[Salesforce] Maximize user adoption with a successful data integration

Salesforce is a powerful CRM tool that provides lots of features to help you get the most value out of your Customer Relationship Management process. Among other things, it can help you with:

  • Establishing a centralised, up-to-date, customers/prospects and contacts database.
  • Giving an overview of your current business pipe and some hints about what it might look like for the coming months.
  • Providing a robust and customisable data-privacy policy mechanism.
  • Getting an unified performance review among team members or between teams.

The list can go on and on, and there are plenty of reasons for a company to decide to start integrating Salesforce in its business process. Nevertheless, one must know that Salesforce is a good return on investment as long as you take care of its integration. A poorly deployed Salesforce platform can rapidly become a thorn in the side of your sales team. Because you are likely to put lots of efforts to customise, deploy and maintain your Salesforce, you’d rather be sure that it genuinely brings business efficiency.

This article will pay particular attention to data-driven user adoption as it holds a significant impact on any tool integration.

If you are in charge of deploying Salesforce in your organisation, you might have to deal with the following:

First, you’ll have to go through a “process translation” stage where you’ll try to find out the best way to match your organisation’s processes with Salesforce features.

Then, you might need to customise your Salesforce interface afterwards to fit your needs and satisfy your expectations.

Once you have a proper platform that can support your business activity, you’ll need to plan some training session for your sales team.

Finally, you would – and should – want to integrate data from other applications of your IT system. You must be able to access and trust data on Salesforce. It will build trust around reports and dashboards, that will become useful helpers for business decisions.

It becomes critical to know how to manage information before going through the development and deployment stages. Make a solid data-integration plan can help you in the future avoid data complexity.

What is data complexity?

Yes, data can be complex. But how can you estimate the data complexity within your organisation? We can distinguish two main axes that will help us identify the situation we find ourselves in. We’ll talk about volume and diversity.

Defining “simple data”: unique or limited number of sources

Indeed, having a single data source to deal with, reflects process simplicity. There is no struggle fetching records nor running analytics. Sharing knowledge is simple as based on a consistent, almost unique, source. There is no room for anything else than trust in information. Also, your data might already benefit from a similar format, and won’t require multiple processing operations to be used uniformly.

Figure 1: Simple data integration

Defining “complex data” :  multiple sources

Multiple sources imply:

  • Numerous formats
  • Various synchronisation processes
  • Potential data dependencies across sources

In this case, it is a bit trickier. Indeed, you’ll have to fetch information from different endpoints in a unified way. At this point, we start talking about data complexity. The more sources, the more complexity. Data can come from very different systems, making synchronisation harder to deal with and you might have to go through multiple process steps before having usable data sets. Not to mention that these actions might need to be scripted depending on your synchronisation frequency.

Figure 2: Complex data integration

Moreover, such scenario let duplicated data exist in different systems. Meaning that a synchronisation issue can directly affect user adoption as it breaks user trust be giving him different information from one tool to another. A good synchronisation policy and a good knowledge of your organisation systems are required to manage such implementation.

Unfortunately, this is what is likely to happen when integrating Salesforce. You’ll want data from other systems (invoicing, staffing, […]) to appear in Salesforce and you’ll want to make links between data in Salesforce and other applications. There is a risk for your organisation data complexity to increase during the process.

What scenarios can drown user adoption ?

A successful Salesforce integration implies users willing to use the platform. Let’s put aside Salesforce ergonomic and functional logic to focus on your company integration. What factors could slow the craze for Salesforce integration?

Not connecting all data sources to your Salesforce organization

figure 3
Figure 3: Drawbacks of missing data source

If you choose not to/are not able to synchronise one or more data sources, you’ll end up with incomplete, wrong computations on your Salesforce platform. It will have several repercussions such as:

  • Not being able to rely on Salesforce dashboards to drive your business decisions.
  • Being forced to log into another tool to get the missing information and lose efficiency in this time-consuming step.
  • Having users work on Excel or other software and then report their work on Salesforce. Which will immediately lead your users to stop using the CRM.
  • Break any trust in data displayed on the platform at all levels of your organisation.
  • One of the worst cases would be; displaying data that is partly correct and use this altered information set to work with. Thus leading to a discrepancy between Salesforce and the reality of your business.

