If effective and simple resource management is the Holy Grail of professional services automation (PSA) software, the ability to produce a revenue forecast worth reading must be the golden Unicorn’s horn.
Why is this so hard?
Well, first let us discuss what is meant by a revenue forecast and the many equally valid interpretations people place on this seemingly simple concept.
The following four definitions are variously used by companies for revenue projections:
- Hard (or Committed): Minimum value of all contracted work in hand, less value of contract amounts already booked (including open balances on prepayment funds, considering also future recurring fund topups to end of current contract)
- Planned: Value of work in hand (as #1) plus changes verbally agreed but not yet documented or entered into the system
- Expected: Expected value of work in hand including known changes, logical extensions and future commitments contracted for but not yet committed (subsequent phases etc)
- Potential: #3 plus scenario modelling of contracts under negotiation and base-loading of regular minor changes from existing customers, where such events are common enough to rely on and be prepared to resource for
All four values are equally valid, depending on the question asked, the intended use of the report and, most importantly, the current business environment (quiet to highly active).
If times are quiet, and costs are a concern, the hard value would be used for staff and contractor management decisions. If business is highly active and the company is growing, “Expected” or even “Potential” values would be used for recruitment planning.
All of these forecasts suffer dramatically from data fade, calibrated by the typical operating model of the business.
If the company engages in large numbers of small or short-term engagements, the ability to look forward will be severely restricted by the short-term nature so the “Hard” value is likely to be small with an event horizon in weeks or even days. Growth planning will be totally driven by an understanding of the repeatable nature of the work and the market conditions rather than allocation of work in hand.
Where a business has a mix of long and short term work, a better balance will exist, however, the likely event horizon even for the “Planned” value of the forecast may still be limited to a few months only.
Generally, annual budget projections will be based upon “Potential” values with logical trend-based extensions to compensate for data fade calibrated by the type of business the company engages in.
This is why the topic is so hard and closer to a conjuring trick than science.
How can your PSA help?
If (and this is a large if), all work in hand is correctly modelled against a timeline and kept up to date with each executed change order or new piece of business signed off, then the PSA will have some chance of delivering a Hard forecast. Even with this exceptional (and frankly unusual) level of adherence to data maintenance, your PSA will still struggle to generate a Hard forecast for prepayment funds, rate card work and minor changes. If these form a material part of your workload, your Hard profile is likely to be missing the natural churn component of your business.
If you have good proposal profile modelling and the ability in your PSA to build what-if scenarios, you will have an opportunity to produce a range of “Potential” work profiles, providing your project managers also do the same modelling for change orders and likely extensions of existing work. However, this latter is a common weak point in organisations, Material proposals get modelled as part of the sales process, but expecting PMs to do the same level of modelling on existing work, on an organisation-wide reliable basis, is tough, very tough.
And, even then we have another problem to overcome that most companies have no way of considering systemically: productivity.
Just because someone is booking to a task with a notional rate, the allocated time multiplied by rate does not necessarily equate to recognisable revenue. Anyone who has worked in a professional services business will understand the difficulties of either achieving the expected margin on a fixed price project, or getting time and materials billing paid without challenge.
Time booked does not equal recognisable revenue and in some cases, it can fall materially short. Even a Hard forecast from planned work time can actually be optimistic as it will naturally assume perfect productivity to budget. Timing of revenue recognition (under ASC606) should also be considered in your revenue forecast and while this should match resource profiles, so often it doesn’t.
Also, projects slip, in particular where customer document reviews and actions (such as testing) are involved. The further out your forecast profile goes, the more errors it will contain and so the less value it has.
Lastly, this is a problem where “state” matters. Some reports naturally have state based on process (such as number of tickets opened and closed in a period, the ticket process automatically maintains state for this report). A revenue forecast report cannot rely on process-driven state as the process of project delivery is not tightly coupled to maintenance of up to the minute resource views. Simply running a report or viewing a graph on-line will give you some numbers, but will not provide the state definition telling you how meaningful the data is.
While your PSA may be able to model a certain portion of your project delivery, beware of taking any forecast report too seriously. At best it will be partial, at worst it could be totally wrong.
Is there an answer?
The most meaningful revenue forecast is one produced by the project manager as part of their monthly reporting cycle (thus achieving consistent month-end state). They understand exactly what is going on and are best placed to call the revenue numbers. They know the productivity and the timing of the recognition milestones, they understand the subtleties of the first three forecast buckets and so can enter that data without confusing the workload allocation views used for resource management. A team manager dealing with churn activities is the one most able to call the expected volume of that activity. Numbers you will never get accurately from resource allocation based views.
Such a profile is also simple to produce on a quotation so the what-if scenarios are also supported in the same process.
The answer is therefore not to expect a resource view to magically generate meaningful forecasts, but rather to operate an independent fact-based process, informed but not driven by resource views.
About the Author: Harmony Business Systems Ltd (HBS) is the company behind HarmonyPSA, the most complete cloud PSA software on the market. Developed with functionality to cater for even the most complex needs of MSPs, VARs, ISVs and Professional Services organisations, HarmonyPSA truly is the next generation of PSA systems. HBS is an independent company based in the UK. Follow HarmonyPSA on Twitter or LinkedIn