“Your MarTech stack is the plumbing required to make personalization work”. Econsultancy founder Ashley Friedlein, said that at a recent breakfast seminar and he has a point. If your marketing technology platforms and tools don’t work well together, then data can’t flow from one system to another.

But before we get to that, it is worth pointing out that 57 percent of marketing leaders say they want to advance in personalization this year. No wonder: Personalization means using data to serve up content that is tailored, relevant and timely.

Shoppers return to your site because you provide offers that resonate with the individual, anticipate their needs and save them time.

Here are 3 examples of eCommerce personalization in a well-thought-out marketing stack:

1. Sending price drop emails

Price drop emails are personalized with a re-creation of the abandoned product. You send an email in response to the customer’s individual website behavior, triggered by a product change. The email is personalized with content based on what the shopper looked at.

2. One-to-one homepage personalization

You serve banners based on customers’ favorite category. A unique combination of product recommendations is based on crowd sourced data and the individual’s behavior.

3.Customized marketing emails

You send a different kind of newsletter to new customers versus lapsed customers. Content within the email is personalized on a one-to-one basis. For example, with a unique coupon code and AI-based product recommendations.

These examples all require coordination between marketing systems. Even if you have the best eCommerce platform, a great ESP, and a Customer Data Platform (CDP)… For personalization, you’ll need to access, aggregate and implement the following data, in near real time.

• Data about the individual customer’s behavior
• Anonymous data about similar customers’ behavior
• Data about the customer’s context (e.g. location and time of day)
• The customer profile
• Live product information (e.g. live pricing, stock levels and ratings)

A personalization engine collects this data and uses it to populate emails and web pages with relevant content.

Approaches to personalization in your MarTech stack

There are different roads to take towards a full-featured personalization stack.

One option is to invest in a dedicated personalization platform. This requires integration with existing technology, such as your ESP, eCommerce system and CDP.

Another approach is to rely on an all-in-one marketing suite. These platforms aim to wrap your ESP, personalization engine and CDP into one software. In theory, this facilitates personalization and reduces the number of systems involved.

At first glance, there are several reasons you want to consider an all-in-one solution for personalization. We’ll deal with each of these in turn and ask if you might be better served by a best-of-breed approach.

1. You want a single source of customer data to aid personalization

You want to create a single source of customer data to feed into personalization and other marketing efforts.

An all-in-one suite may seem like a logical way to move towards a 360-degree view of the customer. Perhaps the solution will even perform the role of a Customer Data Platform.

This approach does introduce a few questions that need to be answered, let’s consider the function served by each part of your marketing system.

Marketing software shouldn’t leave you feel locked in

Your ESP, CMS and personalization platform perform specific parts of your marketing strategy. You’ll want to experiment with these tools, mixing functionality until you get the desired results.

You may find you want to change your ESP after two years or try out different personalization options. If you’re using the same system to send emails, personalize content and manage customer data, it’ll be a huge project to migrate everything at once.

This could leave you feeling locked into features you don’t use, and unable to investigate new tools to help you meet your goals.

Single Customer View and your CDP

A Single Customer View (SCV) is a way of cleansing, matching and merging customer data. The aim is to provide a single source of truth for targeted marketing communications.

What does the Customer Data Platform do in this case?

A Customer Data Platform takes the Single Customer View to the next level. It makes the cleansed data available to integrate with your dedicated marketing software. It should be a core system around which other marketing software integrates. You’ll likely want to keep this system in place for many years.

In addition, your Customer Data Platform should be able to integrate seamlessly with any future tools you need to add.

How CDP works to solve the challenge (click for larger image)

The Customer Data Platform Institute offers the following definition of a CDP:
“A Customer Data Platform is packaged software that creates a persistent, unified customer database that is accessible to other systems”
Will your marketing platform be able to deliver all the benefits of a CDP?

“A persistent, unified customer database…”

Many all-in-one marketing platforms are built around an email marketing solution. They typically have an architecture which uses a fixed data model. This means limited ability to link together multiple database tables in a bespoke schema for each client.

For example, a business may have an ERP system, a CRM system and a web analytics system all from different vendors. These will all have their own unique database structure.