Having inconsistent data between Salesforce and other applications

figure 4
Figure 4: Drawbacks of inconsistent data-sets

When integrating data on Salesforce or from Salesforce your are likely to encounter data inconsistency. This may occur because of :

  • Technical issue during the synchronization process
  • Missing information from one system
  • Different sources used across synchronisations
  • Short delay when syncing target with source data set

This can have a huge effect on user adoption as it breaks data trust. User won’t refer to Salesforce data has they know it might be inconsistent and in the worst scenario they might loose track of which system to trust and therefore reject both.

When dealing with data variations from a system to another, it is essential to communicate with your colleagues once the bug is solved. Make it clear that there are no more discrepancies and don’t hesitate to explain the data transfer process and specify data sources. All these details will rebuild trust in your IT system.

If you happen to have some long syncing periods, you might want to explain it directly on your Salesforce platform or any channel between IT and Sales teams. Aware users are more likely to be comprehensive.

Not reporting Salesforce information to other applications

figure 5
Figure 5: Drawbacks of isolating Salesforce data from other applications

A full integration implies both way syncing! You must be able to view Salesforce activity from other applications as well. Users must feel that the work done on Salesforce is used elsewhere in the organisation. To show that it has an impact on management and business growth. Sharing Salesforce data will give visibility of Salesforce usage and enhance user adoption like a snowball effect. Furthermore, it will reinforce the idea that your organisation has a strong capacity for third-party tools integration and can provide a unified IT service.

Not sharing data sources and business process

figure 6
Figure 6: Having different interpretation of the same data-set

You can have everything correctly plugged in, real-time synchronisation, live reports and automatic overview dashboards and still struggle with Salesforce adoption. Why? Processes.

If you don’t share your processes among your teams, you might end up with a misunderstanding of a correct data-set. The process/logic behind the reports and dashboards should be considered as part of the data integration. If not, different visions from different teams of the same information can lead to frustration between team members. Do not hesitate to communicate about the syncing process, explain the sources, the links, the matching between fields – because a label might differ from one system to another.

Now that we’ve said that, proper risk mitigation is to prepare your data integration. Therefore, let’s answer the following questions.

How to perform a successful data integration ?

How can data from multiple applications, self-hosted or cloud based, can be integrated into Salesforce ? There are several questions to ask before starting data integration :

  • Which information needs to be uploaded?
  • Are data sharing policies respected?
  • How will the uploading process go?
  • How often should it be done?
  • Is it one or both-way synchronisation?

Which information needs to be uploaded?

Get to know the target users. What do they need? This question takes long hours of meetings and discussion to answer. Usually, more than one business, sales, administrator teams are going to cooperate through Salesforce. The idea is to learn what is the current process and to find out which data to use.

Moreover, different team might use the same data-set but with a different interpretation/vision. How to make sure that everyone access and understands the data once on Salesforce. Do they speak the same language? Are they from the same line of business? Do you need to define a standard reference when naming specific field labels?

Are data sharing policies respected?

Data access restrictions are usually already in use in your organisation. Users access data from applications based on their privileges or confidentiality level. You might also have confidential data-sets that are only accessible by a restricted group of users. Make sure that you know these rules when preparing your Salesforce data integration project. You don’t wish to erode user adoption with inconsistent or missing data on the platform, but on the other hand you also want to avoid breaking user trust by bypassing existing sharing policies.

How will the uploading process go?

This step will mostly depend on where is the data hosted? Is it on self-hosted systems on which you have full control? Is it cloud hosted and you’ll face technical restrictions? Or worse, is it on a shared Excel spreadsheet that had been exchanged by emails over and over since the beginning of time?

