Testimonials for Your Business — Recipe for a 4x conversion increase
It’s basic human psychology: testimonials work simply because of the way our brains work. Social proof material (i.e. testimonials and word of mouth) is powerful in impacting human behavior. This was a post I published while I was working for Crossover, after years of executive work for global software companies, I built Exceptionly for revolutionizing the software talent industry through objective testing.
Ironically, even though we’re living in a world of technology and the age of online businesses, we need to go back to a New York City sidewalk in 1969 in order to fully understand today’s customer behavior. According to research by Stanley Milgram, — who hired actors to stop and look at the sky in a specific spot — people emulate the actions of others in an attempt to reflect correct behavior for a given situation in which they are unsure of how to act. Buying a product or service is no different. Customers have been buying products for centuries now. But as life got faster and the world got smaller, we have learned to make purchasing decisions based on the experiences and opinions of our peers. For those who are interested in learning more about the psychological background of this phenomenon, I suggest reading Implicit Egotism by Brett W. Pelham, Mauricio Carvallo, and John T. Jones.
In this post, we’re going to deep-dive into four steps of bringing testimonials into your business: Planning, Production, Distribution, and Monitoring & Improvement. We’re going to discuss this subject based on real figures from a successful initiative we ran at Crossover. We hope that you’re able to learn something from what we’ve worked on over the past couple of quarters, and apply our learnings in your own business.
Knowing your products and services
This might seem incredibly basic, but many professionals do not have a basic grasp of their own company’s products. Try this exercise from the book “Content Inc” by Joe Pulizzi:
Try to ask yourself 40 questions directly related to your products or services. This might not be easy. If you can’t ask yourself 40 questions on a subject, you are not even close to being an expert of it.
Make sure you understand your existing customer base and their behavior. Identify your customers, and understand why they bought your product in the first place. Make sure you are covering every single category or type of products and services you’re offering.
Identifying content types
Once you have clearly mapped your existing customers, the next step is to identify what type of content (and which channels) you should use based on where your target audience is most active. Some examples are:
- YouTube Videos
- Blog Posts
- One-Pagers (short, written content pieces)
- Two Liners (Tweetable, “sticky” quotes)
- Social Media Posts
- Printed Materials
- Conventional Media Materials (TVs, Radios, Newspapers)
In our scenario, since we’re looking for global talent primarily from the software engineering industry, Linkedin was our primary source of “customers” (i.e. talent) and we decided to go with YouTube videos, One-Pagers, Two Liners, and Social Media Posts.
Now we have identified our target audience and what type of content we need to produce. Before you produce any content, I strongly suggest you read “Made to Stick” and “Content Inc”. Both of these books have profoundly impacted the way I look at how I create meaningful, attractive content.
Listening to customers in an organized way
Back to Crossover’s initiative for a moment: our goal was to attract more talent from Eastern Europe and Latin America, but we wanted to start by understanding why our existing talent had decided to join Crossover in the first place. To this end, we ran a storytelling contest among our existing partners. We awarded $10K for the most profound story and collected hundreds of real stories about why people had chosen to join Crossover.
Then we planned a two-week content production tour in Eastern European countries. We met with our existing customers (again, current Crossover partners) and heard more about professional and personal reasons they chose to work with us. This was primarily focused on video content production. We knew who we wanted to film (based on identifying top stories from our $10K storytelling competition). We carefully created a list of questions for our interviews — we selected these questions because they provide structure for our interviews while leaving room for personal stories from our partners:
- What were you doing before joining Crossover? What were you looking for personally or professionally when you decided to join Crossover?
- What are you learning at Crossover?
- Tell me about a specific project you are working on/have worked on at Crossover that you thought was interesting.
- Has Crossover enabled you to do anything personally / professionally that you weren’t able to do before joining Crossover?
- What are your goals for the next couple of years at Crossover?
- Anything else? (miscellaneous / open-ended…)
We visited Hungary, Romania, Poland, Russia, Ukraine, and Turkey. Before we visited, we arranged local video production teams based on clearly defined video recording hardware specs and built a checklist. We also conducted pre-interviews remotely. These were critical in allowing us to identify the most profound components of our partners’ stories and made our filming time more productive. When we were on site, we met our “customers” in a casual way, broke bread together, and filmed over 40 people.
Organizing and repurposing raw material
Pre-interviews and filming should be approached with utmost intentionality; collecting the right raw material are critically important. But the real magic begins during post-production.
In this step, you clearly define what you want to highlight from all of the raw material you just produced. For this, we transcribed the raw footage we collected from over 40 people. It may sound like hard work but it’s worth it, simply because it gives you a full map of your content inventory in a searchable, browsable way. Also, there are a number of tools on the market that allow you to make the transcription process much less manual (we used a hack with Zoom video conference audio transcription for this step). You need the full transcription in order to cut the clips which give you the stickiest story possible.
