Vous êtes confronté à des divergences de données CRM dans les prévisions de ventes informatiques. Comment allez-vous garantir des prédictions précises ?

Start by reviewing the data entries for completeness, consistency, and accuracy, identifying any errors or gaps that could skew forecasts. If you find that sales stages or deal values are inconsistently recorded, standardise these entries and correct inaccuracies. Cross-reference CRM data with other sources, such as sales reports or financial records, to validate the information. By systematically auditing and cleaning your data, you improve the reliability of your sales forecasts, enabling more accurate predictions and better decision-making.

La limpieza de datos es fundamental en todo sistema y el CRM no es la excepción. La correcta sincronización de los datos entre diferentes sistemas es indispensable. Para limpiar los datos, se deben definir reglas de cumplimiento de los datos, que permitan identificar errores y corregirlos. Además, de implementar un proceso, que facilite la integridad de datos a lo largo del proceso. La definición de reglas facilita la auditoría y permite identificar el porcentaje de cumplimiento de los datos y el avance en la aplicación de correcciones. Actualmente con la IA, es posible identificar las discrepancias en los datos que permitan una definición de reglas de forma más sencilla. También, podría utilizarse IA para la prevención de inconsistencias.

It always starts with what's being entered. Who is doing the entering? Have they been taught properly? It's the same old story: BS in, BS out. If there are discrepancies, what is the root cause? What is the data discrepancy? If one is trying to forecast, one must have a starting point if that starting point is corrupted or wishy-washy at best. You can't predict... all forecasting needs to have "SMART" data. But More importantly, all stakeholders need to know and understand what the "SMART" data entails... And that's a training thing.

Establish clear guidelines for how data should be entered into the CRM system. Create standardised fields for entering deal stages, customer information, and revenue estimates, ensuring all team members follow the same protocols. Provide training to your sales team on these standards and set up automated validations within the CRM to flag any inconsistent or incomplete entries. By enforcing consistent data input, you minimise discrepancies, leading to more reliable data and, ultimately, more accurate sales forecasts.

While standards and training are important you can improve data quality by implementing picklists and validation rules on your most important fields. This ensures data can easily by sorted, filtered, and analyzed without the need for cleanup. Additionally, if your CRM is not connected to an external system it allows you to easily blend, edit, and import data through the use of simple Vlookups or primary keys.

Standards will only work if they're embraced company-wide, and that means leadership. If leadership is seen not to be embracing CRM and the 'input standards,' how can one champion CRM accuracy? There is nothing worse than having meetings on forecasting when leadership does not embrace standards. Data is great, but if the data is haphazard and sporadic at best, how can one be accurate?

The only true and effective way is to have a process champion. This champion lives and breathes the message. However, this champion needs the authority to effect the message about effective data and the power of good data, no matter who they are talking to. There's no point in having a line manager (who is the chosen champion) who is required to act on the message if the VPs are not inputting accurate data.

Before even engaging in statistical testing, it's important to ensure that the different departments have the same underlying data. Sometimes (I've seen it happen) you'll run into problems where the CFO, payroll, and sales all have totally different SKUs and definitions of things like ARR, TCV, and ACV. After you've ensured those processes, data points, and record histories line up, then you can move on to statistical analysis.

When I see CRM data not matching up in IT sales forecasts, I start with an "Analytical Review." Instead of just trusting the system, I carefully look at the data from different views. I compare past sales, customer actions, and market trends to find mistakes or missing info. I also ask my team to regularly check and clean the data. This extra step makes our forecasts more accurate, which means we can plan better and reach our goals easier. It might take a bit more time, but it really pays off.

Feedback, first you need: Regular Data Audits: Conduct frequent audits to identify and correct discrepancies early. Cross-Referencing Data Sources: Compare data from multiple sources to validate its accuracy. Data Cleaning: Implement a routine data cleaning process based on your company’s Service Level Agreements (SLAs) to remove duplicates, correct errors, and update outdated information. Data Validation: Use automated tools to validate data as it is entered into the CRM. Data Governance: Establish clear data governance policies that define how data should be handled. Learning & Development: Provide ongoing training for your team on the importance of data quality and how to maintain it.