The audience qualifier tax.
Every vertical software company writes the same kind of page: booking software for coaches, a CRM for therapists, a website builder for photographers. We measured 30 software categories with and without the audience they are sold to, 150 pairs in all, and the qualified phrases keep 0.45% of the demand.
- 30 categories, unqualified
- 834,060
- US searches a month, combined
- The same 30, across 150 qualified forms
- 3,760
- 0.45% of the above
- Qualified phrases with no demand
- 121 of 150
- 80.7% report zero
What we measured
30 software categories that exist as searches in their own right, from website builder down to membership site software. Each was then measured again followed by 5 audiences: for coaches, for creators, for therapists, for photographers, for consultants. That is 150 qualified phrases plus 30 bare ones, 180 in total, all US monthly search volume from the Google Ads Keyword Planner via DataForSEO on 2026-09-12.
The one thing that could have faked this result
Google Ads returns a combinedvolume for groups of similar keywords, and DataForSEO's documentation says to submit such keywords in separate requests. A category and its own qualified form are precisely that kind of pair: batched together, one shared figure could be stamped across both, manufacturing either a null result or a spurious one depending on which way it collapsed.
So no category ever shared a request with its own qualified form. One request carried the 30 bare categories, then one request per audience. Within any single request the phrases are different categories, which is not the close-variant case the warning describes. As an audit rather than an assurance, we then re-measured seven phrases one per request, four of them the largest non-zero results and three of them zeros. All seven returned the batched figure exactly.
It is not uniform, and that is the useful part
A study that only reported the result it hoped for would be an advert. Some audiences cost far more than others, and a handful of category-and-audience pairs have become market terms in their own right and keep real demand.
| Audience | Combined volume | Categories at zero |
|---|---|---|
| for coaches | 50 | 26 of 30 |
| for creators | 1,510 | 27 of 30 |
| for therapists | 280 | 23 of 30 |
| for photographers | 830 | 23 of 30 |
| for consultants | 1,090 | 22 of 30 |
The phrases that survived, and why the big ones are not the interesting ones
The obvious way to read this is by volume, and it is the wrong axis. 9 qualified phrases keep 100 searches a month or more, but a large number can simply mean a large category: website builder for photographers keeps 720 searches a month, which sounds respectable until you see it is 0.358% of a category worth 201,000. It is not an exception to the rule. It is the rule, wearing a big number.
What separates a real counter-example is the SHARE of the category it kept, measured against what an average qualified form keeps. That is not the headline figure and the distinction matters: the headline's 0.45% puts all 150 measurements over 30 categories, so it is not an average of anything a single phrase can be compared to. The mean share across the 150 qualified forms, zeros included, is 0.94%. By that measure 6 of the 150 pairs beat the average, and the top of the list is the thing worth knowing: a qualified phrase that outsells the bare category it qualifies has stopped being a category plus an audience and become a category of its own.
- community platform for creators keeps 122.92% of its category, 590 a month against 480 for community platform alone.
- time tracking software for consultants keeps 4.88% of its category, 590 a month against 12,100 for time tracking software alone.
- scheduling software for therapists keeps 3.06% of its category, 110 a month against 3,600 for scheduling software alone.
- proposal software for consultants keeps 2.78% of its category, 20 a month against 720 for proposal software alone.
- email marketing platform for creators keeps 2.66% of its category, 880 a month against 33,100 for email marketing platform alone.
- booking software for photographers keeps 1.03% of its category, 30 a month against 2,900 for booking software alone.
What to do about it
The wrong conclusion is to stop writing for a niche. The audience does not belong in the search phrase; it belongs in everything else. Aim the title and the URL at the category term people type, then write the page entirely for the audience you sell to, with their examples, their objections and their vocabulary.
We know because we did it wrong. Eleven of our own nineteen marketing pages led with a phrase of zero measured demand, and the tell was the same suffix every time. Fixing that is what prompted this study, and the numbers on this page are the ones we should have measured first.
