Marketing in Atlanta is a high-stakes game. Predicting where your budget will generate the most return requires more than gut feeling. Can and predictive analytics for growth forecasting truly transform your marketing strategy and help you anticipate future trends with pinpoint accuracy? Absolutely. This guide will show you how to leverage the power of Salesforce Marketing Cloud Intelligence (formerly Datorama) to achieve precisely that.
Key Takeaways
- You will learn how to connect your Google Ads account to Salesforce Marketing Cloud Intelligence to import campaign data.
- You will discover how to create a custom dimension within Salesforce Marketing Cloud Intelligence to categorize your marketing efforts by specific Atlanta neighborhoods.
- You will understand how to build a predictive dashboard in Salesforce Marketing Cloud Intelligence that forecasts website conversions based on historical data and seasonality.
## Step 1: Connecting Your Data Sources to Salesforce Marketing Cloud Intelligence
The foundation of any good forecasting model is, well, the data. Salesforce Marketing Cloud Intelligence thrives on data, and the first step is connecting your various marketing platforms.
### 1.1: Accessing the Connect & Mix Tab
Log into your Salesforce Marketing Cloud Intelligence account. On the main dashboard, locate the “Connect & Mix” tab on the left-hand navigation menu. Click it. This is where the data magic begins.
Pro Tip: Make sure you have the necessary permissions to connect data sources. Usually, you’ll need admin access to both Salesforce Marketing Cloud Intelligence and the external platform you’re connecting.
### 1.2: Connecting Google Ads
In the “Connect & Mix” tab, you’ll see a variety of data source options. Locate and click on the Google Ads connector. You’ll be prompted to authorize the connection with your Google account. Follow the on-screen instructions to grant Salesforce Marketing Cloud Intelligence access to your Google Ads data.
Common Mistake: Forgetting to select all the necessary Google Ads accounts. Salesforce Marketing Cloud Intelligence only imports data from the accounts you explicitly authorize. I made this mistake last year for a client; we were missing data from their North Fulton campaigns for a week before we caught it.
### 1.3: Configuring Data Streams
Once connected, you’ll need to configure your data streams. In the Google Ads connector settings, specify which campaigns, metrics, and dimensions you want to import. Be sure to include key metrics like Impressions, Clicks, Cost, Conversions, and Conversion Value. For dimensions, include Campaign Name, Ad Group Name, Keywords, and Geographic Location.
Expected Outcome: After completing these steps, Salesforce Marketing Cloud Intelligence will begin importing historical data from your Google Ads account. You’ll see the data populate within the platform within a few hours (depending on the volume of data).
## Step 2: Creating Custom Dimensions for Localized Analysis
To effectively forecast growth in the Atlanta market, you need to segment your data by specific geographical areas. This is where custom dimensions come in handy.
### 2.1: Navigating to the Dimensions Management Section
From the main menu, go to “Analyze & Act” and then select “Dimensions Management.” This section allows you to create and manage custom dimensions that are specific to your business needs.
### 2.2: Creating a “Neighborhood” Dimension
Click the “Add Dimension” button. In the “Dimension Name” field, enter “Neighborhood.” For “Data Type,” select “Text.” Now, you need to define the values for this dimension. This is where you’ll list out the specific Atlanta neighborhoods you want to track, such as Buckhead, Midtown, Downtown, Virginia-Highland, and Decatur.
Pro Tip: Be consistent with your naming conventions. For example, always use “Virginia-Highland” instead of “VaHi” to avoid data discrepancies.
### 2.3: Mapping Google Ads Data to the “Neighborhood” Dimension
This is the tricky part. You need to tell Salesforce Marketing Cloud Intelligence how to associate your Google Ads data with the “Neighborhood” dimension. The easiest way to do this is by using the “Rules Engine.” Go to the “Rules Engine” section and create a new rule.
For the rule, specify that if the “Geographic Location” dimension in Google Ads contains “Atlanta,” then assign the corresponding neighborhood based on the ad targeting. For example, if the ad targets the 30305 zip code (Buckhead), assign the “Neighborhood” dimension to “Buckhead.” You’ll need to create multiple rules to cover all the neighborhoods you’re targeting. Consider how hyperlocal marketing in Atlanta can benefit from this approach.
Common Mistake: Overlapping rules. Make sure your rules are mutually exclusive to avoid assigning multiple neighborhoods to the same data point. It’s a headache to untangle!
Expected Outcome: After setting up the rules, Salesforce Marketing Cloud Intelligence will automatically assign a “Neighborhood” value to each of your Google Ads data points. You can then use this dimension to segment your reports and dashboards.
## Step 3: Building a Predictive Dashboard for Growth Forecasting
Now for the exciting part: building a dashboard that forecasts future growth based on your historical data.
### 3.1: Accessing the Dashboard Builder
From the main menu, go to “Visualize & Explore” and then select “Dashboards.” Click the “Create Dashboard” button to start building a new dashboard. Give your dashboard a descriptive name, such as “Atlanta Growth Forecast.”