You should be able to configure the CDP to handle this structure. It shouldn’t be a process of mapping your data into a structure that the CDP can support.

“…Accessible to other systems”

A true CDP should allow easy, flexible access to all the data of any system held by the client. This is complicated by the complex and custom data structures mentioned above.

Typically, an all-in-one solution doesn’t have the APIs to efficiently and easily permit this kind of access. Clients will end up accessing the individual systems directly, losing the benefits of buying into a CDP.

If your goal is to clean up customer data and facilitate its use in marketing, you may be better off choosing a dedicated Customer Data Platform. This should be able to integrate easily with whichever ESP and personalization platform meet your needs.

2. You want to reduce costs by moving to one system

It might seem less expensive to adopt an all-in-one suite, rather than buying dedicated software to execute each part of your strategy.
Here it is important to consider how costs will play out in the long run.

Ongoing costs of your MarTech systems

All-in-one marketing clouds aim to replace all or most of your marketing functions. Marketers sometimes feel locked into increasing costs, as changing providers would mean replacing the entire marketing infrastructure.

Best-of-breed solutions operate in an open marketplace, with good competition between providers. As it’s relatively easy to switch solutions, vendors are incentivized to keep costs reasonable.

Suite vendors lack this incentive to price new features competitively. Take behaviorally triggered emails as an example. One client needed complex merchandising rules for displaying content in email. The online retailer wanted to only show a product once, even though a number of recommendation blocks were used, as well as frequency caps on content impressions and the ordering of products. They told us that it worked out five times more expensive to add this feature to their all-in-one platform, compared to a dedicated solution.

Integration with current and future technologies

Ask whether integration will be more expensive with the all-in-one software if you want to go outside the platform’s built-in capabilities. Will the platform’s APIs be well developed enough to work with future tech you might adopt?

Best-of-breed solutions need to be built for integration, so tend to work well with your core technologies – such as your CRM – and other point solutions. A best-of-breed approach frees up marketing teams to experiment with new technologies to meet your business goals.

3. You want to deal with fewer vendors

It’s tempting to rely on your all-in-one platform for everything, including personalization. You’ll only have to deal with one vendor, one invoice and one support team. If something goes wrong, there’s less likely to be finger-pointing between providers.

Yet, users of marketing suites cite lower customer satisfaction than those taking a best of breed approach. According to one vendor satisfaction report, only 53% percent of suite users feel that their feedback and requests are valued. In contrast, 87% of Medium-Enterprise ESP users say their requests are heard.

Find out whether you’ll have access to a dedicated support team to help you make the most out of the software. And consider whether the platform will be flexible enough to keep pace with your needs. Will a larger platform mean longer development cycles for new personalization features?

4. You want to empower your team with one set of tools

A marketing suite could reduce time spent switching between personalization, email and data tools.
By using only one system, you may be sacrificing control over the future direction of your marketing technology. As we’ve seen, it’s difficult to adapt to change when you’re locked into an all-encompassing system.

All-in-one platforms often result from mergers and acquisitions in the MarTech space. And many started out as email service providers. This means that one or more elements of the integrated system might not be best-in-class. You’ll likely need to integrate additional technologies to meet specific needs.

When it comes to personalization, a dedicated platform can open up a range of specialized features. You’d be hard pushed to deliver these through your ESP.

Here are some examples:

Granular, behavior-based segmentation

Suppose you want to send different browse abandonment programs to your customer segments. You can use segmentation based on a mixture of behavioral, customer and product data.

The system will need to collect:

• Behavioral data (e.g. the category and value of recently browsed products)
• Customer data (e.g. the shopper’s location and favorite brand)
• And product data (e.g. live pricing and availability.)

The marketer will need to be able to set rules to define which customers get which email program, based on this data.

The system will then need to identify shoppers who abandon a session, even if they aren’t logged in. Then trigger the appropriate email program within 30 minutes, based on rules the marketer assigned.

Email service providers may struggle to identify browsers who aren’t logged in, and to perform real-time, behavior-based segmentation.

Example of segmented and personalised email (click for full image)

Real-time dynamic content

Marketing suites with a heavy email focus tend to be light on real-time content tools. For personalized images, sophisticated product recommendations, or location-based content, you’ll likely have to add a specialized solution.

Intelligent product recommendations are a part of most eCommerce marketing strategies. Check whether your vendor can draw from multiple data sources. For example, to mix crowd-sourced recommendations, personalized suggestions, and AI-based rules. And what about serving a different mix of product recommendations to different customer segments?

You’ll also need to consider the range of channels and the variety of content you want to personalize. For a seamless customer journey, the customized experience should flow through your website, emails and mobile apps. A marketing suite with an email focus may not be the best solution to personalize your website.

User identification is key to successful e-commerce site personalization. With the right software for onsite personalization, you can identify prospective customers early and throughout their customer journey.

Initial identification can be through a modal form that doubles as an email capture form. Once the person is identified the software will start adding collected behavioral data about them. The customer behavior can be used to trigger personalized offers and guiding messages throughout the entire customer journey as the user moves between their inbox and your website.

Example of a user identification modal and a personalized message at a later stage in the customer journey. (img source: Triggerbee)

Parts of your site experience might even adapt while you are using them. In eCommerce, personalization is essential. Help your visitors along through the phases of the customer journey, guiding them to the final conversion destination.

That’s why a new form of guided selling is an interesting (sales- and) marketing development. Guide your prospect by providing them with personalized information. Show them in a few questions or clicks what product best suits their needs.

An amazing example of guided selling is the Sephora beauty assistant app, which helps customers choose their makeup items. This experience is close to being in-store and trying out different shades of makeup. So customers can develop a preference (without risk).

eCommerce marketers are already realizing that one technology is rarely the best choice for all parts of the ecosystem. This is exemplified by the shift to headless commerce.

Increasingly, brands are using front-end CX technologies to build websites separately from the back end. Marketers are able to build customized eCommerce systems that can quickly adapt to customers’ needs.

Choosing the best marketing personalization platform

Before committing to a software –before even comparing vendors – you’ll need a clear idea of what you want to achieve.

Consider the tools you’ll need to carry out your marketing strategy now and in the future. Ask whether a single provider can deliver on all of those requirements and decide where you’re willing to compromise.

Will a “good enough” all-in-one solution really be good enough for your customer data and personalization needs? Will your team feel empowered, or constrained, by wrapping these tools into one system?

Where best-in-class personalization is concerned, it’s worth considering a best of breed tool for the job.

The post Going all-in with eCommerce personalization? Consider these 4 things first. appeared first on Email vendor selection.

} .divTable.blueTable .divTableBody .divTableCell { font-size: 13px; } .divTable.blueTable .divTableRow:nth-child(even) { background: #F4F4F4; } .divTable.blueTable .divTableHeading { background: #DEE6ED; border-bottom: 0px solid #444444; } .divTable.blueTable .divTableHeading .divTableHead { font-size: 15px; font-weight: bold; border-left: 0px solid #D0E4F5; } .divTable.blueTable .divTableHeading .divTableHead:first-child { border-left: none; }

.blueTable .tableFootStyle { font-size: 14px; font-weight: bold; color: #FFFFFF; background: #D0E4F5; background: -moz-linear-gradient(top, #dcebf7 0%, #d4e6f6 66%, #D0E4F5 100%); background: -webkit-linear-gradient(top, #dcebf7 0%, #d4e6f6 66%, #D0E4F5 100%); background: linear-gradient(to bottom, #dcebf7 0%, #d4e6f6 66%, #D0E4F5 100%); border-top: 2px solid #444444; } .blueTable .tableFootStyle { font-size: 14px; }

A successful triggered email programme might very well be the most engaging tactic in your marketing toolkit. But how to get started and select the best triggered email provider? Let’s look at the pros and cons of different email trigger solutions, and the questions you should ask when choosing a vendor.

Why is triggered email so popular?

Triggered emails treat customers as individuals, sending them messages based on an action they have just taken. They meet marketers’ ultimate goal of getting content in front of customers at the moment when they are most likely to engage. That is why they work so well.

Sales uplift from shopping recovery emails alone can exceed 10%:

(Source: Fresh Relevance Real-Time Marketing Report Q4 2018)

The above figures are for shopping recovery emails, but triggered messages can take a range of forms.

