Lists of marketing mistakes tend to be lists of things that are easy to see. Inconsistent branding. No clear call to action. Not posting enough. Real problems, all of them, and none of them the reason a business spends a year on marketing and cannot point to what it bought.
The expensive mistakes share a property: they produce no symptom. Nothing breaks. The dashboard still loads. Money leaves and the loss shows up as an absence, which is the hardest thing for anyone to notice.
section 01publishing into a site that cannot be crawled
This is the most expensive mistake we see, and almost nobody checks for it, because there is no error message.
Google has to discover a page, crawl it, and decide it is worth indexing before it can rank. A sitemap does not guarantee any of that. A sitemap is a suggestion. Internal links are the actual argument for why a page deserves crawl budget, and a page nothing links to is an orphan regardless of how good it is.
In Search Console the symptom has a name: "Discovered, currently not indexed." Google read the URL out of your sitemap and declined to spend the budget on it. Not penalized. Not rejected. Deprioritized, because nothing on the site argued for it. The usual cause is structural, a navigation or footer that never linked to most of the pages, leaving them orphaned from the day they shipped.
The pages still get written, edited, and paid for. None of them can rank. The cost is not the pages, it is the year of publishing on top of a foundation that cannot carry it.
Open Search Console, go to Pages, and look at the split between indexed and not indexed. It takes ten minutes and it is the highest-value ten minutes available to most businesses.
section 02measuring the wrong end of the funnel
The second mistake is trusting numbers that are structurally incapable of answering the question being asked of them.
Sessions went up. Did revenue? Rankings improved. For queries anyone searches? Cost per click fell. Did cost per customer? Most marketing reporting measures the cheapest thing to measure rather than the thing that matters, and the gap between those two is where budgets die.
The specific version that costs the most is attribution that cannot see the actual path. Someone finds you on search, does not convert, sees you again a week later, then calls the phone number on your site. Nearly every reporting setup calls that a direct visit with no source, which means the channel that produced the customer gets credited to nothing. Do that for a year and you will systematically defund the thing that works. We wrote the full version at the lead you can't measure.
section 03treating a conversion problem as a traffic problem
When results are flat, the reflex is to buy more traffic. It is the easiest thing to buy and the easiest thing to justify.
But if the site converts at half of what it should, doubling traffic doubles the cost of the same outcome. The conversion fix is usually cheaper, faster, and permanent, and it makes every future dollar of traffic worth more. The traffic fix makes the leak bigger.
Telling them apart is not hard. If people arrive and leave without engaging, the offer or the page is the problem. If almost nobody arrives at all, it is traffic. Most businesses have the first problem and buy the solution to the second. The longer diagnostic is at traffic problem or conversion problem.
section 04answering every question with more content
The fourth is the one the industry is most invested in not saying, because content is the easiest thing to sell by the unit.
More content helps when the gap is coverage: real questions your buyers ask that you have not answered. It does nothing when the gap is authority, indexing, conversion, or offer. Publishing twenty posts onto a domain Google is already crawling reluctantly spends budget you do not have on pages that join the same queue as everything else.
The tell is whether anyone can name the question each piece answers and the buyer who asks it. If the plan is a number of posts per month, the plan is a production schedule, not a strategy. The argument in full is at why more content is the wrong answer.
Every one of these four is invisible on a dashboard and cheap to check. That combination is exactly why they persist: nobody is alerted, and nobody looks.
Check the boring things first. The expensive mistakes are never the ones anyone is arguing about.