Tag Archives: decision trees

Marketing Case Study: Global Players

Here’s an example of how Zingtree’s interactive decision trees are helping to improve online marketing efforts in a real education abroad organization.

Global Players helps to place student-athletes in their perfect study-abroad fit, promoting awareness and cultural education as well as athletic endeavors around the world. As an organization growing in popularity, Global Players was in search of finding a streamlined, self-service way to placing students in their ideal location abroad.

Since adopting Zingtree’s solution to integrate their “Program Match” decision tree, Global Players has been able to place over 1,000 students so far, increased the likelihood of a visitor to contact the organization by 5 times, and more!

You can view the case study below, and open a PDF version here.

casestudy-global_players copy

Want to see more? Explore the Zingtree Gallery for examples of how our decision trees and interactive solutions provide value for businesses!

This case study was originally published January 19, 2016.

Create Decision Trees using Microsoft Excel

decision tree excel

If you’re comfortable using Microsoft Excel, you can build the first draft of your Zingtree decision trees in Excel using a spreadsheet, and then easily import them into Zingtree. Once you’ve successfully imported your tree, you can modify it and enhance it using Zingtree’s editing tools, which offer a lot more decision tree related functionality than Excel.

Here’s how our example tree used in the tutorials appears in Excel:

Overview

Zingtree can create decision trees from Excel, or any similarly formatted tabular source. Your spreadsheets just need to be set up in a specific way for this to work.

Start now by downloading and modifying this example .XLS file:

Download Examples for Excel

Note: This spreadsheet has two tabs: Basic and Advanced.

Here are the rules:

  1. The first row is for column headings. This is important, as it tells Zingtree what type of data is in each column. Make sure to use the column headings as described below.
  2. Column A is for the node number. Usually, this is sequential. It’s required. The heading must say “Node“.
  3. Column B is for the title of a node. Your trees will be easier to read if each node has a descriptive title. The heading must say “Title“.
  4. Column C is the question that is being asked. You can leave this blank if you want an answer node. The heading must be “Question“.
  5. Column D is for any content that appears in the content area.  This is imported as plain text, but you can add formatting, images, and videos later using the Zingtree editing tools. The heading must read “Content“.
  6. If you want to include node tags in your tree, insert a column with a heading of “Tags“. This is optional.
  7. If you are using a scoring variable for scoring button clicks, add a column headed “Score Variable“. This is optional.
  8. The last columns are for the button choices.  The heading over the first button column must be “Buttons“. You can have several columns of buttons.
  9. For the button columns, you can make them link to other nodes by adding the node number in square brackets.  In the above example, cell F2 has a button labeled “Yes” that links to node #2.  ( Yes[2]  ).
  10. If you’re using score variables for button clicks, you can include something like “+3” after the node number in brackets.  This would add 3 to the scoring variable mentioned in the Score variable column. (See the Advanced tab in the sample spreadsheet for an example.)
  11. If you want to add a comment to any node, insert a cell on the right that starts with an exclamation point character (!).

Note: Columns can be in any order, but the column headings must contain the proper text like “Node”, “Question” etc.

You can also make Link Nodes and Tree Nodes with special text in the content column:

  • Example: To make a Link Node that goes to Google, the content area looks like this (see cell D9 in the example):
    LINK: http://google.com
  • Example: To make Tree Node that opens tree ID #123456789, the content area is this (see cell D8 in the example):
    TREE: 123456789
  • Example: To make a Tree Node that opens tree ID #999999999 at node #3, the content area should be:
    TREE: 999999999,3

Once you’ve finished your tree, it needs to be exported as a TXT file. This is also known as a tab delimited CSV.

You can also just copy and paste cells from your Excel document into Zingtree. Excel copies tab delimited CSV data to the clipboard automatically.

Build Your Tree

To start, download our “what to wear” example tree, as an Excel.XLS file.

