Despite this, companies often still adopt a “go with your gut” approach to forecasting, foregoing analytics for general trend sentiments. So, to help you start moving towards more accurate, and effective, forecasting, we have compiled a number of key recommendations and tips.
Sales forecasting – What and why?
In simple terms, a sales forecast is an estimate of sales, individual or firm-wide, over a given period of time. That definition is flexible, and can incorporate team-based forecasts, biannual predictions and more.
Firms use forecasts to gauge their business going forward, and can employ this information differently depending on staff level. A manager might work on improving individual predicted rep sales, while a C-level executive might make global strategy adjustments. The point is to have well-defined and reliable information so that the decisions made can be tuned for best effect.
You shouldn’t underestimate the consequences, and benefits, of this information. If a director notices that a department is on a downward trend, they can act to revitalise the ailing sales strategy. That way negative outcomes can be headed off, rather than having to be mitigated later.
This is a fairly intuitive benefit, and one most of you will be familiar with, but it goes deeper than just calamity dodging. Your HR department can rely on sales data to adjust recruitment rates, so that you are fully staffed for a big sales peak. Or alternatively they can increase spending on training and skills, if reps are having a difficult time competing in a market.
Lastly, but perhaps most importantly at the employee level, is the issue of productivity. Sales predictions are tangible, and help staff keep on top of their sales trend. If sales predictions are poor, they can make the adjustments they need to stay on top of their quote. It can also be a potent motivational tool, letting staff know their performance ahead of time.
What’s in a forecast?
A sales estimate is just that. An estimate. It doesn’t need to be perfect, and you can’t expect it to be. Sometimes you will see a spike that never materialised, or a plateau that turned out to be a climb. It happens. There is a difference between real, fallible data and bad data however, and you need to be aware of that fact. On top of that, even good data can be used incorrectly. So, assuming you’re not fudging the numbers, here’s what should feature in any good sales forecast.
- Sales process
It’s critical that everyone is using the same playbook in a sales team. If reps are improvising regularly, or their processes are poorly defined, then your estimates are going to be guesses, and poorly-educated ones at that. Make sure sales are consistent, principled, and use coherent stages.
There’s no point knowing what your sales figures will be like if you have nothing to compare them to. Make sure that you have well defined quotas for your team, so that you can tell whether a dip is a dip or a sinkhole.
What is a lead, really? Where does your funnel open and close? You need to know this if you want to start crunching the numbers. So dust off your definitions off, and make sure they are well understood by staff.
So what happens next? A staff member sells way above quota. Or, they miss several forecasts in a row. You need to be prepared for this eventuality, and to have your resulting actions ready and waiting. If not, the value and force of your forecasts is going to diminish. This will be especially true in the eyes of staff. Remember, a number is just a number if it doesn’t affect anything.
- CRM integration
Surprise! It’s your CRM again. Integrating your sales database gives you great opportunities to really fine-tune your predictive ability, so make sure you are making full use of your data.
Data- what to look out for
The information you have at hand is critical, but many factors can cause it to change suddenly. So don’t panic if you see a sudden trend, and look for a change in one of the following:
One of the single most common issues with forecasts is that they are often left unchanged after a policy shift. It stands to reason that if the way you do business changes, then the level and nature of that business will too. If you drop Rep compensation, expect to see a drop in sales, and make sure your forecasts incorporate this.
The consequences of a change can be complex, so make sure to fully estimate the potential impact before committing to a new forecast. Sometimes a reaction won’t be apparent for a few months, while reps adjust. Or, you might see an opposite effect. So stay vigilant!
Every rep is a unique revenue stream, so don’t be surprised to see some shifts after a reshuffle. If you start rationalising, expect to see downward trends materialise, and take that into account beforehand.
New areas and markets can be unpredictable, so plan accordingly. Your product might sell differently on the other side of the country, or your reps might find it difficult to find leads. So plan well, and feed the relevant data into your projections.
- Markets and Industries
Sometimes things outside your control can start muscling in on expected growth. So while any good forecast will allow for fluctuations, it is probably best to start from scratch if you can foresee a long-term industry shift. This goes for regulatory changes too. If the dynamic of the market changes, outdated estimates will probably do you more harm than good.
- Product and competitors
Unless you are particularly lucky, you will probably be competing within your market. At the very least, there are two agents of change in this set-up, you and your competitors. So stay on the lookout for competitive advantages that you might accrue, or a new product offering from the main competing firm. Forecasts can hinder rather than help if you start using them to reject new realities.
