Optimizely: The reason my web page might not look like yours

To introduce you to the company I chose for my presentation, let’s start off by giving you the 5 W’s and H of Optimizely: (Hint: Use these if someone on our trip says to you, “So, tell me about Optimizely!”)

Who: Optimizely was founded by Pete Koomen and Dan Siroker, both ex-Google employees. Koomen was a product manager for Google’s App Engine and Siroker served as Director of Analytics for the first Obama campaign.

What: In short, Optimizely is an optimization software that provides a platform for A/B and multivariate (multiple variable) testing. Operating under a SaaS model, the service allows for experimentation with robust analytical results.

When: The company was founded in January 2010

Where: The company’s headquarters is in San Francisco, CA (duh!). They have other offices in Netherlands, Germany, the UK, and Australia.

Why: According to their website, by using Optimizely, companies can run, manage, and share ten times the number of experiments across marketing or product development teams.

How: Similar to Airbnb, Optimizely went through the Y Combinator program in early 2010.

Funding Rounds:

giphy (2)

Initial Angel Funding: $1.2 Million

Series A: $28 Million

Series B: $57 Million

Series C: $58 Million

Product Offering:

Currently, there are two versions of Optimizely: Optimizely Classic (Free) and Optimizely X (Premium / Paid). Recently, the company announced it will be discontinuing Optimizely Classic, effective February 28, 2018. In other words, by the time we visit, only Optimizely X will be available.

On the website, Optimizely X is marketed so intensely that I had to conduct a separate Google Search just to find Optimizely Classic. Eventually, I found a small blog post that provides introductory information to users. Essentially, Optimizely Classic can only be used to analyze experiences using A/B testing. For experimentation, personalized experiences, recommendations across digital channels, and better technical support, Optimizely X is the way to go.

How Optimizely is used:

Regardless of which version you choose, the software is implemented by placing a snippet of code in your website’s or app’s code. Additionally, plugins exist for content management systems such as WordPress (!!!) or Shopify. Maybe we should install Optimizely Classic on this site to see if we can bring in more outside users!

Screen Shot 2018-02-05 at 8.51.27 PM

Once the product is installed, Optimizely provides tips for users on how best to use their service. As we have learned, running experiments and collecting data for no strategic purpose does not provide value to a company. For that reason, Optimizely provides rich and diverse content on their site to exhibit their value as a software. This content is especially important for Classic customers thinking of making the switch to the X version. (And soon, they will have no choice).

Below is a graphic on how to create an experiment using Optimizely. Since we are not using the site and I may cover some of this material in my presentation next week, I will only speak to these steps at a high level.

Screen Shot 2018-02-05 at 2.14.08 PM

To start, customers need to decide what variation on their site they want to make. For example, an E-commerce site, Attic & Button, decided to change the picture users saw on their home page, depending on the visitor’s location and style preferences. With a goal of increasing time users spent on their site, Attic & Button launched an experiment and let Optimizely go to work. Using Optimizely’s Stats Engine, the Attic & Button marketing team was able to see which picture(s) was / were performing best. From there, the team could keep using the winners, ditch the losers, increase site traffic, and learn from the experience.

Screen Shot 2018-02-05 at 8.00.03 PM


On their website, Optimizely has published case studies from top companies and organizations such as BBC, HP, Sony, Movember Foundation, TomTom, Bleacher Report, Jawbone, StubHub, and many more. What is so fascinating to me is that each client uses Optimizely to address and solve a different issue. That’s how you gain market share!


google-optimize-1-1Google Optimize

Why it’s better: More compatible with Google Analytics and AdWords

Why it’s worse: You can only run three experiments at a time

Screen Shot 2018-02-05 at 9.17.40 PM


Why it’s better: Strong customer support, more sophisticated personalization tools

Why it’s worse: Less powerful visual editing capabilities


Why it’s better: Hotjar’s expertise lies in discovering areas for improvement and experimentation

Why it’s worse: Unlike Optimizely, it is not a one-stop shop

What I think makes Optimizely so great:

Optimizely seems like such a useful tool because it can be used by those who aren’t well versed in coding, such as members of the marketing team, as well as by full-stack developers. Furthermore, behavioral targeting occurs in real-time and machine learning is used to make relevant recommendations to custom-audiences.

In my presentation, I plan to take a deeper dive into the features that make up Optimizely X. Stay tuned!

