Designer: Eliza Newman-Saul

Business:Lauren Kainski, Tanay Chaturvedi, Seth Bergeson (MBA Candidates, Foster School For Business)  

Engineering: Carlo Del Mundo (Xnor.Ai), Kiron Lebeck (Phd Candidate Allen School), Rosa Thomas (AWS), Max Smiley (Microsoft) Arthur Liang, Ben Ihrig


Duration: 3 Months

Class Project In Entrepreneurship Course On Starting A Business And Pitching To Venture Capitalists

Tools Used: InVision, Adobe After Effects, Adobe Premiere, Sketch, Illustrator, Keynote, Google Slides, Instapage

Primary Deliverable: A Slide Deck Pitch, Executive Summary And Financial Outline


InstaUP using AI technology to boost engagement by automatically selecting the most engaging photo, editing and cropping the image, suggesting hashtags, and automatically posting it at the most impactful time.





With recent changes to Instagram’s algorithm, which prioritizes posts based on past engagement, organic reach has decreased significantly. Only an estimated 10% of audiences see content. In this new environment, both influencers and marketers need more tools to increase their engagement. How might we create better tools for influencers and aspiring influencers to optimize engagement?



Understanding the User:
Who is an Influencer?


An influencer is usually defined as over 2K to 10K followers and earns on average  $180 per post

+  Average influencer engagement rate is 5.3% vs. 1.7% overall
+  $1.6B market by 2018
Influencer posts doubled from 2016 to 2017 to 1.5M+



Finding Influencers & Signal


We targeted micro-influencers and brands on Instagram in three verticals:









We created an instagram crawler that generated 1,105 contacts in the Beauty Influencer


+ Targeted a number of these contacts directly
+  Through personal contacts found several influencers within our network to interviews 
+ Using Instapage and target Facebook ad to create a landing page to target our customer segment
+  Scraped and targeted 200 Instagram Influencers from Facebook
(Audience: 13-19, Female & Male)

Platform #2: Facebook Desktop
Audience: 13-19, Female & Male

Platform #3: Facebook, Instagram Audience (Scraped and targeted 200 Influencers)

*Drove 60 Clicks to the InstaUp website for an average Cost per Click of $ 0.34*



Competitive Analysis


To understand the marketplace we looked at a broad range of products.Instagram has 800M users and ad spend will increase by 88% this year to $6.9B. Seventy percent of businesses (15M) use Instagram in their marketing efforts. Ad spend on influencers will increase by 50% this year to $2.4B by 2019.

We reviewed 74 products. Competitors provide similar tools as InstaUP but none of them analyze the 80M photos posted on Instagram per day or integrate the four functions InstaUP provides. Competitors fall into three categories: social media management, analytics tools, and targeted solutions. Many freemium photo editing apps exist, and influencers and marketers often use several of these apps in conjunction with scheduling or analytics apps. InstaUP integrates these core functionalities into one intuitive app while also leveraging the billions of photos on Instagram to refine its algorithm.

Pitch Deck_v5 (1).jpg
Pitch Deck_v5 (1).png







01.  85% influencers measure success from likes/follows and 46% from comments

02.  50% + of influencers spend 30 mins to 3 hours per post

03.   54%  of Instagram users want to be full-time influencers, but aren’t yet



Conducted 21
Semi-Structured Interviews


We spoke to Instagram Users who had between 2,000 to 208,000 followers as well expert interviews with the the former Marketing Manager of OfferUp and Co-founder of TBH (which was sold to Facebook in 2017).



Expert Interviews

"You're selling the idea that machine learning can provide more value, augmenting their photo selection process. But actually, it's the _belief_ of the outcome that attracts people, users don't care how it's done under the hood."

CoFounder of TBH

"I would have loved a tool that could accelerate my Instagram and Facebook acquisition marketing to save us both money and time. I had to manually search through 1,000s of photos to find creative and we were spending a lot of ad money along the way."

Former OfferUp Marketing Manager




User Interviews

“The hardest part is definitely generating new and exciting content. There's SO much out there now and a lot of it is repetitive. I can only look at so many photos of Glossier You before I just get sick of it and don't even want to think about the scent. I know my page is probably more saturated than your average beauty enthusiast who follow beauty accounts, their friends, their favorite restaurant, etc. But it gets hard, and it makes it hard to set yourself apart."

Participant #3

"Aside from testing, writing the blogpost it takes around three hours, on top of that comes about one hour of picture taking and editing and another hour for scheduling promotional tweets/Instagram/Google+ etc., In total between 4-5 hours. Videos take around six hours, accompanying blog post included."

