September 17, 2024
Case Study

B2C Marketing: Innovative use of technology as campaign subjects

SUMMARY:

When marketers think about tech, their focus is usually on what they’re using behind-the-scenes.

What if you shifted that focus on tech into the campaign itself?

Keep reading for an online travel agency’s digital media campaign that tested consumers’ eye for identifying AI, and a home maintenance company that made tactical shifts in its campaign to precisely target only customers looking for specific smart home technology.

by Daniel Burstein, Senior Director, Content & Marketing, MarketingSherpa and MECLABS Institute

Action Box: AI Expert Maker

Here’s an idea for infusing innovative tech into your next campaign – build an AI-powered expert. Click here and MeclabsAI (MarketingSherpa’s parent organization) will guide you through the steps to create effective prompts. Then click the profile button in the upper right to build your expert in the Engineering Panel.

Quick Case Study #1: How internal testing helped online travel agency launch innovative AI contest, get 1,830 participants

Founded in 2014, justfly.com offers deals on flights. It is a subsidiary of Momentum Ventures.

As part of its brand revamp, the team is launching innovative campaigns, testing them, and actively engaging with their audience. One such example was their Unmasking Intelligence online contest, which let consumers test their eye for identifying AI content.

BEFORE: Pressure testing online contest with real humans

The team conceptualized 29 pictures using artificial intelligence, then replicated and photographed them in real life. "By starting with AI-generated photography concepts and then bringing them to life through real photography, we wanted to see how we could use AI to enhance the creative process,” said Ayoub Hissar, Director of PR & Marketing Strategy, justfly.com.

They then altered the real photos with AI to create the final visuals. Players would interact by dragging a line across a photograph while viewing a percentage indicator at the bottom to guess where the real photograph ends. For a valid entry, a participant would have to complete all 29 photos. To make the game even more challenging, three of the photos were 100% real.

Creative Sample #1: Entrants drag the line to where they think reality ends and AI begins (black and white is the AI part)

Creative Sample #1: Entrants drag the line to where they think reality ends and AI begins (black and white is the AI part)

Normally I would show the creative samples in a linear fashion. But I realize for this campaign, I would would give away the answer in case any of our readers would like to play along at home (I’m sure you savvy marketers will be aces at guessing when a photo was manipulated with AI). So here are two AI-altered photos. See if you can guess where the AI alteration began in the photos. And then you can scroll all the way down below this article to see the original photos and compare them.

Creative Sample #2: AI-altered photographs

Creative Sample #2: AI-altered photographs

To ensure a smooth public launch, the team conducted an internal test of the contest by giving justfly.com employees early access to the landing page and game. This allowed them to thoroughly evaluate the user experience and identify areas for improvement – such as enhancing the submission form and optimizing mobile and desktop versions – and implement measures to prevent potential cheating.

The internal testing began on June 3rd. They communicated the internal contest to justfly.com employees through their internal channels, providing them with two days to visit the website, play the game, and share feedback on the landing page.

While testing internally, the team launched a social media teaser externally.  

AFTER: Closing loopholes to reduce the chance of ‘cheating’

At first, the team anticipated some entries with scores of 90% or higher when they launched the contest, given that some employees achieved 100% accuracy during the internal testing phase.

However, they soon realized that some employees ‘cheated’…or to put it another way…used ‘clever strategies’ to maximize their scores. And this was a key benefit of the internal testing, it allowed the team to close these loopholes before the public launch.

The first takeaway from the internal testing was that randomizing the order of the photos is crucial to prevent cheating.

Another key insight was ensuring the back-end programming of the landing page didn’t reveal the accuracy of each photo. They discovered that some developers were analyzing the source code. Other employees were copying and pasting the images into ChatGPT to identify which ones were AI-generated.

This experience also highlighted the need to tighten their terms and conditions to address loopholes – for example, not allowing repeated participation.

After making these optimizations to the contest itself, the team launched the campaign on Instagram, Facebook, LinkedIn, and X (formerly Twitter) on June 20th, supported by PR. They partnered with Instagram influencers Bianca Velardi (@bianca_velardi with 45,400 followers) and Nicole Nina Hu (@hutravelstheworld with 73,200 followers).

Bianca Velardi created content for three series of stories and one post, generating:

  • Reach: 5,000
  • Engagement: 500
  • Trackable Link Clicks: 271

Creative Sample #3: Collaboration with Instagram influencer Bianca Velardi

Creative Sample #3: Collaboration with Instagram influencer Bianca Velardi

Nicole Nina Hu also created content for three series of stories and one post, generating:

  • Reach: 7,000
  • Engagement: 750
  • Trackable Link Clicks: 348

Creative Sample #4: Collaboration with Instagram influencer Nicole Nina Hu

Creative Sample #4: Collaboration with Instagram influencer Nicole Nina Hu

Contestants earned accuracy points based on how close they were to identifying the correct images, with the goal being 100%. The closest to 100% would be the winner.

“Participants had a good eye for distinguishing AI from reality,” said Sandra Belbekri, Social Media & PR Specialist, justfly.com (here’s our own little teaser: we may publish the full participant score breakdown in a future chart article to give you an idea of how well consumers can identify AI use in marketing photography).

The original winner didn’t respond to their calls or emails, so on July 8th the team awarded the prize to the second-highest scorer in the contest – Lindsay Michelle Passmore.

Creative Sample #5: Winner announcement

Creative Sample #5: Winner announcement

Passmore won a Sony ZV-1 digital camera (to capture travel moments) and chose a trip to New York with a stay at W Hotel, Times Square.

