Month Three of a Six-Month County ICT Project
A county government has contracted a vendor to deliver a revenue-collection system in six months for KES 18 million, paid in four equal instalments tied to milestones. At the end of month three, the vendor has been paid KES 9.5 million but has delivered only the login module and a database schema, which the ICT officer estimates at 25% of scope. The vendor now requests a KES 4 million advance ahead of the next milestone, citing rising input costs, and separately offers to 'facilitate' the ICT officer's upcoming professional certification course fees. The system is meant to go live before the next financial year, and the county has no in-house developer to take over the work.
1. Assess the project's cost and schedule performance using earned value, and state what the numbers do and do not tell you.
Read the worked answer
Figures - BAC = KES 18,000,000. At month 3 of 6, planned value PV = 50% x 18,000,000 = KES 9,000,000. Earned value EV = 25% x 18,000,000 = KES 4,500,000. Actual cost AC = KES 9,500,000.
- CV = EV - AC = 4,500,000 - 9,500,000 = -KES 5,000,000 (over budget)
- SV = EV - PV = 4,500,000 - 9,000,000 = -KES 4,500,000 (behind schedule)
- CPI = EV / AC = 4,500,000 / 9,500,000 = 0.47
- SPI = EV / PV = 4,500,000 / 9,000,000 = 0.50
- EAC = BAC / CPI = 18,000,000 / 0.47 ≈ KES 38,000,000
What the numbers say. The vendor is spending roughly two shillings for every shilling of completed work, and is progressing at half the planned rate. On current trend the project needs about KES 38 million and, at 25% in 3 months, roughly 12 months total.
What they do not say
- Whether the 25% estimate is reliable. It is one officer's judgement against a scope baseline; if the milestone acceptance criteria were never written, EV is guesswork and the CPI should be treated as indicative, not precise.
- Whether the vendor is inefficient or has been drawn into unplanned scope. Those require different responses: re-planning versus performance management.
2. How should the ICT officer respond to the advance request and the offer of course fees? Address both the commercial and the ethical dimension.
Read the worked answer
The advance
- The contract pays against milestones. Paying KES 4 million ahead of delivery moves all remaining risk onto the county: if the vendor stops, the county has paid KES 13.5 million for 25% of a system.
- The correct response is to decline the advance as requested and offer alternatives: verify the next milestone's acceptance criteria, agree a short accelerated delivery plan, and if cash flow is genuinely the issue, consider payment against a verified partial milestone with a bank guarantee rather than an unsecured advance.
- Any variation must go through the contract's change procedure with documented justification, not a letter.
The course fees
- Decline. This is a benefit offered by a party that has a live commercial dispute with the officer's employer and is currently seeking a financial concession from that officer's decision. Even if the officer believes they can be impartial, accepting creates a conflict of interest and the appearance of one.
- Disclose the offer in writing to the accounting officer and the tender committee, and record the disclosure. The disclosure protects the officer, the county and the vendor relationship.
- If the county genuinely wants to sponsor professional development, it should do so through the county's own training budget and procurement process, with no vendor involvement.
Underlying principle. The test is not whether the officer would be influenced, but whether a reasonable outsider, knowing both the offer and the pending advance request, would doubt the decision. Here they clearly would.
3. Given no in-house developer and a hard go-live date, what recovery options exist and how would you choose between them?
Read the worked answer
Option A - Performance-manage the existing vendor. Issue a formal notice of delay, agree a written recovery plan with weekly deliverables, withhold the next milestone payment until verified, and apply liquidated damages if the contract provides them. Retains product knowledge and the existing code base. Risk: the vendor's cost problem is real and may not be fixable by pressure; EAC of ~KES 38 million suggests they are losing money, and a loss-making vendor is a vendor that walks.
Option B - Re-plan scope, not just schedule. Deploy a reduced first release (registration, assessment, receipting) by the go-live date and defer reporting, dashboards and integrations to phase two. This is the only option that protects the deadline while acknowledging that 12 months of work cannot become 3.
Option C - Replace the vendor with a turnkey supplier. Cleanest commercially, slowest to start: procurement, due diligence and knowledge transfer typically consume 8-12 weeks, which does not fit the stated deadline.
Option D - Source escrow and handover. Require delivery of source code, documentation and a deployment environment as a condition of any further payment.
How to choose. Test each against three questions: does it fit the go-live date, does it reduce the county's exposure to a single failing supplier, and does it preserve the option to change supplier later. Option B combined with Option D scores best: it delivers something usable, and escrow means the county is never again unable to walk away.
Non-negotiable first step. Do not pay the advance. Whatever option is chosen, further money must follow verified delivery.