Open this file in Excel, and start modifying it.  You can use the Basic or Advanced tab – most people start with the basic option. Make sure to keep top row column headings in place. Keep questions in the question column, content in the content column, etc.

When you’re done, you can import it into Zingteee via copy and paste, or by exporting to a TXT file.

Import via Copy and Paste

This is the easiest way to bring your decision tree into Zingtree:

  1. In Excel, select the entire range of cells for your tree, and copy to the clipboard (Ctrl+C or Cmd+C).
  2. In Zingtree, go to the Import via Copy and Paste tool.  (You can also get there via My Trees, Create Tree, then select Import from Microsoft Excel.) A screen like this appears:

  3. Choose Microsoft Excel as the source.
  4. Enter a name for your tree
  5. Paste the data copied from step 1 into the data area. (Use Ctrl+V or Cmd+V).
  6. Click Import and Create Tree.

You’ll see your new tree in the Zingtree overview.

Import via a TXT file

For larger trees, you may find it better to upload a file instead of copying and pasting. Here’s how it’s done:

  1. In Excel, go to File, Save As, and select Text (Tab Delimited) as the type.

  2. The file name will become the name of your tree. Click Save when finished.
  3. In Zingtree, go to the Import File tool. (You can also get there from My Trees, Create Tree, and then choosing Import from Excel.)

  4. If you want to overwrite an existing tree, select it via Replace Tree. Otherwise, a new tree will be created.
  5. Click Import File, and locate the file you created in steps 1 and 2.
  6. The new tree will appear in Zingtree.

Notes:

  • You can use this process to import files from any tab delimited CSV format.

Any questions? Reach out to us anytime. 

This article was originally published January 26, 2017.

Make Tests, Quizzes, Assessments & More with Zingtree’s Logic Nodes

adding-search-to-decision-trees-6

Zingtree has the ability to create what we call evaluations through the use of logic nodes. These can be interactive tests, quizzes, assessments, training simulations, or anything else that requires assigning a score to each question answered – then later acting upon the total score.

Coupled with the powerful decision tree capabilities already present in Zingtree, logic nodes opens up a whole new realm of knowledge engineering possibilities. To see a working example of an interactive decision tree quiz using logic, try this 10-question Personality Test found in the Zingtree Gallery.

Here are the basics on how to build a tree with logic nodes:

  • Add a value to each button click when designing your tree.
  • As the end-user goes through your tree, Zingtree keeps a running total of the score values of all buttons clicked.
  • When you’re ready to act upon the running total score, use a new “Logic Node”. From there, you can specify rules which redirect to specific nodes.

Want to build your own? Here’s a step-by-step guide:

Step 1: First, go to Overview > Edit Node and open up the button editor by clicking Edit Buttons. Select Score Button Clicks, and enter a Variable Name that will be used to tally scores and branch on later.

581d3092c6fe48-46396098-359664

Step 2: Next, add your desired score values to button clicks in your lead qualification decision tree.

button-scoring

Step 3: Now, add a new Logic Node using This Tree > Add Node. In the Logic Node, you can specify which node should appear next (and which to branch to) according to rules you create.

Step 4: Link the last button click of your test/quiz/assessment to this new Logic Node.

Note that the Logic Node never appears on the screen. When you direct the flow of your Zingtree to this node, it looks at the sum total of all the button click scores and redirects to the proper node.

Need to See More Examples? The Zingtree Gallery has three trees that use Scoring. You can play or examine them to see how they work. See the Logic Demos.

Bonus Tip! 
Want to display the score on a node? Just add the text #score# into the content area.

Other Notes:

  • You should make sure you are using the latest embed code or hosting URL for your tree.
  • If your end-user presses the back button, the running total score will adjust.
  • If the tree is restarted, the running total score resets to zero.

Any questions? Please reach out to us at any time

This article was originally published August, 2015.