Here are some types of triggered emails:

• Browse abandonment emails
• Post-purchase emails
• Wishlist emails
• Back-in-stock emails
• Price-drop emails

How do I want my triggered emails to work?

Before you decide on an approach, let’s look at the moving parts of triggered emails.

Triggered emails automatically serve relevant marketing messages to customers, based on their (inter)actions. Customers who raise a certain signal (such as abandoning a shopping cart) are put into a trigger programme. Trigger programmes perform actions, such as sending the user an email or series of emails.

Here are some actions that could trigger an email programme:

 
Looks simple so far!

But you won’t want to send a one-size-fits-all email to every customer who completes an action. For example, it may be more effective to send browse abandonment emails only to your most engaged customers. Or you might like to run a different trigger programme for high spenders versus low spenders. Conditions and rules can be set to determine which trigger programme a customer should receive.

Imagine your aim is to increase the effectiveness of browse abandon campaigns. You could treat customers differently depending on whether they have previously made a purchase:

Run trigger programme A for browse abandoners who have previously purchased, and recently spent a lot of time on the website.

Run trigger programme B for browse abandoners who have previously made a purchase, but haven’t spent much time on the website recently.

Send no email to browse abandoners who haven’t previously made a purchase.

All this means that to get started with triggered emails, your eCommerce system and ESP need to be able to talk to each other in real time.

Online brands have options when it comes to setting up triggered email campaigns. You can:

• Automate triggered emails using your existing Email Service Provider (ESP).
• Adopt a purpose-built triggered email platform to do the heavy lifting.

Here, we’ll explore the business case for each approach, and look at the questions you should ask when choosing a provider.

Setting up triggered emails using your ESP

To automate triggered emails using your existing email system, you’ll have to create an integration process to transfer eCommerce data into your ESP.

The system will need to collect, transfer and implement data in near-real-time so that triggers don’t misbehave. Where you allow multi-stage programmes, there must be conditions to ensure they get canceled if behaviour changes. You won’t win shoppers’ loyalty by sending them a cart abandonment email after they’ve already made a purchase!

5 things to consider when sending triggered emails from your ESP

1. Can you read data from your eCommerce system into your ESP?

Emails containing product information should be up-to-date, and consistent with what the customer browsed on site.

This means that your ESP will need access to the following information:

• Customer list (e.g. first name, last name)
• Transactional events (e.g. cart events, purchases, returns)
• Behavioral information, (e.g. product browsed, view cart contents)
• Product data (text, image urls, prices)
• Stock level, for in-stock products

There are a few ways to connect the data with your ESP.

1. Use custom data extensions
Popular eCommerce providers such as Magento can use custom-written extensions to pass data to an ESP or other platform. This can work well for transactional, product and stock information. However, behavioral information is usually only captured when the shopper has logged into the eCommerce site.

Most shoppers don’t login until they’re ready to purchase, and non-purchasers never log in, so you’ll miss out on behavioral data for most of the shoppers. Remember that plugging in an extension can disrupt your existing eCommerce system, so check with your eCommerce system manager.

2. Scrape data with JavaScript
Another approach is to scrape data from the front end via JavaScript. This works well for behavioral, transactional and product information. However, stock levels and more confidential product information is usually not exposed on the front end, so you’ll miss out on marketing opportunities like back-in-stock alerts and optimizing recommendations based on product margin.

This JavaScript scraping approach is easier and more reliable if your eCommerce site uses a standard data layer format (such as the W3C Digital Data Layer) to expose data in machine-readable format on the web page for use by other providers.

Many eCommerce providers can use a standard Digital Data Layer to expose stock levels, product information, and user details. If you have to re-implement a new data layer each time you add a new vendor (e.g. an email service provider), the implementation time and server load will increase, so it’s a good idea to use a standards-based one like the W3C version.

3. Use e-commerce data feeds
A third approach is to take a data feed from the eCommerce platform. Most platforms can send feeds of transactional / product data for use by other systems. This approach doesn’t usually attempt to handle behavioral data.

