Harsha Kotthapalli
← All 34 outputs

Research · output 5 of 34

Assumption log: what your plan treats as true.

Every claim your plan already rests on, marked by how much evidence sits behind it. Harvested from your deck, your homepage, your roadmap and your notes; the fatal untested ones get the cheapest possible test.

See the real example ↓Take the free diagnostic →
Assumption log11 claims
Proven
Partly known
Nobody checked
Disproved
Expats will pay monthly
Employers will hand it to staff
They don't know there's no pension
Everything the plan has checked is a fact about the market. Nothing it has checked is about whether anyone pays.

The full output carries all eleven claims with their source, and a test with its pass number for the four that are fatal.

See the full output →

The output at a glance · built on WealthKarma

How it gets made

What goes in

Deck and homepage
The roadmap
Call notes
Internal docs

Every line traces to a sentence somebody wrote

What comes out

The living log

The report

Updated as tests run

How long

1day

to build

then minutes as evidence arrives

The output, in full

Built on WealthKarma, a retirement app for Gulf expats. Every count is real and every gap is marked as a gap.

Assumption log11 claims · 5 untested · Aug 2026
01 · What the plan takes as true

Proven

Partly known

Nobody has checked

Tested, and it was false

The claimIf wrongFrom
Expats will pay monthly for a retirement planFatalPricing page
People will move the end-of-service money into a planFatalDeck, slide 6
Employers will hand the app to their staffFatalDeck, slide 9
Someone who signs up will still be paying in year twoFatalDeck, slide 12
Saudi and Qatar will follow BahrainCostlyDeck, slide 7
Enough of them earn enough to saveCostlyDeck, slide 4
Trust is the thing standing in the wayCostlyHomepage
No Gulf state gives expats a pensionCostlyDeck, slide 3
They have been putting it off for yearsCostlyHomepage
They don't know there's no pensionFatalDeck, slide 2
We lose to the other appsCostlyDeck, slide 10
02 · What we already disproved

They don't know there's no pension

Four of five knew, and had done nothing anyway. The plan was built to inform people who are already informed.

We lose to the other apps

Not one of twelve losses went to a competitor. Five chose nothing at all.

03 · What nobody has checked

The four that are fatal, in the order they'd kill it. Each number was written before the test.

Expats will pay monthly for a retirement plan

Cheapest test
A price on the page and a card field. No product behind it.
Passes at
5 of 20 reach the card field
If it fails
The revenue model is the employer, not the saver.

People will move the end-of-service money into a plan

Cheapest test
Ask ten people who just changed jobs what they did with it.
Passes at
3 of 10 still hold it uninvested
If it fails
The gratuity is a talking point, not a source of funds.

Employers will hand the app to their staff

Cheapest test
Take it to eight HR heads and ask for a pilot, not an opinion.
Passes at
2 of 8 agree a dated pilot
If it fails
There is no distribution and every user costs money to find.

Someone who signs up will still be paying in year two

Cheapest test
None yet. Nobody has been a customer for a year.
Nearest read
Whether month-three contributions are still going in.
04 · The finding
Everything the plan has checked is a fact about the market. Nothing it has checked is about whether anyone pays.
05 · What can't be known
Year-two behaviour, until there is a year two.
What was assumed and never written down anywhere.
Whether a test that passes at these numbers holds at scale.

Where it sits

Research

Compete

Position

Message

Launch

Measure

you are here · 5 of 34

Want this for your company

This is one of thirty-four. The diagnostic tells you which ones you're missing.

Take the free diagnostic →Book a call →
← Why now

Next in the journey

Pre-mortem →