Address Auto-Complete and Verification in Decision Trees

Many of our call center customers are using Zingtree to collect data from their customers, and in a sales or shipping application, there’s always an address form or two to fill out. Our customer Tom K. asked us if there was a way to make this process faster, and also to verify addresses to eliminate costly shipping errors. The short answer? Yes! Zingtree has an automatic address auto-complete and verification capability available to you.

Note: There is an extra cost for each address verification lookup: two lookup credits, or $0.02 per lookup. And currently, this is for addresses in the USA only.

Here’s How It Works

As you start entering an address, Zingtree gives you selections on every possible match:

The more text you enter, the fewer matches appear:

Once you select a match, the city, state and zip code are filled in automatically, and the address is verified against the USPS database. If the address is valid, a “USPS Verified” tag will appear in the form:

Try a demo here

How to Set Up Address Auto-Complete and Verification

Setting up a form with address verification requires two parts:

  1. Enable the Verify USA Addresses option in the Tree’s Settings.
  2. Set up your address entry forms, using the new address 1, address 2, city, state and zip field types.

Step by step, here’s how to do it:

  1. Select your tree, and go to the Settings tool.
  2. Make sure Verify USA Addresses is checked. This will enable the special address field types.

  3. Click Update Settings.
  4. Now, go to Overview, and edit a question node where your address form resides.
  5. Locate the Data Entry Fields area, and click Add to add fields.

  6. Add the first line of address entry, using the field type Address 1:

  7. Add the second line of the address, using the Address 2 field type.
  8. Add a field for the city, using the City field type.
  9. Add a field for the state, using the State field type.
  10. Add a field for the zip code, using the Zip field type.
  11. When you’re done, the data entry fields summary will look something like this:

    Important: Make sure the field types are all set properly.

  12. Save your node by clicking Save Changes.

That’s it! If you like, you can copy our Gallery demo tree and modify it for your application. Or, just examine it to see how it’s built.

If you like this feature or have any ideas on how to make Zingtree even more useful, please send us a message.

Decision Tree Maker: Troubleshooting for Tech Support

Troubleshooting has been a key problem-solving technique used by technical support teams and agents for ages. For every type of product or service in existence, there are countless types of issues that can potentially occur; troubleshooting aims to help narrow down what that problem could be for each customer, based on other contextual factors.

Zingtree makes it easy to build custom, interactive troubleshooters for tech support that are simple and straightforward enough for anyone to use and understand. When YouMail, the leading provider of visual voicemail, realized that they needed to resolve their support inquiries more quickly, they looked to automated troubleshooters to help. Using a clever mix of our interactive troubleshooters and YouMail’s own technical support staff, the company was able to amp up their customer support efforts, and assist customers in a wider range of questions and issues – all while saving money.

Here’s how our interactive troubleshooters have allowed YouMail and other companies to streamline their tech support departments.

Improving Self-Solving for Customers

While traditional methods of customer support are still in practice today, the vast majority of customers (72%, in fact) prefer to use self-service support options before reaching out to a live agent to resolve their issue. Turning a complex interaction into a simple click-click-click provides an easy-to-digest platform for people to find solutions all on their own, at any time or place. Offering these types of avenues for customers to find answers, understand technical processes, and troubleshoot problems, will greatly improve how you serve your customers, and, in turn, how they will feel about your company.

Providing a Complete, Guided Path

Information is always easier to follow and retain when you’re being gently guided through bite-sized pieces of information. Interactive decision tree troubleshooters are a more user-friendly way than traditional troubleshooting methods to guide people through a series of in an interview-style way, which helps to keep the process moving along and customers more engaged with the information being presented. You can even add customizations and visuals like GIFs, videos, and photos to help display information in a more fun and engaging way.

Connecting with Existing CRM

Effective technical support strategies should integrate every method and means of customer communications in a centralized way. For most companies, this means having a built up knowledge base, FAQ, and live support that can share information with one another seamlessly.