Here are the available options:

  • Manually: Not the best way to integrate a large amount of data, not efficient, time-consuming and highly likely to induce human errors in the process. But might be a suitable alternative for a small amount of data that is either part of a system that doesn’t provide any data-pulling process or if the process is too complicated and require more integration time than the data duplication itself.
  • Uploads with a data-loader: There are two ways for uploading data on Salesforce; using Salesforce data-loader or Salesforce upload feature. The data loader is a rich client application that can manage data insertion, update & deletion on a large scale. It provides success and error reports. This solution accepts CSV files and allows key-binding configuration. It is an excellent alternative for non-tech users but requires an understanding of the platform & company process.
  • Using Salesforce API: The most “techie” way. Not as suitable as the data-loader for large scale insert or edit operation though. You can configure an OAuth-Connected application on the platform, it will provide OAuth keys and client code. The API is an excellent choice for data pulling as you can execute SOQL queries directly from the API call.
  • Using a third-party tool: Some software might provide a Salesforce module integration for data syncing.

How often should it be done?

Before diving into data integration, you should define how often the syncing process needs to run. It might be criteria for choosing your data transfer process. For instance, you’ll prefer an automated way that requires a bit of development instead of manual interactions if the process has to run daily. It will cost time and efforts in the beginning but will be effective over time.

Is it a one or both-way synchronisation?

Last but not least; what is the type of data integration. Do you need to insert elements on Salesforce for the project launch? Will you need to inject data on the run? Will some systems require to read data from Salesforce as well? A best practice would be to have a graphic representation of all systems that need to communicate with Salesforce and for each, report the synchronisation rate.

These questions tend to avoid data complexity issues. Having these elements will make your data integration easier to perform and maintain.

How to ensure a sustainable data quality?

As you might suspect, data integration helps you reducing your data complexity and gives you a better start when choosing Salesforce as part of your information system. But this is not enough; the next challenge you will have to face is ensuring data quality. Hopefully, your CRM will be widely used in your organisation, and you’ll end up with lots of user input. You want to be sure that newly created records are meeting the expectation you have set during the data integration stage. Some control and cleaning operations will be required to maintain a high business value in your data.

Ensure data access policies are applied

You will have to make changes in your configuration. These evolutions might bring new questions on data sharing policies. Make sure your team members have access to the records within their scope and think twice before adding or removing sharing rules.

Ease & clean user input :

As far as we know, we can’t wholly rely on user input to ensure data reliability. Some systems require a lot of data to function and this time-consuming task can tired users out, making fewer efforts in filling information correctly. There are options to save time and efforts:

  • Make use of default values, checkboxes & Picklist if the field has a restrictive value list.
  • Mark field as required for mandatory information only
  • Fill the Help Text attribute if the field’s label if not self-explanatory
  • A bit more complicated but very helpful : build a Path to guide users
  • Use Process Builder to set automated update rules and reduce user interactions
  • Use Validation Rules to enforce field formatting and values

Prevent data duplication :

Your Salesforce record number will grow over time. It is essential to control this growth to maintain a consistent data referential and avoid data duplication.

  • Define your model to follow the DRY dogma – Don’t Repeat Yourself – prefer object references over duplicated fields on separate objects.
  • Use Salesforce’s duplicate Management module to define matching rules and duplicate rules.

Constantly control data quality

It is important to control data quality in the long run. That way you’ll make sure that your reports serve reliable information. There are a few App Exchange applications such as Data Cleanser, Data Quality Analysis Dashboards or Validity Demand Tools that can help you cope with data monitoring. The data loader utility is very handy when performing large scale data updates.

In conclusion

Data integration is only a part of the whole process of moving to Salesforce. Your team will face many challenges to fit your organisation logic into Salesforce machinery. Remember that user adoption is the key to a successful transition to Salesforce. Willingness to use the platform will naturally lead the CRM to produce more and more business value for your organisation.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: