DataGroomr and DemandTools both offer Salesforce data deduplication, data quality and data verification solutions. While they boast similar capabilities, their implementations differ significantly.
Table of contents:
- DataGroomr and DemandTools Introduction
- DataGroomr vs DemandTools Pricing
- User Interface
- Matching, Deduplication, and Transformations
In 2020, DataGroomr released its data quality management tool designed to identify and correct errors, inconsistencies, and duplicates in datasets, ensuring data accuracy and integrity. DataGroomr is a SaaS-based app for Salesforce that requires no installation. (An optional Lightning component is available, so you can work directly in Lightning.)
DataGroomr’s unique value proposition is that it combines both ease of use and a powerful set of features that allow users – from the smallest shops to the largest enterprises – to quickly and easily create, maintain and automate deduplication, data management and verification within Salesforce without requiring deep technical understanding or extensive training. It can create this “magic” due to its powerful AI-driven engine, which uses sophisticated machine learning to quickly evaluate records, combined with an intuitive, visually-driven UI. DataGroomr also provides tools for power users who wish to take deep control of merge rules at a field level or create their own algorithmic models, which the visual UI still simplifies.
DemandTools was originally created in 2004 by CRMFusion, which was purchased by Validity Software in 2018. Validity’s DemandTools data quality and data management software offers a wide range of features to help businesses maintain clean and accurate data in their Salesforce org including deduplication, data import/export, data cleansing, data normalization, and data validation.
DemandTools started as a Windows desktop tool (now available on Mac and Linux). There is also a SaaS-based version called DemandTools Elements, but it only allows for record import and export, record deletion, and deduplication of records.
DemandTools has always been focused on power users, providing fine-grained control of deduplication and merge processes. Prior versions were very spreadsheet-like and not easy to use. DemandTools V, released in 2021, overhauled the entire UI to make it easier and more visual. While significantly improved, its user interface has not quite escaped its row-and-column origins.
DemandTools has a traditional rules-driven engine, meaning there is no true “quick start.” You always have to set up the rules for the system to follow. If you are willing to invest the time in learning, DemandTools provides some extremely powerful capabilities to manage and automate your deduplication and data validation efforts.
DataGroomr vs DemandTools Pricing
Pricing is important to examine because there are subtleties when making an apples-to-apples comparison between DataGroomr and DemandTools pricing.
DataGroomr has four versions: Micro, Standard, Professional and Enterprise. DemandTools has three: a free version, DemandTools Elements and DemandTools (with their complete functionality). Comparing is challenging since there is no direct correlation between vendor packages. DataGroomr Micro sits somewhere between DemandTools’ free version and DemandTools Elements. It has all the functionality of DemandTools Elements (for 10,000 records). DataGroomr Standard has more functionality than DemandTools Elements but is probably the closest for cost comparison purposes. DataGroomr Professional and Enterprise versions are most closely aligned in functionality with DemandTools’ complete platform.
On the surface, DemandTools pricing looks like a bargain at $132/seat per year for the complete version. However, DemandTools pricing is based on the number of seats you have in your Salesforce instance, whereas DataGroomr pricing is for the number of DataGroomr users. Since this is administrative functionality, most organizations need just one license.
Since DataGroomr pricing is “fixed” for Standard and Professional plans, DemandTools pricing quickly becomes more expensive. Again, keep in mind that these products have differing functionality available at each tier.
|Free Trial||Complete product||Free Trial|
|Micro||Includes all features to dedupe, merge, import and verify your core Salesforce data for 10,000 records||Free Version||Import, export, delete|
|Standard||Includes all features to dedupe, merge, import and verify your core Salesforce data||DemandTools Elements||Free + deduplication|
|Professional||Standard + custom/cross object deduplication, advanced merge, mass transform, delete and more||DemandTools (complete)||Elements + all other use cases|
|Enterprise||Professional + undo/restore for merges, role-based permissioning and API integration|
Both DataGroomr and DemandTools are relatively easy to install, but given their different origins, their installations are quite different.
With DataGroomr, you simply click on the “Get It Now” button (or the “Test Now” button) which takes you to the DataGroomr signup page, where you sign up with your Salesforce login, are immediately connected with your Salesforce account, and are in the DataGroomr SaaS interface ready to go, along with instructions for how to install the Lightning component.
The Validity DemandTools download from AppExchange takes you to the DemandTools site, where you can register. For the complete product, you download the installation package and install the application like any desktop application. However, the download page is confusing. Although it directs you to download the application, most of the page is dedicated to a record quality display – when you don’t yet have any records to assess – and to promoting additional products.
Lastly, because DemandTools is rule-driven, you need to set up your validation rules as a first step.
DataGroomr and DemandTools take two different approaches to the user experience.
DataGroomr’s home page is set up in three main sections: the left navigation, default dashboard, and a section of widgets. Everything on the dashboard is based on modular widgets; so you can customize the dashboard by moving elements around, resizing, or deleting as you see fit. An “Add Widget” button adds other predefined charts or adds back widgets you removed.
The dashboard provides you an overview of your data – with predefined charts, which can each be customized for datasets, time period, and groupings.
The left navigation shows modules with sub-elements for connecting to core functions like the log browser, jobs, or transform rules. Icons show the number of records in each category, and color-coding shows whether there are records that are awaiting action.
The site section widgets explain what they do and include clearly visible direct links to specific documentation and tutorials.
The DemandTools interface is grid-based with actions clearly shown at the top of the page, making them immediately accessible. There is also a quick visual summary of your data quality and a button to access a more detailed analysis.
Below this are custom scenarios, either pre-built or built by the user, from which to choose showing what action they take, what Salesforce object they apply to, when last run and if automated, and what the run schedule is. There are filters allowing you to reduce what can become an overwhelming list.
There is a major difference in the user interface for creating rules for each scenario. The user experience for designing rules in DataGroomr is visually-driven and drag-and-drop in a virtual palette on a single page. DemandTools is a two-page process following more traditional field selection and configuration. DataGroomr’s approach makes sense in the context of their AI-driven deduplication process. DemandTool’s interface makes sense in their rules-based approach. It’s a matter of style.
The ability to Import record batches manually and deduplicate those records before they create deduplication problems is one of the key features of any deduplication capability. With records constantly arriving from trade shows, hosted events, purchased lists, friends-and-family lists, etc., it’s critical to check those records against the existing database and make quality decisions in “real-time” as the data is uploaded.
DataGroomr’s import feature is about as simple as it can get. Select a source dataset as a .csv file (or any object in Salesforce) to import, and the interface tries to automatically match your input column headers against fields in Salesforce. If it cannot, you can match the fields manually, or choose not to match a field. The system then does the comparison. When completed, you can import the unmatched records (non-duplicates) and even update using the matched (duplicates). Other options include field-level transformation and verification of the import.
DemandTools’ import function again trades off power and control at the cost of simplicity. Select the .csv file you want to import, choose your Salesforce object, and the field mapping screen appears. There is an automap capability, but it is more restrictive. This prevents some import errors but then things start to get complicated. First, you must decide whether you are going to perform an insert, an update or an upsert (confessing that I had to look up “upsert.”) Then for update or upsert, you need to select whether to always update, update if empty, do not update, or combine fields. The combine field option is another tricky aspect which requires the user to reference the user manual to determine how exactly this works.
DemandTools lets you filter records based on any field or combination of fields – to only import records that match the filter criteria. You are also able to add transformations at a field level and or add multiple mappings to a single field. Additional options are available to filter, match and clean records, but many administrators simply want to make sure the data is deduped and up-to-date upon import.
Matching, Deduplication, and Transformations
Now we get into the core functionality: identifying duplicates (matching), deduplication (merging), and standardizing formats for fields (transformations). Both products are so robust that we cannot cover them in detail, but we will give you enough information to distinguish their approaches. In many ways, the differences again come down to an AI-based algorithmic approach versus a rules-based approach and a visual UI versus a more traditional grid-and text-based user experience.
