Fed up with adverts from software companies and consultancies that are going to change the world of Planning and solve all your project problems? This series of occasional articles promises nothing but thought provoking questions on how well we understand the basics of our profession from the perspective of a retired planner with 50 years’ experience in the industry whose worked in a variety of UK EPC roles for Clients and Contractors with no axe to grind. We are going to look at PEOPLE, PROCESSES, and PROCEDURES (the old 3P’s – not 4P’s, 5P’s, 6P’s or 7 P’s)that we all have been educated and trained to use in our everyday work (5W’s + 1H) and POKE THEM IN THE EYE / LOOK UNDER THE STONE (the Inverse Universe) and get into trouble with the planning profession because for loads of reasons were not doing so well. If you want to contribute to this soul searching drop me a line with either POSITIVE or NEGATIVE comments (yes, we are PLANNERS that can hold both views at the same time!)
PROCESSES No 9 – Earned Value Management – now over 60 years old – Is there anything new to say about problems in its implementation?
Having looked at how the team can be working hard but recording no progress in blog 1, and some of the problems in using Critical Path Networks in blog 8, we are again looking at our PROCESSES. Underpinning all our analysis and evaluation of progressed work is the use of Earned Value Management methodology. Let’s start by looking at the fundamentals that build an EVM System, point out some of the pitfalls and see if we already have the tools to do better in the future.
Quick history – Following on from the development of CPN (time control) and more powerful main frame computers (in the early 1960’s) it was realised that the addition of budgetary information to the activities could lead to potentially cost control information. EVM in various guises become the leading contender. Led by the US DoD PERT, CSCS, CCCS etc versions were developed throughout the 1960’s (Mosaic do a particularly good history paper on EVM). The original descriptors were all budget function orientated (BCWS, BCWP, ACWP) but sometime in the late 1960’s it got shortened to PV, EV, AC losing the budget function link (pity). Perhaps PBV, EBV, ABC would have been better descriptors. Over the years, it has become clear that EVM has its limitations, but to be of use, EVM has to be applied in an organisation that has a mature project management system. So, as EVM cannot be ignored, lets kick off with some fundamentals.
WHAT ARE THE FUNDAMENTALS REQUIRED OF YOUR EVM SYSTEM –
- Running an EV system within a project / company requires a whole series of Standard Operating Procedures (SOP’s), detailed Working Instructions, skills and experience and employee training. These state how the EV methodology is going to be applied within your organisation.
- We are going to assess the progress of work so we will require a comprehensive Progress Measurement System (PMS) / Rules of Credit (RoC).
- Work will require to be identified within common Project Breakdown Structures through a variety of hierarchical coding structures (Work, Risk, Resources, Costs, Contracts, Assets, BoQ, BIM etc). At some hierarchical level these must all integrate with each other (usually at level 5).
- Costs, both direct and indirect, will have to be recorded at this low level (granularity) within the WBS, (e.g. timesheet bookings / actual manhours, expenses, subcontracts, PO’s etc)
- The finance system (receipts, invoices, payments, cash in / out etc) will have to have this same level of granularity.
- There must be a fixed current Performance Measurement Baseline Schedule on a project which has the same granularity.
- There must be a fixed current Performance Measurement Budget / Estimate on a project which has the same granularity.
- The company and project must have a mature risk allocation system with the same granularity.
- All the Systems must be compatible for data exchange.
- Employees must have the Skill Sets and Experience in using and interpreting the systems outputs.
Remember the Strap Lines about Data
GIGO – Garbage in, Garbage Out
Data must have the 5 C’s – Clean, Consistent, Conformed, Current, Comprehensive
What exactly is EARNED. We are saying that we have achieved a specific objective
What exactly is VALUE. We are using any increasing project function that can reflect progress on a project (e.g. budget, document numbers, material quantities, manhours etc).
And just to complicate matters real project delivery will alter the base function numbers and subsequent CHANGES can REDUCE these achieved percentages or numbers
It must be stressed that the function data - Planned, Earned and Actuals - must be CONSISTENT whether its budget, manhours, quants etc
So, rather than argue about any specific value (which will vary depending on your chosen rules) it’s the TREND in the values from period to period, produced on a consistent basis, that’s important.
That is – they must occur in the same format, to the same value on each granular line item with the same underlying meaning. They must follow the SoP’s laid down to maintain this consistency. So
Not in the approved budget no PV or EV, but AC e.g. unapproved scope changes being worked on
Actual progress but no EV, but AC e.g. RoC have 100% on completion
Less progress recorded, less EV but AC e.g. Approved scope change reduces % complete by increasing total work volume
Not on approved schedule no PV or EV, but AC e.g. rework, work left unfinished
Change not approved, no PV, no EV but AC e.g. change not accepted or cancelled
Valuation items submitted and paid but not budget planned, no PV or EV, but AC e.g. materials on site but not incorporated
No invoice or no accrual recorded, AC e.g. those annoying suspense accounts, wrongly allocated invoices, internal costs from somewhere that will eventually be removed
Quite a list before you start
THE PITFALLS OF EVM is that in the process
it is very good at looking backwards and producing metrics but as a DIAGNOSTIC TOOL for managing the project going forward it is of little use.
