How to Price Software in East Africa Without Racing to the Bottom

Every developer in Nairobi has had this conversation. A client asks what it will cost. You say a number. There is a pause on the other end of the call. Then: "That is a bit high. My cousin's guy said he can do it for eighty thousand."

What happens next decides what kind of business you are running for the next five years. Most of us discount. Then we discount again on the next deal, because we have quietly reset our own idea of what the work is worth. Two years later we are working sixty hour weeks, chasing invoices, and wondering why the company never has money in it.

Here is my thesis, learned the expensive way: underpricing does not kill you loudly. It kills you slowly, and by the time you notice, you cannot tell it was the cause of death. A bad project you priced properly is survivable. A good project you underpriced is what forces you to take the next bad project.

First, decide which market you are actually in

"The East African market" is not one market. There are at least three, they pay differently, and they buy on completely different signals. Pricing badly usually means pricing for one market while selling into another.

Market	Who they are	What they buy on	Budget shape
Local SME	Distributors, schools, clinics, retail chains, family businesses	Trust, referral, "will you still answer my calls in month six"	Small, cash sensitive, wants one number and monthly instalments
Local corporate, NGO, donor, county	Banks, SACCOs, insurers, INGOs, public sector	Procurement compliance, references, paperwork, perceived risk	Larger, slower, budget cycles, 30 to 90 day payment, withholding tax deducted
Offshore or diaspora	European, US, Gulf clients buying remote engineering	Communication, timezone overlap, portfolio, English fluency	Hourly or monthly retainer, benchmarked against global rate tables

The local SME market will punish you for hourly rates because the client has no way to verify hours and every reason to fear them. The offshore market will punish you for fixed prices because they are used to buying capacity. The corporate and donor market will punish you for anything that is not a formal, itemised, VAT-inclusive proposal on letterhead.

Work out your floor before you look at anyone else's price

Most pricing advice starts with the market. That is backwards. Start with the number below which the business does not survive, because that number is not negotiable and the market's opinion of it is irrelevant.

Worked example for a solo consultant in Nairobi who wants to take home KES 300,000 a month.

• Target take-home: KES 3,600,000 per year.
• Gross-up for PAYE and the cost of paying yourself properly: the exact figure depends on your structure, so ask your accountant, but budget roughly 1.35x. That is about KES 4,860,000.
• Overheads: internet and power 15,000; hardware amortised 8,000; hosting, licences and tools 10,000; accountant and compliance 10,000; marketing and proposals 10,000; medical cover and insurance 8,000; contingency 10,000. That is KES 71,000 a month, roughly KES 850,000 a year.
• Revenue required: about KES 5,700,000 a year.

Now the part everyone gets wrong. You do not have 2,080 billable hours. Take 52 weeks, remove 4 weeks of leave and 2 weeks of sickness and public holidays, and you have 46 working weeks, or 1,840 hours. A solo consultant who also sells, writes proposals, invoices, chases payment, does support and keeps learning bills about 50 to 60 percent of that. Call it 1,000 billable hours.

KES 5,700,000 divided by 1,000 hours is KES 5,700 per hour, or roughly USD 44 at about 130 shillings to the dollar. Add 15 percent for bad debt, unbilled rework and the client who disappears after milestone two, and your honest floor is close to KES 6,500 per hour, about USD 50.

Sit with what that number means

Published rate aggregators put senior Kenyan developer rates somewhere in the USD 30 to 35 per hour band, with the wider range running from about USD 20 for less exposed developers up to USD 45 and above for people with strong cloud and modern JavaScript experience. Those are directional numbers, not gospel, but they are the numbers your offshore clients are anchoring on.

So the arithmetic says something uncomfortable: the market's median rate does not fund the life most people think it funds, once you carry your own overheads, your own downtime and your own tax. Employed developers do not see this because their employer absorbs the 40 percent of the year that is not billable. Freelancers who price like employees are silently paying that cost themselves.

You have exactly four levers. Lower the income target. Raise utilisation. Raise the rate. Or stop selling hours.

