Hard to Market Is a Diagnosis, Not a Condition

"Our product is just hard to market."

I hear that sentence early in most conversations with a founder whose growth has flattened, and it is usually delivered with a shrug, as though difficulty were a property of the product like weight or color. What follows is almost always a messaging sprint. New homepage, sharper headlines, an agency with a deck about narrative. Sometimes that works. More often the company spends a quarter rewriting sentences and the pipeline in March looks like the pipeline in December.

The reason is that hard to market is not one condition. It is four, they have different repairs, and the expensive mistake is running the repair for a problem you do not have.

The four causes I look for:

• The product sits in the wrong competitive set.
• You are selling to the wrong buyer.
• The problem is real and nobody has funded it.
• The category is too young for anyone to be shopping in it.

I published a version of this argument in June that ran those as sequential wedges, one leading into the next. Reading it back, the ordering is the part I would take out. These are independent. A company can have exactly the right buyer sitting inside the wrong competitive set. Another can have a sharp frame wrapped around a complaint nobody has budgeted to fix. Treating them as steps means you start at step one no matter what is actually wrong, which is how positioning work becomes the default answer to every stall.

So the sequence is not four fixes. It is four tests, and they run in parallel in about an afternoon. The diagnosis is the cheap part. What comes after is a real build, and each cause calls for a different one.

Wrong competitive set: the explain-back test

The test I trust here is whether a buyer can explain your product accurately to a colleague after one conversation. Not repeat your tagline. Explain it, in their own words, to somebody with budget authority who was not in the room. If they cannot, no amount of copy is going to fix it, because the problem is that they have nowhere to file you.

The second signal is what your lost deals compare you to. If prospects keep naming competitors you consider irrelevant, you have not been misunderstood. You have been correctly filed into a set you do not want to be in, and that set is deciding your price.

Positioning is not what you say about the product. It is which competitive set you place yourself in, and the set determines the price you can charge, the buyer you sell to, and the urgency they feel.

At GlacierGrid, back when it was called Therma, the product monitored refrigeration temperature. As a refrigeration monitor it competed in a commoditized niche against other monitors, which capped the price and put the decision with a facilities manager. We moved the frame to a grid-responsive energy platform, cooling and refrigeration treated as a battery. The technology did not change much. The competitive set changed completely, and with it the buyer, the budget line and the urgency. Qualified pipeline grew 600%, CAC improved 60%, and the reframe supported the company's move into a broader energy platform and the story behind its $19M Series A.

The build here is not a new headline. It is a new comparison set carried through pricing, the ICP, the sales talk track and the qualification criteria at the same time. Change the words alone and the old set snaps back within a month.

Wrong buyer: thirty interviews and a hundred names

The interview rule I run is ten current customers, ten churned, ten prospects who evaluated you and bought something else. The churned and the lost are where the answer lives, and they are the two groups every company skips.

The threshold that matters: seven out of ten saying the same thing is a pattern. Fewer than that is an anecdote, and I have watched entire repositionings get built on three enthusiastic calls.

Then the arithmetic check. The Named Account Test asks you to write down 100 specific companies that fit your stated ICP. Real names, not a segment description. If you cannot get to 100, you do not have an ICP, you have a hypothesis, and any CAC number you produce against it is noise.

The ICP format I use has five slots: [Role] at [Company Type] who is [Situation] and needs [Outcome] because [Stakes]. If any slot is empty you are targeting a demographic rather than a buyer. The stakes slot is the one that goes missing most often, and it is the one doing the work.

The build is a rewritten ICP with spend rerouted behind it, not a new persona slide.

Unfunded pain: budget, timeline, owner

This is the cause most often misread as a messaging problem, and it is the one where a messaging sprint does the most damage. Better copy on an unfunded pain raises reply rates and meeting counts while nothing closes, which reads like progress for about two months.

The test has three markers. Does a budget exist. Is there a timeline. Is there a named owner. A prospect with none of the three is describing a complaint. A complaint is talked about. A pain is paid to make go away. If they have not tried to solve it already and have not put money against it, you are not selling into demand, you are trying to create a budget line, which is a much longer motion than most plans assume. I laid out the full ladder in an earlier piece on Level 3 and Level 4 pain.

You have two honest moves. Elevate the consequence until it reaches something already funded, or change who you sell to, because the same problem is often Level 2 for one role and Level 4 for another.

The GlacierGrid homepage was converting at 2.1% behind an abstract hero about monitoring. We replaced it with the consequence the buyer already had a number for, preventing a $50K freezer failure. Demo requests went up 47%. The copy got more specific, but what actually changed was that the page started naming a loss somebody was already accountable for.

Young category: check whether anyone is searching

Before you blame the funnel, check search volume for the category term. If nobody is typing it, capture channels have nothing to capture, and a search campaign will faithfully report that a market you have not created yet is not converting.

The decision rule is a budget shape, not a channel ban. When the category term has no volume, most of the money belongs in discovery, where people can encounter the idea without knowing to look for it. Once the term exists, paid becomes the fastest way to own it, and I would rather be the one already holding that inventory when the searches start.

At Cubii the category did not exist. Nobody was searching for an under-desk elliptical because most people did not know that was a thing you could buy. We cut AdWords from roughly $20K a month to $2K and moved into discovery, and the useful move alongside it was borrowing a mental model people already trusted, so it sat next to physical therapy and active sitting rather than next to gym equipment. That spend shift was part of a turnaround that took the business from losing $40K a month to profitable in about 60 days.

What to do first

Block one afternoon. Run the four tests on your own business and write a one-line verdict for each: can a buyer explain us back, can we name 100 real accounts, does the pain have a budget and a timeline and an owner, is anyone searching for the category.

Two of them will usually come back yes. Fix the cheaper one first, then re-run the other test rather than assuming it moved.

What I would resist is the reflex to start rewriting the homepage before those four are answered. Hard to market is rarely a story problem first. It is a story problem last, after you know which of the four you are actually in. If you want a second read on which one your company has, that is a good use of a 20-minute conversation.