Both products can operate on individual records and an entire dataset through mass actions, but DataGroomr seems to have a more balanced approach. You have a great deal of power to do mass actions, but it also has a very clean interface in the application to do manual editing. DemandTools’ manual merging interface seems like an afterthought and the focus seems to be on providing powerful mass action capabilities.
Another key difference has to do with the relationship between the data and the rule sets applied to them. In DemandTools, a subset of records from the Salesforce database and the rules that apply to them are tightly bound in a scenario, which means you end up storing a huge number of scenarios. For DataGroomr, the dataset is separate from the actions taken on it, which provides greater user flexibility.
DataGroomr runs a matching algorithm to detect duplicates on your Salesforce database when you first open the product; and displays the results. This algorithm is based on traditional AI and machine learning, using a sophisticated multivariate model that runs off all fields in a Salesforce object to determine likelihood of match, displayed in the match confidence field in the result set.
The default rule seems to work very well in the tests we performed, but if you prefer to use a model you design, DataGroomr allows you to create and train your own models. Of course, you have to have a large enough dataset to train the model. DataGroomr’s model has been trained on millions of records across a plethora of datasets. The model you build will be trained only on the data in your dataset.
DataGroomr allows you to build two kinds of models: a machine learning-based model or a more traditional rules-based matching model. In general, a well-designed machine learning-based model is going to perform better than a rules-based model. But since some companies require a rules-based model, DataGroomr provides that option.
One of the neat features of DataGroomr is the Undo button. If you decide that you made a mistake, whether for an individual record merge or a mass merge, you can hit undo. That backs out the action and reconnects the discarded records to all their previously connected information. This is an incredible “safety net” that allows you to feel safe that if you do something wrong, especially as you are learning, you are not locked into the mistake.
You can merge records manually in DataGroomr, which is fine for small datasets, but large datasets need to be handled using a set of mass activities and rules.
DataGroomr lets you establish both record and field-level merge rules. This is where the visual merge rule designer comes into play, which makes rule building simpler and allows you to visualize the impact more clearly.
First you select the fields that will be part of the rule; then drag and drop scenarios, logic and functions into the appropriate logical structure; then set any values for the fields for the rules to run on. Hit “Run” in the rules preview and you see the results as if the rule was being applied. If the preview meets your expectations, you then save it.
The problem here is that this is not a complete merge rule. There is a second location where you need to specify rules for a merge – the Dedupe use case screens. After you have selected the subset of records, set the match criteria, run the scenario, and get a set of potentially matching records back, there are two buttons – Merge Options and Winner Rules at the top of the screen.
Merge options include:
- The ability to keep a non-winning record by adding a prefix
- Creating a follow-on task after the merge is complete
- Changing record ownership based on a set of assignment rules as records are being changed.
Winner Rules lets you determine the scope of the rule and apply the specific winner rules to this scenario. Where you would think you would apply just one rule, you can actually apply multiple – and this is where it gets complicated. Do I create a complex winner rule in the winner rules editor and apply, or should my strategy be to create lots of small, granular rules that I can combine for all scenarios as appropriate? This is an unnecessary complexity I don’t really need, and it doesn’t increase the control and power I have with merge rules. In my mind, this split and confusing implementation is one of the great weaknesses of DemandTools.
Transformation rules in DataGroomr are created with the visual rules editor. The fields for which formatting can be applied and the options in logic and values cover most situations for field-level standardization.
There are also options to create ad-hoc rules like the one shown below that transforms field-level data.
DemandTools once again has a very complex approach, located across two screens. The first location is in the Cleanse Rules screen, accessed from the settings menu, where you have very detailed control over critical field values and can standardize names and cases.
You can also transform data in the Modify user case screens. Here you can replace fields with values from other fields, specify field values, replace values from an extensive list of formulaic definitions, use aggregations, among other capabilities. It is in no way easy to use.
There are a lot more features we could compare, but hopefully this brief overview provides a good sense of how these solutions differ. While they both cover a range of data cleaning use cases, they have clearly different design philosophies and target customers. Which are you?