Questions for our visit:

giphy (3).gif

1. How did the company’s experience in the Y Combinator program play a role in its success?

2. With the discontinuation of Optimizely Classic, how does Optimizely’s lead generation strategy change?

3. What are some of the smallest changes customers have made that have led to immense success?

4. Which tools within Optimizely have helped nonprofit clients the most?

5. At what stage of a company’s website / app is it best to implement Optimizely or a similar tool?

Before you go:

If you don’t already, please follow Optimizely on Twitter! Be sure to share your favorite tweet in the comments section below!

11 thoughts on “Optimizely: The reason my web page might not look like yours

  1. Nice post Kipp! I’m interested to hear the main reasons behind discontinuing the Optimizely Classic software offering, does it have to do with increasing ROI? It’s cool reading about the competitors and how their offerings differ. It definitely seems like Optimizely has the broadest capabilities with a greater convenience to customers. Looking forward to your presentation to learn more about Optimizely!

    Liked by 1 person

  2. I can’t wait to hear more about Optimizely during your presentation! I’d love to hear about their experience in the Y Combinator program. It had such a significant impact on Airbnb, I’d have to imagine it’s role in shaping Optimizely was just as important. Also, like Camille, I’m curious about the decision to discontinue Optimizely Classic. It will leave approximately 70,000 websites without the service. This was one of the only articles I could find, which is an interesting read: https://venturebeat.com/2018/02/05/optimizelys-decision-to-ditch-its-free-plan-suggests-a-b-website-testing-is-dead/

    Liked by 1 person

  3. Nice post and very well-researched! I like how Optimizely seems so simple to use. Like you said, not everyone can be an expert web developer. Websites and experimental results can definitely help entrepreneurs to refine their products or business models. It seems like a bold move by Optimizely to discontinue their free Classic service. I typically like to try things out before committing to them. However, Optimizely’s credibility and record of success seem to justify an investment by any fledgling or established company. Looking forward to learning more!

    Liked by 1 person

  4. Great post, Kipp! A/B testing is a term a bit like big data: always thrown around, but many people don’t fully grasp the concept. However, it seems clear that Optimizely is designed to be friendly to all user levels and easily adoptable. I loved the analysis of their competitors, I’m interested to see if they decide to focus on things like discovering improvement areas, customer service, and personalization in the future. Can’t wait to hear more!

    Liked by 1 person

  5. Really great job, Kipp! Noting that Optimizely gives employees other than the programmers easy access to this powerful experimentation really sold me on the value the company offers. It’s a very interesting choice to discontinue the free version. While I’m not hosting websites that could necessarily benefit fully from their optimization software, I would have loved to play around with it. I’m most excited to hear about the most creative ways customers have used their software as the possibilities are really endless.

    Liked by 1 person

  6. I think Question #5 hits the exact nail on the head for what I’m most interested in for this business: when is it best to implement an A/B test? Overall, I’m fairly interested in where A/B test outsourcing makes more sense vs. developing in house. At some point in growth, a company has to move these insights in house to complement efficient product design (think about how often Facebook runs tests: 1000’s of times per day) but what stage is that typically?

    Liked by 2 people

  7. This was a really clean, informative blog! I came away from this article with a great sense of Optimizely’s offerings and its place in the market. The Attic and Button experiment is really interesting and a great example. Does Optimizely have integrations with CRM software, and can it customize A/B testing down to the individual level? Hypertargeting like that could be complex, but would also garner great insights about complementary products based on past purchases.

    Liked by 1 person

    • Thanks, Katie! From what I have seen, CRM integration is possible but not widely publicized. That said, a partnership with a company like Salesforce or HubSpot seems logical. This is certainly something to watch as the company shuts down Optimizely Classic this month. Regarding hypertargeting, audiences can be highly customized in the software. I was unable to find specifics, but I would imagine there is a filtering cap of sorts (Let’s say 5 or 6 filters) to prevent tiny sample sizes and therefore inaccurate results.

      Liked by 1 person

  8. Kipp, great post! It’ll be interesting to see if these websites start integrating AI alongside with basic user feedback to identify ways to improve your site and appeal to a greater audience. Maybe they already do. Also, I’ve always wondered if there’s any example of over optimization. By this, I mean times when companies have tried to optimize there websites so extremely that it’s ultimately created more confusion for customers through brand inconsistency than good.

    Liked by 1 person

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s