Each of these 3 approaches has some limitations. The best approach to fill in the gaps is to use a combination of the different methods, to create a seamless capture of all available data. For example, capture behavioral and product data using JavaScript, and combine that with historical transactional and stock level data from a data feed from the eCommerce platform.

2. Does your ESP allow you to set flexible business rules?

You’ll have to add conditions/rules to ensure that emails go to the right people. These rules should take into account customers’ behavior and your current business objectives.

Imagine a marketer is using a simple browse abandonment trigger programme, but wants to use a new programme for customers looking for deals on holiday gifts. Rules need to be set so that regular customers receive the standard browse abandonment email, while customers who browsed lots of holiday deals receive the new festive programme.

3. Can you manage frequency and exclusion?

It’s crucial to set marketing rules to make sure browse and abandoned cart recovery emails aren’t sent too often, and customers don’t get the same email twice.

For example, if you’re sending multiple types of cart abandonment emails, you need to be able to prioritise trigger programmes. Each customer should receive only the most relevant email series for them, and not get enrolled into multiple programmes simultaneously.

Shoppers will be irritated – and rightly so – if their inbox fills up with multiple, similar emails at the same time.

4. How will the email series respond if the situation changes?

One trigger programme can send a succession of emails at different times. For example, a cart abandonment email could be sent 30 minutes after abandonment, 24 hours later, and then after one week.

If the customer makes a purchase before the email series is complete, the programme should stop sending emails. Otherwise, customers will be frustrated by an irrelevant email. This means that the rules you set in your ESP need to be informed in real time by purchase data from your eCommerce platform.

5. How will your ESP personalise the content?

For many types of triggered emails, you need to include personalised content about the context. For instance, the products which have dropped in price after being viewed by the shopper.
For best results, you need to go further. Will you be able to increase the converting power of triggered emails by including smart content based on the individual’s history and preferences?”

For instance, a personalised hero image based on the shopper’s preferences, or product recommendations filtered by the shopper’s favourite category.

When to use an ESP as triggered email software

Using a general-purpose tool like an ESP can work well if you’re content with sending basic triggered messages such as purchase complete emails. Cart abandonment can also work, but the conversion rates will be lower if you can’t identify customers who weren’t logged in when they browsed or carted.

If your ESP is lacking browse and form abandonment, or any type of sophisticated personalisation, you’re likely better off considering a purpose-built solution.

Find the best specialised triggered email software

There are several dedicated triggered messaging systems which help tie your ESP and eCommerce systems together. Some handle just cart abandonment, while more sophisticated systems handle more applications including email and website personalisation.

Here’s what you should consider when choosing a triggered messaging provider:

1. How much does the triggered email software cost?

First, check the cost of integration of the triggered email solution with your eCommerce platform. Support for data layers may reduce this cost. The W3C Digital Data Layer is a standard way of exposing eCommerce data such as cart contents to third-party providers and partners on your site.

This makes installing plugins and tools much simpler. There’s also less strain on your technical teams, as there’s no need for lots of custom coding.

Also, consider how the solution is priced.

There are three main pricing models to choose from:

• Affiliate pricing model: Some solutions are priced based on the revenue generated by the triggered programmes. This can work well if you’ll be sending emails on a small scale, as you’ll pay only for what performs. However, it will be difficult to forecast costs and plan your budget if you’re sending emails at scale.
• Basic/Pro/Premium pricing bands: Paying for functional packages makes it easier to forecast costs and control your budget. But you may end up missing out on some features out of your price range, or unnecessarily paying for functionality you don’t use. Often, there is an additional CPM cost for email sends or other personalisation capabilities, which makes costs unpredictable.
• All you can eat model: Some solutions let you take up a subscription based on the size of your business (e.g. your website’s monthly page views), with no additional variable costs such as cpm or impressions. You get control of your marketing budget, and the cost scales as your company does. This model is mostly fit for mid-market and enterprise level users with a large customer base.

Many triggered email providers offer a range of other email and web functionalities. Find out if you can pick and choose from the available modules. This will avoid duplicating your existing functionality or paying for features you don’t need.

