The spreadsheet is the competitor: spotting manual workarounds in creator content
The most valuable thing in a practitioner video is usually the step the creator apologises for, speeds through, or cuts away from.
The short answer: hunt for the steps practitioners perform without explaining — the copy-paste, the export-then-reimport, the side spreadsheet, the naming convention nobody wrote down. Then keep only the ones that are frequent, error-prone, time-pressured or expensive to get wrong. That filter is the difference between a product and a plugin nobody needed.
Most idea research looks for stated problems: the complaint, the wish, the “I really wish something would just…”. Those are real but they are also rehearsed, and rehearsed opinions drift toward whatever the person read last week. The unrehearsed evidence is behavioural — what people actually do when the camera is running and they have stopped performing.
Your real incumbent is usually not a company
Founders benchmark against funded competitors because those are visible. The customer benchmarks against what is on their screen right now, which in most categories is a spreadsheet, a shared document, a recurring calendar reminder, or a colleague who knows how to do the thing.
That incumbent is formidable in ways a software competitor is not. It is free, it never has an outage that you get blamed for, it can be reshaped in thirty seconds, and the person using it already knows how. Any product entering the category is asking someone to give up all four advantages, and the pitch has to be worth that trade.
Not “is my product better than the spreadsheet” but “is it better by enough to justify learning something new and giving up total flexibility”. Products that lose to a spreadsheet usually cleared the first bar and never considered the second.
Four markers that identify a workaround on camera
| Marker | What it looks like | Why it is reliable |
|---|---|---|
| The apology | The step is excused rather than explained | People only apologise for steps they know are silly |
| The jump cut | Time disappears mid-workflow | Editing removes the boring part, which is the manual part |
| The second window | A sheet or doc open beside the main tool | Mirrored data is a missing feature made visible |
| The verbal convention | Naming or filing rules recited from memory | Convention held in a head is structure with no system |
The second window is the densest of the four. When someone keeps a spreadsheet next to a tool that supposedly does the job, the sheet contains exactly the fields the tool refuses to model, and those fields are a product specification handed to you for free. Reading them alongside the vocabulary people use is how the schema falls out, which is the method in designing a data model from domain vocabulary.
Which workarounds are worth building for
- ✗Rare steps treated the same as daily ones
- ✗Flexibility replaced with a rigid flow
- ✗Workarounds that are someone's expertise productised over their objection
- ✗Scope set by the breadth of the corpus
- ✓Frequent: performed at least weekly
- ✓Error-prone: people double-check the output
- ✓Time-pressured: done while someone waits
- ✓Expensive when wrong: rework, refunds, trust
The double-check is the most portable of the four, because it is visible without any domain knowledge. When a practitioner verifies their own output before acting on it, the step is both risky and untrusted, and both properties raise the value of automating it — and the cost of supporting it, which is the other side of the same reading in estimating support burden from practitioner complaints.
Observed beats reported, consistently
Ask people what they need and you get a feature list shaped by what they have recently seen advertised. Watch them work and you get the actual cost structure of their day. The two diverge often enough that treating them as interchangeable is one of the more expensive research mistakes available.
This is the same argument for sourcing over-the-shoulder footage rather than talking-head commentary, set out in workflow mapping from day-in-the-life videos, and it is why validation that never watches anybody work tends to confirm whatever the founder already believed — the failure mode examined in validating a SaaS idea without surveys.
When the workaround is the person
Sometimes the manual step is where an individual’s value lives: the judgement call, the client-specific tweak, the thing they are known for. Productising it directly will be resisted, and the resistance is not irrational.
The better move is usually to automate everything around that step and leave it as a deliberate human decision point, which converts a threat into leverage. The same instinct explains why some categories hold customers tightly for reasons that have nothing to do with features — the territory covered in switching costs and why people stay.
Workarounds are a scoping tool, not just an idea source
Once the list is ranked by the four costs, version one is easy to draw: take the two or three highest-cost workarounds and remove the need for them entirely. Everything else in the workflow stays manual, on purpose, and the product is judged on whether those two or three steps actually disappeared.
That is a narrower first version than most research suggests, and it is the right one, for the reasons in scoping a first version from build-along videos. It also gives you a demo with a spine, because the before-and-after is a specific step the viewer recognises.
Count the workaround before you build for it
A workaround becomes a business case when it has a number attached, and practitioner content usually supplies enough to estimate one. Three quantities are normally recoverable: how often the step happens, roughly how long it takes, and how often it goes wrong badly enough to need redoing.
None of those will be stated precisely, and precision is not the point. A step performed twice a day at two minutes each is an hour a week; the same step performed twice a month is twenty minutes a year and should not shape a product. The order of magnitude decides, and the order of magnitude is legible even from vague narration.
| Recoverable quantity | How it surfaces | What it decides |
|---|---|---|
| Frequency | Mentioned cadence, or implied by the workflow | Whether it is worth any engineering at all |
| Duration | Elapsed time on camera, before the cut | The size of the promise you can make |
| Failure rate | Redo stories, verification habits | Whether accuracy or speed is the pitch |
The failure rate is the one that changes positioning. A slow but reliable workaround wants a speed pitch; a fast but error-prone one wants an accuracy pitch, and the two produce completely different products from the same observed behaviour. Getting that backwards is how a genuinely useful tool gets marketed at the wrong problem.
What this costs to run
No separate corpus is needed — the workaround read is a second pass over the videos you already gathered. As of September 2026, Hobby is $19 a month with 25 videos and 2 projects, Pro is $59 with 80 videos and 8 projects, and Studio is $199 with 250 videos, 20 projects and 3 seats, each with a 7-day free trial; see the pricing page.
Run a synthesis that surfaces manual bridges, mirrored spreadsheets and repeated double-checks across your whole corpus. 7-day free trial.
Closing thought
The tutorial tells you how the tool works. The ten seconds the creator edited out tells you what to build.
Frequently asked
Why is a spreadsheet a competitor?
Because it is what the customer uses today, and any switch you ask for is measured against it rather than against another vendor. A spreadsheet is free, already understood, infinitely flexible and owned by the person using it, which is a stronger position than most funded startups hold.
What exactly counts as a manual workaround?
Any step a practitioner performs by hand to bridge two systems or to compensate for something a tool does not do: a copy-paste between tabs, an export followed by a re-import, a naming convention held in someone's head, a recurring reminder, or a shared sheet that mirrors data living somewhere else.
How do I spot one in a video when nobody calls it out?
Watch for the apology. Practitioners narrate workarounds defensively, with phrases that excuse the step rather than explain it, and they tend to speed up or cut away during it. A jump cut in the middle of a workflow is one of the most reliable markers in the entire medium.
Are all workarounds worth automating?
No. A workaround that takes twenty seconds once a month is a fact of life. The ones worth building for are frequent, error-prone, performed under time pressure, or carry a real cost when done wrong, and those four properties are usually audible in how the person talks about the step.
What if the workaround is the practitioner's competitive advantage?
Then it is a template rather than a feature, and productising it will meet resistance. People defend the parts of a manual process that make them individually valuable, which is a useful signal that you have found a real job rather than busywork.
How is this different from reading feature requests?
Feature requests describe what people imagine would help. Workarounds are what they already do, at cost, without being asked. The second is evidence and the first is speculation, and they frequently point in different directions.
What does this analysis cost?
As of September 2026, Hobby is $19 a month for 25 videos and 2 projects, Pro is $59 for 80 videos and 8 projects, and Studio is $199 for 250 videos, 20 projects and 3 seats, with a 7-day free trial on every plan. Workaround-spotting runs over the same corpus as validation, so it costs a synthesis rather than new sourcing.