Some content is a target for your newsletter and some is your legal stuff. But if you plan on getting search traffic for your content, you can’t write generic content and then go “link building”.
Eventually, great content gets links without any effort. But initially, you need to help others discover it. Once you've published something genuinely valuable, identify the people maintaining sites where your content would help their readers, and let them know it exists.
This is different from traditional "link building." You're not spamming or begging. You’re helping others see how your content will help their readers.
As a writer, it’s part of your responsibility to verify that the content you’re about to write has a good chance of succeeding. Don’t assume that any content idea that you raise or you receive is thoroughly evaluated and is ready for writing.
You need to check and see that:
- There’s unmet demand for this content
- There’s available for you a realistic funnel to drive traffic to this content
- You’re able to produce this content in high quality and on time
We always want to produce content that many people need and is not available. Here are some cases showing demand and supply.
| Demand | Supply | Vertict |
|---|---|---|
| People need to know about a new release of your software (certain demand). | There’s no announcement blog post for it (no supply). | Write the announcement blog post. |
| People need to know your terms and conditions (certain demand). | There’s already a page that describes the terms and conditions (available supply). | Don’t write additional documentation for the terms and conditions. |
| People are searching for how to do SEO (a lot of demand). | There are many blog posts and tutorials on how to do SEO (a lot of supply). | Don’t write another tutorial on SEO. |
| People need to know how to operate software X for purpose Y (solid demand). | There are only a few, outdated and incomplete tutorials on the topics (poor supply). | Write your own tutorial on operating software X for purpose Y. |
If you look at search results and see many high-quality guides from authoritative sites, you can't compete by 'trying harder.' Find a different angle or don't write it.
Bottom line, just “demand” (search traffic) is not enough. Don’t go into a content project if you can’t compete with the current “supply” (existing content).
Once you’ve established that there’s enough interest in your content and not too much competition, check that you can actually drive traffic to it. Sometimes, driving traffic is almost trivial. You’re writing to your clients and can email them to let them know about your new content.
A traffic funnel isn’t always as straight forward as sending a newsletter to your clients.
If you’re planning on SEO for your traffic, see that your domain has enough authority and is on a relevant topic. Your traffic funnel might be based on affiliates, partners and other arrangements. They’re all fine, as long as you analyze them realistically, not applying wishful thinking.
I wouldn’t be able to produce high quality content about life on planet Pluto. I’m not an expert on the topic and I can’t try it myself. Before you commit to producing content, make sure you have the resources to do so, on time. There’s no project with an infinite schedule. Someone is always waiting for your content and there’s always a list of other projects waiting for you.
If the schedule to complete the content isn’t clear to you, ask. If the amount of time you can spend writing this content isn’t clear to you, ask. When you know the parameters, check that the writing project is realistic.
You wrote great content, it gets traffic and converts into sales. That’s great. Now what?
Things change and you can’t assume that what works today will continue working, without any change forever.
Great writers maintain responsibility for the content that they produce. They make sure that supply/demand still works in their favor, that the content is up-to-date, that the on-page metrics still look great, that the traffic funnel still works and that the page converts to its objectives.
There are many ways you can do this. Different companies have different systems in place to automate as much as possible and to alert you when action is needed. What’s important is that you remember that maintaining the content that you wrote is your responsibility. You need to account for this ongoing maintenance work when you plan your capacity for writing new content. Of course, you should aim to automate as much as possible, so that monitoring the performance of existing content takes a tiny amount of your time.
After a while, many content items need changes. One of these changes can be “it’s time to remove this content from our site”. When this happens, don’t try to salvage the content, repurpose it or bend reality. Some content stops being relevant and needed and it’s time to remove it from your site.
For example, there’s not much demand today for content that explains how to prepare your software for the Y2K bug. It was a really hot topic in 1999, but nobody needs it today. If you wrote a brilliant piece on preparing for the Y2K bug in 1999 and in 2001 you noticed that it gets no traffic, the best is to remove it. Accumulating old content that gets no traffic hurts your site in many ways. Prune it.