Aryma Labs

How we signed 5 big ticket MMM projects this year. This year, till date we have done 11 pilots. - 5 clients signed up for a full fledged MMM project. - 4 clients said they will get back by early 2025. - 2 clients said they don't think they have the need or are not MMM ready yet. 📌 So what is a comprehensive MMM pilot? A comprehensive MMM pilot involves doing a full fledged MMM project but at 50-60% of the full fledged project cost. How this helps clients 📌 Answering - Are we MMM ready? A comprehensive pilot project starts off with detailed evaluation of client's data readiness and then iteratively proceeding ahead with building models, setting up model calibration and finally using MMM insights. Contrary to common misconceptions, MMM isn't restricted to large marketing budgets. We've tailored MMM solutions for influencers with budgets in range $300k - $1Mn, as well as for established companies with budgets ranging from $10-100 Mn. Statistically there are no constraints that says there has to be 'this high marketing spends' to do MMM. 📌Stakeholder Buy-in Most MMM projects gets stalled because of lack of stakeholder buy-in. A comprehensive pilot helps the client to showcase the results to the upper management and thereby prove its utility. 📌Cost saving: A comprehensive pilot helps client save thousands of dollars and at the same time helps them make the Go / No-Go decision on full fledged MMM. 📌What is in it for us? MMM sales life cycle is perhaps one of the longest. On average, it can take anywhere between 1-6 months to close the deal. At Aryma Labs, we noticed that the time taken from initial demo/service request to contract signing reduced significantly to around 3-4 months once a comprehensive pilot is undertaken in between. A faster MMM sales cycle not only benefits us but also the clients, since quicker MMM also means quicker insights and on ground action. 🛑 What to avoid. Many think that throwing their data at some product which is a wrapper on top of open source MMM library is akin to doing a 'Pilot'. Sorry but this is another 'wasting-dollars-but-getting-no-insights' program. An initiative like this gives no answers on data readiness or the process readiness. Rather it creates a false impression that one can do MMM easily. MMM is also more about statistical accuracy and interpretation. Most of the bespoke changes required at the modeling level, are hardly doable with such wrapper products. Not to mention the inaccurate results one could easily end up with. Getting a open source MMM library to work requires more work. Paradoxically one needs more data scientists as opposed to the promise of not needing one. 🎯 Bottomline - A comprehensive pilot is a win-win deal for both clients & vendors alike. If you want to be MMM ready for 2025, now is the right time to start. Reach out to us.

Ask MMMGPT - Episode 2 (Why is Multicollinearity a big problem in MMM?) Hello Friends, We are back with this week's popular question in MMMGPT. "Why is Multicollinearity a big problem in MMM?" As promised I provide a detailed explanation. Check out the complete video. If you want to learn about MMM accurately, subscribe to MMMGPT. Link is here: https://lnkd.in/gGtdVaN3

An Independent Review of MMMGPT. Thank you Dr. Wei Hutchinson, PhD for your positive remarks as well as the suggestions for new features. Want to try MMMGPT, Pls sign up here : https://lnkd.in/dHu6xpBe

We are excited to have you join our team Keerthana K V. Let's do some great work together 🚀

We are excited to have you join our team Shreyash Adappanavar. Welcome aboard !! Let's do some cool work 😎

Introducing Ask MMMGPT Live, a fantastic new initiative by the founders of Aryma Labs. This showcases the power of this cutting-edge tool by addressing popular questions asked of the GPT and sharing the responses in short videos. This week, the spotlight is on two trending questions: - Is there an ideal R squared value? - How do you determine the accuracy of your model? with Venkat demonstrating the tool. If you're curious about how the GPT performs or have questions you'd like to see answered, signup here : https://lnkd.in/duutkgkh, comment below, or message me or the team at Aryma Labs.