The full table
All 30 categories and 150 qualified phrases, so you can check any figure quoted above. US monthly searches, 2026-09-12.
| Category | Alone | coaches | creators | therapists | photographers | consultants |
|---|---|---|---|---|---|---|
| website builder | 201,000 | 10 | 40 | 110 | 720 | 10 |
| crm software | 165,000 | 10 | 0 | 0 | 20 | 110 |
| project management software | 165,000 | 0 | 0 | 0 | 10 | 210 |
| accounting software | 90,500 | 0 | 0 | 20 | 30 | 110 |
| social media scheduler | 90,500 | 0 | 0 | 0 | 0 | 0 |
| email marketing platform | 33,100 | 0 | 880 | 0 | 0 | 0 |
| live streaming platform | 22,200 | 0 | 0 | 0 | 0 | 0 |
| time tracking software | 12,100 | 0 | 0 | 0 | 0 | 590 |
| online course platform | 9,900 | 0 | 0 | 0 | 0 | 0 |
| invoicing software | 5,400 | 10 | 0 | 10 | 10 | 30 |
| podcast hosting | 4,400 | 0 | 0 | 0 | 0 | 0 |
| payment processing | 4,400 | 0 | 0 | 10 | 0 | 0 |
| scheduling software | 3,600 | 20 | 0 | 110 | 10 | 10 |
| sales funnel builder | 3,600 | 0 | 0 | 0 | 0 | 0 |
| booking software | 2,900 | 0 | 0 | 10 | 30 | 0 |
| landing page builder | 2,900 | 0 | 0 | 0 | 0 | 0 |
| form builder | 2,900 | 0 | 0 | 0 | 0 | 0 |
| appointment scheduling software | 2,400 | 0 | 0 | 10 | 0 | 0 |
| marketing automation software | 1,900 | 0 | 0 | 0 | 0 | 0 |
| customer support software | 1,900 | 0 | 0 | 0 | 0 | 0 |
| affiliate marketing software | 1,600 | 0 | 0 | 0 | 0 | 0 |
| online store builder | 1,600 | 0 | 0 | 0 | 0 | 0 |
| video hosting platform | 1,300 | 0 | 0 | 0 | 0 | 0 |
| survey software | 1,000 | 0 | 0 | 0 | 0 | 0 |
| email newsletter software | 880 | 0 | 0 | 0 | 0 | 0 |
| client portal software | 720 | 0 | 0 | 0 | 0 | 0 |
| proposal software | 720 | 0 | 0 | 0 | 0 | 20 |
| community platform | 480 | 0 | 590 | 0 | 0 | 0 |
| digital product platform | 90 | 0 | 0 | 0 | 0 | 0 |
| membership site software | 70 | 0 | 0 | 0 | 0 | 0 |
Questions
Does this mean you should never write for a niche audience?
No, and that is the most common misreading. It means the audience does not belong in the search phrase. Write the page for coaches, speak to coaches throughout, use their examples and their objections, and aim the title at the category term they actually type. The audience belongs in the copy, not in the keyword.
Could the zero figures just be missing data rather than no demand?
Google Ads reports nothing below a threshold, so a zero means demand too small to report rather than literally nobody. That distinction does not change the decision: a phrase Keyword Planner will not report is a phrase you cannot build an acquisition channel on. We publish the whole table so you can see which are zero and which are merely small.
How do you know the batching did not create this result?
That was the one thing that could have manufactured it, so the design rules it out. A category never shared a request with its own qualified form: one request held the thirty bare categories, then one request per audience. We then re-measured seven phrases one per request, four of the largest non-zero results and three zeros, and all seven returned the batched figure exactly.
Why these five audiences?
They are the ones vertical software is most often sold to, and they vary enough to show the effect is not uniform. Creators and consultants retain meaningfully more demand than coaches and therapists, which is the more useful half of the finding.
Can I use this data?
Yes. Quote it, chart it, or check it. The full table is on this page and every number is derived from the same dataset rather than typed into the prose, so what you see is what was measured. A link back is appreciated.
This is one of our original studies. We publish the method and the whole dataset with every one, so the number can be argued with rather than taken on trust.
We build the thing this research is about
Drry is a platform for running a paid community: courses, memberships, digital products and email in one place, 0% platform fee on every plan, paid into your own Stripe.