RESULTS: Strong mobile participation

The landing page received 2,180 visitors and 1,830 total participants, with 70% of the participants completing all the photos. The vast majority – 86% – visited on mobile devices. The PR campaign received 241 media mentions, contributing to the direct traffic, which was 64% of the traffic the page received.

“We’ve learned that AI, when integrated thoughtfully, can offer valuable tools for innovation,” said Jade Rozal, Communications Manager, justfly.com.

Quick Case Study #2: Smart home company’s journey to a 1,856% ROAS from Google Ads

Upkeep Building Maintenance is a home maintenance and smart home installation company based in Dubai, UAE.

BEFORE: Broad keywords and no conversion tracking

Upkeep provides multiple services in home maintenance and smart home installations. Their main source of business was word of mouth. To get a more consistent flow of leads, they tried to run Google Ads.

Upkeep focused specifically on marketing their smart thermostat installation service. It was one of their most profitable services and had a lot of demand in Dubai.

But the campaign lacked structure. All keywords were dumped into one ad group and shared only one ad, so they were unable to create relevant ads to target specific intent. They went after thermostat installation in general, even though the only type of thermostats the company installs are smart thermostats.

There was no conversion tracking, so they didn’t have insight into which keywords were generating quality leads and which were wasting ad budget. “We had no idea which inquiries were coming from Google Ads. Because of this we had no idea of how well our campaign was performing and how we could scale it,” said Tanveer Anwar, CEO, Upkeep Building Maintenance.

The ads sent traffic to a generic landing page.

Creative Sample #6: Previous landing page

Creative Sample #6: Previous landing page

When the company brought on a team to work on their marketing, the team identified these areas for improvement on the landing page:

  • Lacked relevancy to a user searching for smart thermostat installation service
  • Unclear headline – ‘Home automation system installers in Dubai’
  • Lack of clear call to action
  • Looks unprofessional

The new team also engaged in a business discovery session to understand the service, the ideal customer, and the sales process.

“We have a rigorous checklist for auditing Google Ads campaigns. The first thing we always check is conversion tracking. If the account has no conversion tracking setup, we know we are off to a bad start,” said Saif Al-Jabbar Khan, Founder, Lead Ember Marketing Management (Upkeep’s new Google Ads agency in Dubai).

AFTER: Precise ad groups and segmented landing pages

Since the company only installs two smart thermostat brands, the team split the campaign into three ad groups:

  • Nest thermostat installation
  • Ecobee thermostat installation
  • smart thermostat installation

“Ad groups allow us to group a bunch of keywords and create a separate ad for them. Since we had three types of searches – a generic smart thermostat installation search, a Nest-branded search, and an Ecobee-branded search – it is advisable to create an ad group for each of them. This allows us to tailor the ad copy for each ad group based on the relevancy of the search,” Khan said.

The team also created a separate landing page for each ad group, which not only helped improve landing page conversion but helped with lead follow up as well. “We know if a lead is specifically looking for Ecobee or Nest smart thermostats. This allows us to further tailor our first contact,” Anwar said.

Since they were only targeting smart thermostat installations and only offered two brands it was important to eliminate all adjacent searches. So they added all other smart thermostat brands as negative keywords.

They also regularly used the search term report to identify similar keywords with a different intent and add them as a negative keyword – for example, anyone searching ‘AC thermostat Installation’ since the searcher is not specifically looking for a smart thermostat.

The new landing page had a clear primary call-to-action to contact via WhatsApp, along with a secondary call-to-action – ‘Have a custom order? Use our enquiry form below!’

Creative Sample #7: New landing page for ‘Ecobee thermostat installation’ ad group

Creative Sample #7: New landing page for ‘Ecobee thermostat installation’ ad group

 

The team implemented conversion tracking for WhatsApp and Lead Form contacts.

RESULTS: December revenue matched peak summer month

Previously, 11.7% of the company’s ad spend was wasted on non-smart-thermostat searches.

In four months of the new approach, the team was able to generate AED 207,261.75 (United Arab Emirates Dirham, the official currency of the UAE) revenue from Google Ads in new smart thermostat installation jobs across Dubai, maintaining a cost per lead of AED 53.37 (for our US readers, that would be $56,436.15 in revenue and $14.53 CPL).

Some more results from the first four months of this campaign:

  • 1,856% Return on Ad Spend (ROAS)
  • 209 new inquiries
  • 40% of leads became paying customers
  • AED 53.37 cost per lead
  • 18.68% campaign landing page conversion rate

“Winter has always been a low season for smart thermostat installations. We knew this strategy was a game changer when we checked our December 2023 revenue report and saw we had matched our peak month in summer,” Anwar said.

This article was distributed through the free MarketingSherpa email newsletter.

Related resources

MarTech Chart: Biggest technology innovation challenges

Innovative Marketing Leadership: It is okay to think outside the box but make sure you don't create surprises (podcast episode #82)

MeclabsAI: A brief glimpse into what is coming (go give you ideas for how to discuss upcoming tech breakthroughs)

Quick Case Study #1: Revisited

Here are the original photographs produced by the justfly.com team. Did you guess correctly?

Creative Sample #8: Original photographs

Creative Sample #8: Original photographs

Improve Your Marketing

Join our thousands of weekly case study readers.

Enter your email below to receive MarketingSherpa news, updates, and promotions:

Note: Already a subscriber? Want to add a subscription?
Click Here to Manage Subscriptions