I was recently presenting a lecture speaking about how a large number of social networks cause companies to lose their control over communication channels with their customers. I was also explaining why it is a problem and how Azure serverless services can reduce the problem dramatically for less than 50$. The reaction from the audience was great. So I decided to share it with you. Here is the explanation:

Before social networks existed on the internet (Facebook, Twitter, Instagram and so on), the traditional communication channels between companies and customer were telephone, fax, email, mail or even a company website. Basically, companies had several common channels of communication. So, they could decide which of the communication channels to use.

Nowadays, social networks have become a legitimate communication channel between the customer and the company. As a result, the customer expects the company to respond within the social network. If the company does not, it becomes irrelevant. This means that companies no longer have the privilege to decide which of the communication channels to use.

Social networks are not the problem. Their quantity is. There are too many social networks and they continue to grow. If companies want to communicate with customers through all social networks or some of them, companies will have to pay an enormous amount of money. If they don’t use social networks, they will lose customers. Companies have no more the privilege to decide which of the communication channels to use.

Companies that understand how important this issue is. Spend time and money to develop their private solutions from scratch. Other companies are using external services such as Hootsuite or SproutSocial. These services are excellent, and companies should consider using these kinds of external services. Another option I would like to suggest is to consider using Azure serverless services to create your tailor-made solution. Let’s look at a diagram (1), it demonstrates an automated communication channel for customers that are using Twitter to interact with the company.

An explanation of a diagram:  

The flow trigger is a tweet with the hashtag #expertsliveisrael (Experts Live Israel was a  conference that I organized and took part in during July 2018 in Tel Aviv). Next, it is splitting into using Azure Cognitive Services (Text Translate and Text Analytics) to analyze the tweet sentiment (is it a positive or negative tweet) in order to create a tailor-made response. It also stores the data inside an Azure Cosmos DB for future analyzing of the data.

Note: I used a database to store the data for demonstration purposes, it is not necessary for automatic response. I used it to give you an idea of possible insights that you might use, such as analyzing complaints and improving tailor-made responses.

You may follow my steps to achieve a similar result (I won’t describe all the workflow steps, only those that I think will be important for you to understand the solution):

1. Create a Twitter account or use an existing account, if you don’t have a Twitter account click here to signup.

2. Open Azure portal, and create a new Logic App flow ( I’m assuming you already know Logic App, if not I suggest to read this simple guide or start playing with it.)

My main Logic App flow is ExpertsLiveIsreal-Demo-Twitter, and here is how it visualized on designer view.

To simplify the visuality and for future maintenance purpose, I create a three separated Logic app flows.

1. Add a Twitter trigger and fill in the fields

2. Create a new Logic App flow for cognitive services

Note: You will need to create two Cognitive Services, a Text Translate and Text Analytics (If you don’t know what it is and how to use it, click on this link.)

1. Create a new Logic App flow for a tailor-made response

2. Currently, to reply a tweet, there is no simple built-in action in Logic apps, therefore, at first, you will need to create a Function app and a developer account in Twitter.

3. Once you created a developer account in a Twitter, you may continue to build a Function. 

How I use a free LinqToTwitter.net API?

To be able to do so you need to download it and save it. Next, upload a LinqToTwitter.net.dll file into the bin folder inside the Function app folder. To open the function folder, click on the function name in my case it is an ExpertsLiveIL,  then change the tap to a Platform features and click on an App Service Editor.

You will open a new window of an App Service Editor.

Select a bin folder and Right-click on a mouse

Select Upload Files, browse to the location where you saved a LinqToTwitter.net.dll, select it and click on Upload button.

1. Create a Cosmos DB follow the 5 minutes guide to creating it.

2. Now, you are ready, to complete other actions on main flow.

Note: I didn’t mention a small function GenerateGuid, that I needed since Cosmos DB has a mandatory property “id” and it must be unique. So, I created a function that outputs a unique value to pass as “id” value.

It took me a day maybe a little more to complete a whole flow. And here is a list of my costs:

As you can see the total price was 45.87$. It’s a low price, wouldn’t you agree?

You should try it, be creative and productive 🙂

You are welcome to write me anything (a question or a comment or feedback).

And don’t forget to subscribe! Thanks.