- The customer
The key element of your sales forecasts, the customer should take prime position in your list of changes to look out for. Whether it’s your firm reacting to customer needs, or just simple seasonality, you can expect to see a lot of dynamism here. So while you shouldn’t feel the need to throw out your dataset every time a feature rolls out, keep an eye on the trends following to see if your projections are robust.
The different shades of Forecasting
One forecast is not the same as another, and you should make sure that your process is working for you. Here, we have assembled 5 of the most widely used systems for you to consider.
- Sales Cycle Forecast
A sales cycle forecast allows you to predict when your individual sales leads are likely to close, based on their age. This is a fairly straightforward analysis, relying on average sales cycles to help predict deals. So if a sales cycle is 4 months, and you have been engaged with it for 3, a sales cycle analysis would suggest a fairly high chance of success on that deal.
The main draw of a Cycle forecast is that it prevents “gut” feelings from interfering with you analysis. Too often a rep might, for any number of reasons, make a decision on the likelihood of a deal early in the process. A sales cycle nips that in the bud by providing a firm structure to gauge against. So no more “surprise” failures that seemed so promising 1 week into a 6 month cycle.
It is recommended that you have a CRM that can interface with your marketing software for this process, as you are going to need a lot of raw data to fuel predictions. Make sure to segment based on markets and sales categories, so that the predictions you get are practical. Broad averages here won’t do you much good.
- Opportunity stage
This is a similar technique to the sales cycle, but this one relies on your pipeline designations to create predictions. It creates average success rates per stage, and compares them to the current stage of a lead. You might find that your demo customers generally come back, so that should be a good indicator going forward.
One peculiarity of how this system creates its estimates is in the value. Say you have a €2000 deal, in a stage where it is 50% likely to close. The value you would assign to that is €1000. This is a fairly quick way to get a general figure for your forecasts, and is also quite easy to derive from existing data, assuming that you have it.
However, you should be wary of the accuracy of the results you get. A deal might have stalled in a stage, but this analysis won’t take that into account. So make sure that your pipelines are well maintained so that junk data isn’t entering your predictions.
- Historical forecasting
This is one of the simpler forms of forecasting, and while its results can be suspect, it is a useful way to get a quick preliminary forecast.
The way this works is that, to predict a period’s sales, you take a similar historical sample and “translate” its figures into the present. So if you made €40,000 in sales last October, you can use that data to assume a figure of roughly €40,000 this October.
This of course does not take into account a wide variety of factors, such as market conditions or seasonality, so this can leave you vulnerable to sudden shifts. It’s best to use this as your yardstick before the serious forecasting begins.
- Intuition forecasting
While some of you might be wincing at this one, especially given what we have said about intuition, this is still a very common form of forecasting. Like the name implies, it relies on reps using their experience and intuition to gauge the potential length and value of a deal.
While reps can be very good at making estimates, this is in essence a guessing game. The data you get will be unverifiable and ahistorical, so this is best suited to firms with limited existing data sets.
- Multi-Variable testing
This is the most complex forecasting technique generally employed by firms, and usually involves combining the other systems mentioned here. So you might take age and stage into account, and then allow for variation based on individual rep performance.
The benefits of this are clear. A rep in the early stage of a short sales cycle length will get a boost in success likelihood, while a rep who has progressed to a later stage will see a more reserved success rate if it is a typically long sales cycle.
This approach allows you to combine the various metrics listed here, or even introduce your own, to really drill down into your data when forecasting. A key challenge in implementing the technique, however, is that it will require a significant amount of data on your part. This is unlikely to be feasible for smaller organisations, but more than justifies itself to a well-developed sales firm.
So there you have it. Sales forecasting is a great way to boost your sales performance, and is something that can be achieved by any firm, in one form or another.
If you are a new firm looking to get into forecasting, or even a more mature firm looking to tighten things up, then you may have noted the recurrence of factors like pipelines and likelihoods. It is important to keep on top of this, so our partners at HubSpot have helpfully prepared a Pipeline tracker that you can use to get started. It includes spreadsheets for deal tracking and revenue forecasting, and lets you get familiar with all the variables you will be looking for. If you are looking to go beyond this, quality CRM product is recommended. This will greatly streamline the process of forecasting for you, and help prevent data duplication and junk data.
So get forecasting, and watch your sales efficiency trend upwards!