Participant #4

"When it comes to optimizing engagement through Instagram, there are so many more challenges now. As an artist it has been so challenging to release content that is not getting recognized due to the new update"

Participant #9




01. Influencers want to create unique, authentic content that doesn't feel repetitive.

02. Content creation was time intensive and many influencers wanted to spend more time engaging directly with followers or managing administrative tasks, especially with cross platform demands.

03. Algorithm changes at Instagram cause a lot of unrest and confusion. Influencers are not clear why some posts do better than others.





 InstaUP’s algorithm uses convolutional neural networks (CNNs), a popular computer vision technique, for image selection. To evaluate the technical feasibility of CNNs on the influencer space, InstaUp's first CNN model focuses on self-portraits ("selfies") since well-curated datasets for selfies already exist. InstaUP's CNN is trained on 46,836 selfies. An image is scored from 0-100% with scores closer to 100% suggesting higher engagement.

CNNs provide a data-driven foundation for inferring "how" engaging an image will be, but CNNs do not adequately describe "why" an image will be engaging. To close this loop, InstaUP uses Google's Cloud Vision API to further analyze image content based on smile, eye contact, clarity, position, empathy, posture, and movement.



Training the Model


I worked with the engineers to look over the photos selected by the model. I noticed that the algorithm seemed to only select smiling photos which seemed antithetical to the goals of many instagram followers. We retrained the model so it was not as biased towards smiles. I suggested that as the product developed we would to scape actual Instagram images so the photos would be more inline with the users goals on the platform.  

A  selection of selfies used to train the model. These photos were rated highly by the model, good light and clear photos.

A  selection of selfies used to train the model. These photos were rated highly by the model, good light and clear photos.



Working with a Variety of Stakeholders

The group was committed to using Machine Learning, but had no other clear ideas about the design of the product. We considered a very lean product with a quick to market strategy as well as a more robust design with a number of features. We decided to add a few existing core features to the machine learning to automate the whole Instagram posting process. Auto-scheduling and hashtag optimization are existing features in a few available products and the engineers felt they would be able to recreate these features with a little bit of time. Our key insight was users expressed fatigue managing a range of apps.  


From user interviews I hypothesized that people wanted a multi-purpose app that could offer a variety of professional quality experiences--like a photo studio app on their phone to streamline posting to instagram. In retrospect, because this product offered a new technology where there was a rush to market, it might have made more sense to launch InstaUp as a single feature product and add features slowly. 


Building the Product:

3 Core Features


Automatically Selects


the most impactful photograph with custom filters and auto-filters based on the Machine Learning


Hashtag optimization

offers a list of popular hashtags currently trending on Instagram



recommends the perfect time to initiate your post


Building the Product:



Early Usability Testing:

Gathering User Feedback


With the engineering team we determined that the features were technically feasible, but could take some time to create. We also were able to follow up with two influencers  for a preliminary remote usability test. One user was very excited about the features, but struggled to maneuver the app. We made several core changes:

01. Create a seamless integration where the best photo is selected in your camera role and you don't need to upload anything

02. Filters were added based on user feedback

03.  Words like "analysis" were deleted and language was replaced with photography terms and icons

04. Preview was deemed irrelevant because feature exists in Instagram


Designing with Automation:

How to Help the User Trust the Algorithm

An interesting challenge with automating a complex decision-making task, was would users trust an algorithm. What did users understand about Machine learning and what did they need to know in order to rely on the feedback of the app. I focused on two feedback mechanism

01. Assuming that if the algorithm was successful users would see the feedback in the product’s analytics and Instagram comments and posts

02. Using the model of Google’s Cloud Vision API I designed feedback which included both somewhat subjective ideas like “joy” and “eye contact” which was a more objective. Though I was not able to sufficiently test the idea, my hope was that this would help explain new users why they should trust the algorithm. Seeing this information was optional and could be avoided once users felt secure.

03. Offering manual customizations so users could override the algorithm and feel they still had control over their images, even if the algorithm new better.



Selling to Venture Capitalists



Making the Case for Design


Working with a combination of engineers and business students, on a short time frame, meant there was often a tendency to ignore the design process. It would have been great to have more time with our users to understand their needs. Often my colleagues were satisfied with customer validation and did not necessarily want to did further. The engineers were busy testing the validity of the system. At the end of the project the engineers came to me and mentioned it would have been better to wait on the engineering end until the product was designed. I took this a big win! It was great learning from everyone and building up some business accuity will help me better develop viable products.