Keeping an inventory of sentences and time tagging them on Google Sheets will help you create compilation videos or videos with different themes — female-only videos, geography-specific videos, etc — with significantly lower costs going forward.
A full inventory of the testimonials we produced can be found here. Following these production steps brought us quality testimonial material like this as an example:
Building ready-to-use testimonial material inventory
After the production stage, we used Google Sheets and Airtable to collect all of our ready-to-use material in one place. We tagged them with keywords and categorized them. If your business has more than one product or service, you may find this practice useful.
Defining the distribution channels
First, we listed customer-facing communication channels;
- Homepage — Index of Crossover.com
- Social networks — Linkedin, Facebook, Twitter, and Glassdoor
- Follow-up emails — Emails our candidates (customers) start to receive after they apply to any of our positions
- Automated emails — Informative emails being sent by our platform informing customers regarding their application process.
- Product-specific landing pages — We had about 50 landing pages for every single product (position) we have to offer.
- Index page for social proof material — A homepage for all of the testimonial material you just created, the place where testimonials live. Currently, we have about 9 high-quality video testimonials, 39 long-form one-pager stories, and 167 short-form two-liner stories published.
Integrating social proof material into your business flow
Once we’re done with mapping all the possible integration channels, we started integrating one by one. First, we added the highest quality testimonial material to our website home page.
Then, our potential customers started to receive our messages integrated into our automated emails as a part of the signature:
Every day, we strive to improve in ways that enrich our teams, our products, our services, and ourselves. Innovation is the fabric of Crossover’s culture. Hear more, directly from our Partners.
We also integrated our messages into the signatures of the emails going out from our candidate experience team, who are working in the front-line of our business engaging with our customers:
Our mission is to connect the most talented professionals in the world with unprecedented access to high-impact, high-paying roles. What is it like to work at Crossover? Hear more, directly from our Partners.
Then we used our two-liners for our product-specific landing pages:
We created a section dedicated to human stories in our landing pages, stories related to the specific product we’re offering. If the position is in the Engineering domain, we presented human stories from the Engineering domain.
We also prepared an extensive social media calendar (using Hootsuite), where we scheduled social proof material to be published and shared with our followers every week.
We’re still discovering paid and earned distribution opportunities for our testimonial content. We’re motivated to go forward with testimonials because of the figures I’m going to present in the next section; Monitoring and Improvement.
Monitoring and Improvement
After starting distribution, the simplest approach would be running A/B tests, comparing the metric performance of landing pages with social proof content with the ones that don’t have it. Because we have so many landing pages and so many moving parts in our landing pages, we started with a broader approach.
Mastering Google Analytics
After ensuring that we pushed content at every stage and every channel available to Crossover, we created three simple goals on Google Analytics, based on customer (i.e. candidate) behavior in our sales (i.e. application) process:
- Application for the role (equivalent to adding a product to a cart for most businesses, or expressing an intent to purchase)
- Completing mandatory tests (filling in required information)
- Making it to the marketplace (equivalent to purchasing a product for most businesses)
Using cookies data and URL variables on Google Analytics, we are able to count how many customers are reaching each of these stages. We created two cohorts using the dynamic segments feature on Google Analytics (with a little Regex magic using ga:pagePath dimension). Based on the URLs visited, we created:
- Cohort 1 — Customers who saw our testimonials
- Cohort 2 — Customers who didn’t see our testimonials
For this report, we used an integration between Google Sheets and Analytics. If you do not know how to configure this integration, check out my recent blog post on this topic here.
What we discovered was fascinating. Based on behavioral analysis data over 700K website users in 10 weeks…
- Customers engaged with our social proof material experienced a 30% lower bounce rate.
- Customers engaged with our social proof material were 3.6 times more likely to apply to our jobs.
- Customers engaged with our social proof material were 2.5 times more likely to complete their testing process.
- Customers engaged with our social proof material were 4.8 times more likely to successfully complete the testing process and get into the marketplace.
Crossover is handling up to 100K unique users and tens of thousands of applicants every single week. Our next challenges are to increase the number of customers engaged with our social proof content and push our coverage from 5% to 30% in the Q2 of 2019. The data already indicates that testimonials are highly effective in driving the behavior we want; now, we want to spread their impact.
It’s Easter season so here are your Easter Eggs; Free material straight from the field... Grab your copies today!
About the writer: Founder of Exceptionly, revolutionizing the software talent industry by leveraging his unique big dataset of over 2M hands-on tested software engineers around the world.