Comparing seven health plans without a spreadsheet

What has to line up before a comparison means anything?

Four things, none of which any two insurers write the same way.

Age bands. No standard grid. Across the charts Rabt holds there are 22 distinct bands, so the same 34-year-old is priced off a 26–35 row on one chart and a 31–35 row on another.

Member categories. Some charts carry one FEMALE column; some split single and married female; Abu Dhabi charts price by investor, employee and dependent instead. A comparison has to decide which cell each member reads on each chart.

Network tiers. NextCare's RN3, NAS's Super Restricted Network and Dubai Care's N4 do not map onto each other. There is no ladder they share.

Benefit rows. Every table of benefits orders its rows differently, names them differently and states limits in different units — a figure, a percentage of the annual limit, a session count, or "as per policy".

Until all four are normalised, putting two premiums side by side is putting two unlike things side by side and letting the smaller number win.

Why does the spreadsheet version go wrong?

Not because it cannot be done. Because it has to be redone.

Building the grid once for seven plans and a family of four is 28 rate lookups plus four tables of benefits transcribed by hand. That is an afternoon, and a careful broker gets it right.

Then the client adds a member. Then one plan's renewal terms change. Then the client asks what the 20% co-pay option looks like instead of nil. Each of those invalidates part of the grid, and the part it invalidates is not visibly different from the part it does not. The second version of a spreadsheet is less accurate than the first, and the fifth is a document nobody quite trusts but everybody quotes from.

What does an honest comparison have to show?

Price is the easy row. The rows that change decisions, and that a rate-chart-only comparison cannot produce:

• Network, named exactly as the TPA writes it, including the parenthetical about out-patient restrictions.
• Co-pay structure with its caps — 20% with a AED 50 per-visit cap is a different product from uncapped 20%.
• Annual aggregate limit, and the sub-limits underneath it that clients actually claim against.
• Maternity, with its waiting period. Nil to 40 days across the plans in Rabt's demonstration inventory.
• Pre-existing conditions — covered from inception, covered after 180 days, or subject to declaration. All three appear at similar premiums.
• Whether a medical application form is required, and above what age. A plan needing one cannot be quoted as a firm price.
• Eligibility filters — maximum entry age, whether members below the AED 4,000 salary threshold are accepted, whether a standalone child is allowed.

That last group is the one spreadsheets omit entirely, because they are not numbers. They are the reason a quote cannot be issued, discovered at submission instead of at comparison.

What is the common objection — "the client just wants the cheapest"?

The client wants the cheapest plan that works. The two diverge at exactly the points a price-only comparison cannot see: the network their existing doctor is on, the waiting period against the baby they are planning, the declared condition on the medical form.

A comparison that shows those alongside the price does not slow the sale down. It removes the second conversation, three months later, about why the card did not work.

How many plans is the right number to show?

Enough to demonstrate the range, few enough to decide from. In practice a client cannot hold more than about five columns in their head, and a broker who shows seven is usually showing two they have already discounted.

The useful discipline is to compare many and present few — which requires the comparison to be cheap to build, because the ones that do not make the cut still had to be priced.

What Rabt does instead

Rabt holds each plan's own bands, categories and networks as they are printed, and resolves every member against every plan at quotation. Adding a member re-reads 22 age bands rather than requiring anyone to find the right rows again.

Eligibility rules are held as data on the plan — maximum age, salary threshold, whether a medical application form is needed and above what age, whether standalone children are allowed — so a plan that cannot be issued for this family is excluded from the comparison with the reason stated, rather than quoted and withdrawn later.