### 3.2: Adding a Time Series Widget
The core of your predictive dashboard will be a Time Series widget. Select this widget type and configure it to display Website Conversions over time. Set the time range to at least the past two years to capture seasonality.
Pro Tip: Experiment with different time granularities. You might find that weekly or monthly data provides a more accurate forecast than daily data.
### 3.3: Applying Predictive Analytics
This is where Salesforce Marketing Cloud Intelligence shines. Within the Time Series widget settings, enable the “Predictive Analytics” feature. You’ll see options for different forecasting models, such as ARIMA, Exponential Smoothing, and Prophet. I recommend starting with ARIMA, as it’s generally well-suited for marketing data. For more on this, see our article on predictive analytics to grow revenue.
### 3.4: Adding Filters and Segmentation
To refine your forecast, add filters to segment the data by your custom “Neighborhood” dimension. This will allow you to see predicted growth for each specific area of Atlanta. You can also add filters for other dimensions, such as Device Type and Ad Campaign.
Expected Outcome: After configuring the widget, Salesforce Marketing Cloud Intelligence will display a forecast of future website conversions based on your historical data and the selected forecasting model. You’ll see a trend line extending into the future, along with confidence intervals indicating the range of possible outcomes.
## Step 4: Refining Your Forecast and Taking Action
Predictive analytics isn’t a crystal ball. It’s a tool that requires constant refinement and adjustments.
### 4.1: Monitoring Forecast Accuracy
Regularly monitor the accuracy of your forecast by comparing it to actual results. If you notice discrepancies, adjust the model settings or add new data sources to improve accuracy. You can use the “Forecast Accuracy” widget in Salesforce Marketing Cloud Intelligence to track the performance of your predictive models.
### 4.2: Incorporating External Factors
Remember that your forecast is only as good as the data you feed it. Consider incorporating external factors that could impact your marketing performance, such as economic indicators, competitor activity, and seasonal events. You can manually adjust your forecast based on these factors or integrate external data sources into Salesforce Marketing Cloud Intelligence.
A report by Nielsen [https://www.nielsen.com/insights/](https://www.nielsen.com/insights/) showed that consumer spending in Atlanta is highly correlated with the unemployment rate.
### 4.3: Taking Action Based on Your Forecast
The ultimate goal of growth forecasting is to inform your marketing decisions. Use your forecast to allocate your budget effectively, adjust your bidding strategies, and optimize your ad creative. For example, if your forecast predicts strong growth in Buckhead during the holiday season, you might want to increase your ad spend in that area during that time. Thinking about budget allocation? See our post on smarter customer acquisition.
I once had a client who used Salesforce Marketing Cloud Intelligence to predict a significant drop in demand for their services in Midtown during the summer months. Based on this forecast, they decided to shift their marketing budget to other areas of Atlanta and focus on building brand awareness during the slow season.
Predictive analytics with tools like Salesforce Marketing Cloud Intelligence empowers marketers to make data-driven decisions and achieve sustainable growth. Don’t be afraid to experiment, iterate, and continuously refine your forecasting models. The results will be well worth the effort.
## FAQ Section
What if I don’t have enough historical data to build a reliable forecast?
If you’re just starting out, focus on collecting as much data as possible. In the meantime, you can use industry benchmarks and competitor analysis to inform your marketing decisions. As you accumulate more data, you can gradually transition to data-driven forecasting.
How often should I update my predictive models?
It’s best practice to update your predictive models at least monthly, or even weekly, to incorporate new data and account for changing market conditions. The more frequently you update your models, the more accurate your forecasts will be.
What are the limitations of predictive analytics?
Predictive analytics is not a perfect science. It’s based on historical data, which may not always be indicative of future events. Unexpected events, such as economic downturns or competitor disruptions, can throw off your forecasts. It’s important to use predictive analytics as one tool in your marketing arsenal, alongside your own intuition and experience.
Can I use Salesforce Marketing Cloud Intelligence for other types of marketing analytics?
Yes, Salesforce Marketing Cloud Intelligence is a versatile platform that can be used for a wide range of marketing analytics tasks, including campaign performance analysis, customer segmentation, and attribution modeling. It can integrate with various data sources beyond Google Ads, like LinkedIn and Meta.
Is Salesforce Marketing Cloud Intelligence the only tool for predictive analytics in marketing?
No, there are other tools available, such as SAS and IBM SPSS Statistics. However, Salesforce Marketing Cloud Intelligence is specifically designed for marketing data and offers a user-friendly interface and integrations with other Salesforce products. A recent IAB report [https://iab.com/insights/](https://iab.com/insights/) highlighted the increasing adoption of marketing-specific analytics platforms.
The power of and predictive analytics for growth forecasting lies not just in the insights it provides, but in the actions it inspires. Stop reacting to market changes and start anticipating them. By implementing these steps with Salesforce Marketing Cloud Intelligence, you can transform your Atlanta marketing strategy from reactive to proactive and gain a significant competitive edge. So, are you ready to take control of your marketing future?