Zingtree troubleshooters Integrate directly with Salesforce, Freshdesk, and any other CRM platform, making it easy to pass along proper information from the troubleshooter straight to an agent. If the troubleshooter doesn’t solve the problem, the entire Q&A session history is sent to you, or imported into your CRM, saving tech support from a lot of back-and-forths.

Here’s an example of how a Zingtree session transcript appears inside of Zendesk:

Optimizing with Reports & Analytics

With built-in, real-time analytics and powerful usage reports, businesses can view the direct impact of their interactive troubleshooters for tech support.

  • Performance/Results Report: see how your help system is performing.
  • Session List Report: see all the different customer sessions during a date range, and easily click for more details such as what parts of your troubleshooter they visited, their node-to-node path, and for how long they stayed on each node/question.
  • Traffic Map Report: quickly access a complete view of the content in your decision tree nodes, as well as a breakdown of the traffic to each, shown in both number of clicks and percentage of total traffic.

The Basics: How to Build a Troubleshooter

To get your troubleshooter up and running as quickly as possible, we recommend using the Zingtree Wizard tool.

Set up in a guided, form-based way, the Wizard allows you to supply every question, answer, and other pieces of content for each stage of the troubleshooting process. As you dive further into the creation of your troubleshooter, you can fill in more of the nitty-gritty details, and see a live preview of what the page will look like:


Companies that integrate Zingtree troubleshooters into their customer support processes have seen a 20% reduction in technical support costs on average, as well as improvements in the overall volume of incoming tickets. 

Get started with an interactive troubleshooter, check out our Gallery for more inspiration, or reach out to us with any questions!

Sending Slack Notifications from Decision Trees

slack+zingtree

Slack has become the go-to collaboration platform of choice for so many companies, and people are also using it to track events. We have also recently adopted Slack and wanted a way to send messages to Slack channels whenever something occurred in one of our own Zingtrees.

So, as part of our mission to make Zingtree interoperable with as many popular platforms as possible, we’ve added this capability to our most recent release.

How it Works

When an end-user visits a node in one of your decision trees, you can have that node trigger a custom Slack notification. Once you set up Slack for your organization, any node in any tree can send a message to Slack.

Notifications in Slack look like this:

Slack notification from Zingtree

Setup

First, you need to give your organization access to the Slack app:

  1. Go to Account > My Apps.
  2. Click Slack under Add Apps.Add Slack to Zingtree
  3. Slack will ask you to authorize Zingtree access. Select the Slack channel you want to receive notifications (“post to”), then click Authorize.Authorizing Slack
  4. You will be returned to Zingtree, and Slack will appear in your list of apps.Slack installed in Zingtree

Now you can configure any node in any tree to send a custom Slack notification:

  1. Pick a tree, and go to the Overview tool.
  2. Pick a node, and click the Edit button.
  3. Locate the Send Message To option, and select Slack.Slack messaging added to a Zingtree Node.
  4. Enter a message that will be sent to your Slack channel.  You can also include any form variables (i.e. #name#).

Now, Preview your tree, and visit the node where you just added a Slack message. The custom message will appear in Slack, along with a link to the session transcript:

Slack notification

 


 

Any questions about this, or any of our other integrations? Please contact us anytime.

This article was originally published on April 20, 2016.

Decision Trees for Sales Retail: Phoenix Bats Case Study

Phoenix Bats + Zingtree

Helping customers find and choose the right products in an interactive way

The Phoenix Bat Company has been crafting high-end baseball bats for over 20 years. Using only premium grade woods with the most advanced bat-making machine in the world, they deliver the most consistent and high-quality bats to their customers, including pro baseball players. With painstaking attention to detail and custom manufacturing, Phoenix Bats produces some of the best bats in the country.

Phoenix Bat’s Challenge

With customers at all skill levels and various requirements, visitors to the Phoenix Bats website were confused by the difficulty of researching and selecting the right bat for their needs. In search of a better solution, Phoenix Bats wanted to find an easy way for potential customers to find and select a wooden bat tailored to both the customers’ game and hitting style.