Above the floor, price the outcome

The floor is cost-plus. The ceiling is value. The gap between them is where your business either makes money or does not.

Three questions that find the value number, in order of usefulness:

1. What does the problem cost them per month right now? A distributor whose sales reps take orders on WhatsApp and lose two percent of them is losing real money every month. Get them to say the number out loud. If a fix is worth KES 400,000 a month to them, a KES 900,000 build is a two and a half month payback and you should say so in the proposal.
2. What is their alternative? Not "my cousin's guy". The real alternative: a proprietary ERP at KES 2.5 million plus per-user licences, or eighteen months of doing nothing. Price against the real alternative, and put it in the proposal so the comparison is on your terms.
3. What happens if it goes wrong? Payroll, tax filings, stock valuation and anything touching KRA carry risk premium. You are being paid partly to be liable and reachable. That is legitimately more expensive than a marketing site.

Quote three options, never one

A single price invites a yes or no, and clients under budget pressure default to no. Three options change the question from "should I buy" to "which one".

Tier	Contents	Anchor
Essential	The narrowest thing that solves the bleeding problem. One process, one integration, no reporting beyond the standard.	Roughly 60 percent of the middle price
Recommended	What you actually think they need. Named as the recommendation, described in the most detail.	Your real price
Complete	Adds the things they will ask for in month four: extra integrations, dashboards, training, twelve months of support.	Roughly 180 percent of the middle price

Two rules. The tiers must differ in scope, never in quality or care, because "cheap version where I try less" is a promise you cannot keep. And you must be genuinely happy to deliver any of the three.

The price is not the money. The terms are.

A KES 1.2 million project paid in one lump ninety days after go-live is worse than a KES 900,000 project paid 40 percent up front. Things to fix in your terms before you fight over the headline number:

• Deposit before work starts. 30 to 50 percent. This is not aggressive, it is standard, and clients who cannot pay a deposit usually cannot pay milestone four either.
• Milestones tied to deliverables, not to the calendar. Calendar milestones make you the debtor when the client goes quiet.
• Withholding tax. Resident professional service fees in Kenya attract withholding tax at 5 percent, which many corporate and public clients will deduct at source. That is a credit against your income tax, not lost money, but it is cash you do not have this month. Confirm treatment with your accountant and state clearly in the quote whether prices are before or after withholding.
• VAT. The standard rate is 16 percent. If you are registered, say "plus VAT" explicitly. Nothing sours a signing like a client discovering the number grew by 16 percent.
• eTIMS. Corporate clients increasingly need a compliant electronic invoice to claim the expense at all. If you cannot issue one, you are not a supplier they can use, whatever your price.
• Currency. If you quote foreign clients in USD, invoice in USD. If you quote local clients in USD, expect an argument every month the shilling moves.
• Late payment. A stated interest rate on overdue invoices rarely gets charged, but it changes the order in which your invoice gets paid.

When they say it is too expensive

Do not discount the price. Remove scope. "I can get to that number. At that number we do the order capture and the M-Pesa reconciliation, and we leave the dashboards for phase two." You have now taught the client that your price is connected to work, not to how hard they push.

The moment you discount without removing anything, you have told them your first number was invented. Every future number you give them will be treated as a starting position.

There is one honest exception: a genuine discount in exchange for something with real value to you. A public case study with real numbers. A referral commitment. Payment fully up front. Name the exchange in writing so it stays a trade rather than becoming the new price.

Raise prices on a schedule

Do it once a year, on a fixed date, for new clients first. Existing clients get sixty days notice and a smaller increase. If you wait until you feel brave enough, you will never do it, because the day you feel brave is the day you are least busy and most afraid.

The signal you are underpriced is not a full pipeline. It is a full pipeline that never says no. If nobody has walked away from your price in the last ten proposals, your price is too low and the market is quietly thanking you for it.

Your rate is not a measure of your skill. It is a measure of how much of your life you have agreed to sell at that number.