Forecasting accuracy can be improved significantly. Accuracy is often poor since forecasting is usually done on the basis of subjective judgment of the sales people, not necessarily on verifiable material.
Installing a sales force automation system or CRM will not improve your accuracy. You see, CRM depends on your inputs into the system and if your inputs are subjective, then it does not makes a difference. A CRM system can track historical closing (winning) rates of sales people and then apply a certain probability to all current opportunities. But, then you need historical data. Even then, its predictions cannot be accurate, as circumastances and even sales people change.
Unless you have a sales process and the CRM system incorporates this process it will not be successful. Most people buy a CRM system, without having a sales process and then struggle with the CRM.
Your revenues in a year will consist of the following:
1. Repeat business from existing customers (easy to forecast, just have dialogues with them);
2. Non pipe line revenue (such as contract renewals, unsolicited orders, walk in customers, license fees, upgrades, etc). These are easy to forecast, just extrapolate from the previous year;
3. Active opportunities, where you are actively competing with competition, and your winning chances depend on a host of elements such as your relation with key decision makers, your value proposition as understood by customer, the risk factors, etc. Here it is possible to make a realistic assessment of your chances if you are able to assess each of these elements. You might need to involve the management in the process and you could even turnaround opportunities where you are not well off. But I would recommend that you compete in opportunities where your chances are good and qualify every active opportunity, before deciding to compete. If you do this consistently, you will be able to make accurate predictions.
4. The pipeine opportunities, where you are deveoping an account from the beginning. Here your chances are very high. But the forecasting can be accurate only if you follow a consistent sales process, which has measurable and verifiable milestones. Based on the verifiabe evidences you have, you will be able to measure the revenue that could be realsitically expected from each account you are developing. Ideally you shall be focusing on these accounts, rather than on active opportunities where your chances are very low. The milestones will classify each opportunity into A, B, C, D, etc depending on the stage you have reached and the evidence of reaching that stage. And each stage will have a certain probability. This estimate could be done for the length of the sales cycle, say 4 months, and for the remaining months in the year, you could extrapolate.
It is realistically possible to improve forecasting accuracy to around the 90s. It is true that, like you said, forecasting accuracy is often around 50% for most companies. This is because they dont follow a scientific method as above, and just depend on the subjective assessment of their sales people, which is subjectively modified by sales managers, and then intuitively/ subjectively modified further by senior management.
Please let me know if you need some more clarifications. Kindly check my blog
www.missionselling.blogsource.com for similar discussions.