Zingtree’s Solution

Building a decision tree bat selector tool has enabled Phoenix Bats to guide customers to the baseball bat perfectly suited to their needs. With this interactive platform in place, customers are now able to quickly find a bat, leading to an increased conversion rate and improved sales. Phoenix Bats is also enhancing the end-user experience by using the Zingtree API.

“Without Zingtree, [our customers] would have to read through a number of different bat descriptions in order to figure out which bat is the best fit,” says Seth Cramer, General Manager and Co-Owner of Phoenix Bats. “Now, they get the answer in seconds.”

Key Statistics: 

  • 13 Trees: 13 decision trees in use
  • Lots of Bats Sold: Over 10,000 customers helped

About Zingtree

Zingtree is the most user-friendly platform for creating and implementing interactive decision trees that deliver answers faster. Quickly create a decision tree that your site visitors, leads, trainees and/or customers navigate by clicking buttons to answer questions, and receive detailed analytics on how trees are being used. Zingtree makes it easy to guide anyone through complicated processes – there’s no better way to help people get answers and solve problems.


Want to see more? Explore the Zingtree Gallery for examples of how our decision trees and interactive solutions provide value for businesses!

Automatically Including Location Data in your Decision Trees

Zingtree can tell a lot about your end-users just by gleaning information from their browser. For example, you can determine their preferred language, and where they are located.  Zingtree gives you an option to collect this information, and use it in your decision trees.

For example, you can pre-fill a Zip code field in a Data Entry field, or use Logic Nodes to branch to a node depending on the end user’s language or country.

Using the Collect Location Data option in the Settings tool, you can gain access to the following information:

  • Language
  • City
  • State/Region
  • Country
  • Zip or Postcode
  • ISP
  • Origin/domain of server
  • User’s IP address
  • User’s browser type

Getting Set Up

Adding this data to your Zingtree session is easy. Here’s how to do it:

  1. Select your tree, and go to the Settings tool.
  2. Select Collect Location Data.


    If you want to do this for ALL your trees, also click on all my trees. (You can disable enhanced data collection on all your trees in the same way.)

  3. Click Update Settings.

From now on, any sessions with your tree will also include these variables:

  • zt_language (The 2 letter code of the user’s preferred language. “en” is English.)
  • zt_country
  • zt_regionName (this is a USA state)
  • zt_city
  • zt_zip (also Postal Code)
  • zt_isp (the user’s Internet Service provider)
  • zt_org (the origin domain of the user)
  • zt_ip (the user’s IP address)
  • zt_browser (the user’s browser type: IE, Firefox, Chrome, Safari, etc.)

When you examine a session (like from the Sessions List report), you’ll see these variables as a part of your session, like this:

Got any cool uses for this? Let us know!

New Data Entry Form Updates and More

Besides turning Scoring Nodes into more powerful Logic Nodes, we’ve done a lot of little improvements and fixes in the last 30 days.  In particular, we’ve had a few requests for enhancing Data Entry capabilities.

Here’s what’s new:

  • New: Logic nodes
  • API: delete_form_data returns better error message if deleting already deleted form data (Jay H.)
  • Fix: Link nodes now wait for session data to save before linking out. Makes API calls from linked node work 100% reliably.
  • Fix: SSO URLs are now REST compliant (no & or ? characters) (Priya)
  • New: Added browser_language built-in variable (Eric J.)
  • Update: Logic node editor can set root node.
  • Update: Tree nodes can be set as root node (Chris P.)
  • Update: Only allows letters, numbers and underscores in tags. (Morgan A.)
  • Update: Edit Content Node shows advanced options by default if tags or any other advanced option is set (Bill Z.)
  • Update: Simple Overview shows node tags on each line (if they exist). (Bill Z.)
  • Update: Link nodes now show a “please wait” message, in case new page is slow to open. Also improves email-session-info.php experience.
  • Fix: No longer able to inadvertently create node #0 from Document Node editor. (James W.)
  • Update: Tree search (and search in Zingtree FAQ) now shows node title and tag matches first. This gives more relevant results for tree searches.
  • Fix: host-trees integration URL with a space in the agent name source= parameter now gives correct results (Juan C.).
  • Update: Create Tree for Excel, Sheets options now includes direct links to Excel, Sheets example files.
  • Fix: Blank lines in bulk agent submit no longer show an errant error message. (Allen J.)
  • Fix: removed extra white space at the bottom of embedded trees (Josh S.). (Requires getting new embed code.)
  • Fix: Now saves form data when jumping to a new tree and not continuing past the first node (Diana T.).
  • Fix: Session List CSV export now uses local time.
  • New: Session List (All Trees) Report (for Koltyn)
  • Fix: Agent feedback button didn’t show or work properly in embedded trees (Morgan).
  • Update: Button labels can now be 255 characters.
  • Update: Increased database performance by moving to SSD storage.
  • Fix: Now allows tags to be entered without having to press ENTER.
  • Update: Default tag choices that appear in tag list drop down in Settings, My Agents are from all agents and trees, and only from current organization.
  • New: Email and password field types now included in Data Entry Forms (Imran, Maximiliano)
  • Update: Now supports “required” option for radio buttons in data entry forms (Imran, Diana T.)
  • Fix: Tooltips now always appear
  •  Fix: Preview for authors with apostrophes in their name now works properly (Joe O’)
  • Fix: Nopermalink now gets passed as a parameter for host-trees.php (Jenn V.)
  • Fix: Uses https:// for hosting when going from https://zingtree.com/host-trees.php (Jenn V.)
  • Update: SSO restriction is set by default for new trees if SSO is set up for organization (Jenn V.)
  • Update: Better CSS for panels style tree rendering.
  • Update: Spaces in search parameters sent to host-trees.php turn into underscores so tag matching can work properly (Jenn V.)
  • Fix: Clicks and Usage report wasn’t showing true engagement due to using last click time.
  • Fix: Empty scoring variables no longer cause JS errors when rendering trees (Bob H.)
  • Fix: Emailed session transcripts from multi-tree sessions now get the proper questions before the tree jump (Travis A.)
  • Fix: Webhooks with no returned data no longer make JS errors while rendering tree (Bob H.)
  • Fix: Editing data entry fields – can now properly set Required attribute for multiline text fields (Bob H.)
  • Fix: Launching trees from host-trees.php now properly passes along agent name (Juan C.)
  • New: /api/get-history-data.php for use with updated Zendesk Support app
  • Update: Agent feedback now includes link to session transcript (Katheryn P.)
  • Update: Added last_click_time fields to get_session_data API call (Ricardo)

Phew!

As always, let us know if you have any suggestions, comments, or brilliant ideas!

Zendesk Decision Tree Support App: Version 4

zendesk_zingtree-BLOG@2x

Travis A., who is implementing Zingtree for a video technology company, was keen on using Zingtree to try to reduce his team’s support ticket load. He also needed to import a synopsis of the Zingtree session data into his Zendesk tickets so his tech support people could get some time-saving background information.

Travis wanted to emphasize the condensed Q&A part of the session. Done!

Here’s the new look of the Zendesk Decision Tree Support app:

Here’s what’s new in the Zendesk Support App, version 4:

  • The Q&A view is now the default.
  • Any “back” or “restart” button clicks no longer show in Q&A view.
  • The answers clicked are highlighted, so it’s easy for a support person to see where the customer went at a glance.

You can download the latest Zingtree Decision Tree Support app for Zendesk here.

Need help installing the app? Instructions are here.

Travis' photo
This is Travis.

Thank you, Travis, for the awesome suggestions!