LET’S LIST SOME PROBLEMS (akn - Bertrand Guerard - this is what planners do - steal the best ideas from anywhere)
Too granular for executives –
The Miss: Metrics without meaning.
The Fix: Translate EVM into business terms: Will we meet key milestones? Are we burning more than expected? What decisions need to be made now?
Measures progress, not outcomes
The Miss: Progress not Purpose.
The Fix: Connect EVM to business outcomes: Map EV milestones to benefit realization. Add ROI and value delivery KPIs to your dashboards. Review alignment to strategic goals regularly
Ignores risk
The Miss: Performance not Predictability.
The Fix: Combine EVM with: Schedule Risk Analysis (Monte Carlo) Cost Risk Forecasting Confidence curves (P50/P80 dates and budgets)
Shows variance, not root cause
The Miss: No insight not action.
The Fix: Add root cause and corrective action analysis: Link EV variance to issues and risks. Provide forward-looking recovery scenarios. Use traffic-light indicators with commentary
Lives in a silo
The Miss: Fragmented truth.
The Fix: Embed EVM into enterprise systems: Integrate with finance, ERP, and PMO tools. Use portfolio dashboards with EVM layers. Adopt a common data model and cadence
Breaks with scope changes
The Miss: Misleading metrics.
The Fix: Link change control to the PMB: Every approved change updates schedule + cost baselines. Maintain traceability of scope growth/shrinkage. Reflect change impact in executive reporting
Relies on flawed baselines
The Miss: Accuracy illusion.
The Fix: Strengthen your foundations: Use schedule quality checks (DCMA 14-point, GAO). Benchmark cost estimates Review and re-baseline when needed
Ignores resource constraints
The Miss: Progress without feasibility.
The Fix: Integrate resource planning: Load schedules with resource demand. Compare with availability and skill mix. Add resource health indicators to reports
Assumes linear progress
The Miss: Skewed metrics.
The Fix: Choose the right Earned Value Technique: 0/100 or 50/50 for short tasks Physical % complete with QA gates. Weighted milestones for stage-based work
Illusion of Control
The Miss: Reporting comfort not delivery reality.
The Fix: Add leading indicators: Quality metrics and defect rates. Team health scores. Stakeholder sentiment
WHAT CAN WE DO BETTER IN THE FUTURE - NOW!
- Earned Schedule – now over 20 years old it addresses some of the time aspects of EVM when the project approached completion and SPI trends towards 1, but by then you’re probably working off a snag / punch list / database anyway.
- Improve indices nomenclature –
- WHEN was that indices / variance produced – use a superscript to indicate Period 6 of Period 29
- WHICH curve (ES=0%), 50/50 (ES=50%) or Late Curve (ES=100%) is being used for PV – use a subscript to indicate
- or you can include CRITICAL activities (TF<20 days) only in your calcs.
- Look at other cost based functions – WBS level commitments happen long before progress is recorded – If you don’t commit at the right rate, you’ll never progress at the right rate.
- Why use budget function at all when they can be so unreliable
- Include other functions – manhours and quantities are obvious candidates
- Improve forecasting – What do management have to do to get back on plan and by when do they have to decide? - introduce intermediate date forecasts indices for schedule recovery work – (see IPSYS Analytics work on 30, 60 ,90 day what-if look aheads)
- Integrate the basic systems that provide the EVM data – PMIS, Data Mining, AI. Oracle Cloud, Finance, Procurement, Quality
- Look at the data quality and granularity – Are management getting the right information at the right level to make decisions in a timely manner?
- Do are KPI’s / Indices focus on prospective work?
- Increase skills and knowledge and educate senior management in EVM
STANDARDS
Various EVM standards exist
EIA – 748 – D (E just published) – Primary for USA DoD, which requires an Interpretation Guide (EVMIG – 2019) and fits into a whole suite of documents, recommended practices and pamphlets (too numerous to list) published by GAO, DCMA, PMI, AACE, DoD etc. and is very much about compliance for DoD projects.
ISO 21508 – 2018 is very much a guidance document.
APM, RIBA, CIOB, RICS all publish EVM material besides the hundreds of books, academic papers and conferences held on the subject.
Reference here response file - Peter Holroyd – via LinkedIn
Question to you all –
- Is EVM mandated on your project by contract or company?
- Is your time spent 90/10 analysing the past or 10/90 optimising the future?
- Do you find that its conclusions are acted upon or dismissed?
For my next blog (No 10) in PROCESSES I am taking a look at – REPORTING.