2. Does the platform have access to enough data?

Make sure the triggered email tool can collect and coordinate data from multiple sources, such as your website, ESP, and eCommerce system. Methods of data collection might include APIs, file transfer by FTP, or web scraping as appropriate.

Coordinated data ensures that the information displayed in triggered emails – such as the current price and product image – is up-to-date and matches what the customer will see when they click through to your website.

Only a platform that sits comfortably between your eCommerce platform and ESP will allow you to collect enough behavioural data to personalise and target emails effectively based on customers’ preferences and lifecycle stage.

For instance, you could run different variations of a triggered email programme depending on the customer’s level of engagement, or for new visitors versus established customers. And when triggers are based on real-time data straight from your eCommerce site, there’s no risk of a customer receiving a cart abandonment email when they’ve already made a purchase.

Also, check if there are any API or other usage limitations which may prevent your solution from working correctly or restrict the amount of data available.

3. Will it work alongside your existing (and new) technology?

For minimal friction when introducing the new system, adopt a platform that can work with, rather than duplicate, the software that you already use. This includes:

• ESP for designing and sending emails (e.g. SendGrid, MailChimp, Adestra, Copernica)
• eCommerce platform for collecting behavioural and transactional data (e.g. Magento, Shopify, Salesforce Commerce Cloud)
• Customer Data Platform (CDP) for a holistic view of the customer (e.g. BlueVenn)
• Ratings and reviews provider to harness social proof in your email content (e.g.Trustpilot, Feefo)

Implementation will be more disruptive if your marketing team has to learn to use a new ESP functionality. Ideally, your third-party triggered messaging system should be able to easily switch with you if you decide to change ESP.

4. How are customers identified?

To reach the maximum number of shoppers with relevant emails, you’ll need a platform that helps you identify as many customers as possible. Logically, the more shoppers you are able to recognise, the better you can follow up with them when they interact with your company.

Taking sales uplift from cart abandonment as an example, this chart shows a clear correlation between identification rate and sales uplift.

Data from Fresh Relevance clients deploying cart abandonment emails only.

It’s not a straight line, because your most loyal customers will also be the easiest to identify. But if you can increase your visitor identification rate from, say, 20% to 50%, your sales uplift will increase significantly.

All marketing platforms recognise shoppers who log in or register, and some recognise shoppers who have identified themselves in the past, for example using cookies. But what about visitors who normally use their desktop but who are using their mobile phone this time? They’ve never identified themselves on their current device, can you identify them?

Consider a solution that allows you to identify shoppers on whichever device they are using and tie their browsing behaviour back to their history, even if they haven’t identified themselves in this session.

5. Can you re-build a shopper’s cart across devices?

Suppose a shopper clicks a cart recovery email on a different device to the one they were using when they abandoned the cart. Because it’s a new device from before, your eCommerce system may start a new session with an empty shopping cart.

Some triggered email providers offer multi-device cart rebuild to automatically refill the cart on whatever device the customer is using. This significantly reduces friction in the customer journey.

5. How easy is it to personalise emails?

Over one in four consumers is more likely to be loyal to a brand that sends emails tailored to their needs and interests. This includes a custom header image based on the shopper’s favourite category, and relevant product recommendations based on previous searches or similar products. It’s not as complicated as it sounds: the ideal third-party platform will collect the data and run the necessary analytics, while marketers concentrate on how the content should be displayed.

6. Can the triggered email software handle sophisticated triggers such as price-drop and back-in-stock?

Sophisticated emails like price-drop and back-in-stock require a system which can target shoppers who have viewed a product which has then dropped in price. Or who viewed an out-of-stock product which has since returned to stock.

These trigger types have even more moving parts and require data analysis, so check carefully that the provider handles them properly.

Conclusion: Getting started with triggered emails

The impressive ROI from triggered emails means it’s worth taking the time to weigh up all the options available.

If you’re looking at adapting your existing ESP, consider how much resources will be needed to get the system up and running and whether you’ll be able to deliver all the functionality you need to meet your business goals.

There’s no one-size-fits-all approach to email marketing. But if your aim is to send highly relevant, personalised emails to as many customers as possible, adopting a purpose-built solution could be the way to go.

The post How to choose the best triggered email provider appeared first on Email vendor selection.