In good old days, before the proliferation of online shopping, the neighbourhood shop owner would remember you by your name, know your family members, ask about your kids, knows your last purchase, ask you to consider new arrivals and gives you sound advice to purchase next under your budget and well suiting yours needs.

That’s personalization.

We all love personalization.

As a consumer, it makes us feel more human.

It works for the brands as conversion rate, and loyalty improves.

What is web personalization?

Web personalization is chasing the north star of human to human personalization.

Now, we have data points and technology to make it happen. But in reality, the situation is far from even close to human personalization.

We have the information, but we get creepy sometimes. The downside is enormous for a messaging gone wrong.

Big brands like Amazon are famous for their personalization on every part of the customer journey.

So considering the effort and risk, rewards are enormous for personalization.

Personalization has improved conversion rate, time on site, average order value and even customer loyalty for the website.

Challenges of Personalization

Doing on scale comes with many challenges -

Not having enough data to run personalization.

(Image)

Not handling the cases where personalization does more harm than intended good.

Personalization has many viral creepy stories. (Case + Image)

Data stitching to show effective personalisation.

As data collected from multiple sources have a tough time getting stitched together. Google, Facebook all run their walled garden which makes

And finally -

The biggest challenge is personalization is very very costly for the brand from execution and return on investment with fear of getting backfired.

Personalization is a costly affair in terms of effort for a small group. Now suppose for a hero image banner, once you start with. Consider this -

One Hero Banner Headline will become -

For 4 traffic sources (Paid, Social, Organic, Referral) -> Total Effort = 4 Headline.

Also, include 2 gender customisation -> Total Effort = 4 X 2 = 8 Headline

Also, include 4 timezone -> Total Effort = 4 X 2 X 4 = 32 Headline

And so on.

So you see things go out of hand very quickly for personalization in terms of efforts.

Also after this mammoth effort, you might get 10 people in a bucket to experience your personalized experience hardly showing any measurable impact on the bottom line.

So before going on the personalization techniques, we might need to have a framework to get this done.

Framework For Personalization Program

As you are always spoilt for choices in personalization. It’s always better to work around a framework to get things going.

You can’t do everything in personalization.

Consider this simple framework for running program -

1. Identity Most Effective Customer Cohorts -

A massive effort needs justification and numbers at the end of the day, so you need to start with a large audience possible for personalization.

Customer cohort needs to be substantial to worth your time.

So familiar places to look at the audience and make an intersection of all the parties possible -

Customer cohorts are mostly -

  1. Traffic Segments
  2. Demographic Segments
  3. Geographic Segments
  4. Technology Segments
  5. Behavioural Segments

Will discuss them in detail in the next section.

2. Identity the experience change for testing

There are multiple things you can customize for each segment.

Good brands just personalize the landing page, and Great brands personalize the whole journey, including multiple subpages.

For the sake of simplicity, ask this question -

What we can change to make the experience more personalized for this customer?

Think along these lines for improving the experience. These are some of the areas to start with.

  1. Giving more clarity to the customer
  2. Reduce clicks for the customer
  3. Increase brand trust for customer.
  4. Decrease fear for the customer.

Change Areas might include:

  • Banners
  • Content
  • CTA’s
  • Layout
  • Offers
  • Overlays
  • Emails
  • Push Notifications
  • Merchandising Rules
  • Messaging
  • Recommendation Strategies

3. Build a Hypothesis and Anticipate Impact

4. Prioritize

5. Execute

Website Personalization -

Without customer data -

With customer data