Launching Ask MMMGPT Series 🚀 Hello Folks, Thank you for your overwhelming response for our MMMGPT last week. Many of you have subscribed to our product already. Our day is made when we users from different countries praise our product and tell us that they have learnt something intuitive in MMM. This week we are starting a new initiative called "Ask MMMGPT". Every week, we will take 2 or more popular questions that are asked by our users and highlight it here. We will also provide our commentary or explanation going forward. If you have a particular question or questions to ask to MMMGPT, do reply in comments. We will try to include it in our next series. Or You could sign up to MMMGPT and give it a spin 😎. Sign up here : https://lnkd.in/gGtdVaN3

Is your Budget Optimization linked to your MMM model? When it comes to MMM, a lot is written about MMM model specification, calibration, validation, causality and incrementality testing. However one thing that is often overlooked is the budget optimization. Budget optimization is one of the most used outputs by clients in the MMM lifecycle. Quite literally million dollar decisions are taken based on it. But little is written about whether the budget optimizations are accurate or not. How to test its accuracy, how do you know you have ended up with a bad optimization etc. are never asked nor answered. So here are some tips: 🎯 Always compare your budget optimization with saturation curves. In an unconstrained scenario, budget optimization tends to allocate spends from low ROI to high ROI mediums. This mimics the saturation curve behavior where the low ROI mediums have a curve that diminishes faster as compared to high ROI mediums. Hence drawing a parallel between the behavior of optimization and saturation curve will tell your if your optimization is in the right direction or not. 🎯 Check the ceiling and floor of the constraints. Be it excel solver, any open source budget optimizer or your own in-house optimizer, there exists an option to levy constraints. For e.g. up to 20% minimum spends can be taken out from a particular medium. This is known as the floor value. An accurate optimizer will not always take away 20% spends from that medium but rather it could take out something like 12% or 15%. If your optimizer by default always takes out the exact floor or ceiling value, then the optimizer is not really accurate. 🎯 How do you ensure accuracy of Budget Optimizer? At Aryma Labs, we have a proprietary technique that tallies the results of the budget optimizer with that of the saturation curves. We have a metric that lets us know how accurate or inaccurate the optimization results are. Further our technique also checks the floor and ceiling of the budget constraints. We check if the budget optimizer by default is always hitting the floor and ceiling value instead of choosing values in the range of floor or ceiling.   These checks ensure that our Budget optimization is accurate too. Hence not only are our MMM models well calibrated and validated but our budget optimizations are also well calibrated and validated.

How Google's announcement to not deprecate third party cookies impacted us positively 🚀. In information theory, there is a concept called Entropy. If you want a good refresher, I would highly recommend the article by Naoki (Link in comments). Basically, Low Entropy means receiving very predictable information. High entropy means receiving very unpredictable information (some also define it as an element of surprise). Earlier, there was very high entropy in the marketing measurement and attribution space because Google kept dilly dallying over cookie deprecation. Nobody was exactly sure when it would happen. And even if they went through with it, people had a lot of questions over their privacy sandbox. Would it be accurate? How trustworthy it would be ? etc. were the questions posed by the marketers and advertisers. But On July 22nd,2024 the high entropy turned it to low entropy. Meaning a lot of uncertainty was removed. Google announced that it won't deprecate 3rd party cookies but rather give the option directly to the users to opt-in or not (much like Apple's ATT solution). This signal was enough for marketers and advertisers to really double down on privacy proof marketing measurement methods for 2025 and beyond. We at Aryma Labs saw a significant jump in the number of calls for MMM services/ demos in days just after the Google announcement. But we wanted to be more scientific about this proclamation. 📌 The ITSA method We have been implementing ITSA (Interrupted Time Series Analysis) method for our clients to help them understand the effect of special campaign launches or product launches on the KPI. We decided to apply ITSA to our own data set. This data set comprises of number of calls for MMM services/Demo requests from our website starting from Jan 1st to Aug 16th of this year. Overall out of 8 months, there were 32 days where we got at least 1 service / demo request (pls refer the image). But the most interesting thing to observe is that there has been an increase in flurry of activities post July 22nd. The Highlights of ITSA analysis are as follows (we will be publishing a detailed research paper on this soon). ▪ On an average, before intervention (July 22nd event), we had approximately 1 service request/demo request. ▪ Post intervention, we see that we are getting additional 1 more request !! We also ran a BSTS model to compare, and found similar results as ITSA. The estimate of causal effect of intervention is 0.91. An additional approx. 1 request on average is observed in the post-intervention period compared to the pre-period. We are sure we are not the only ones to experience this uplift in service/demo requests for MMM. Our rivals too could be experiencing the same. As they say, the boat rises with the tide. The tide of privacy proof marketing measurement and attribution is on the rise.

I am Excited to share the complete case study on our recent versatile MMM project - Modeling Crowd Attendance through MMM !! Earlier last week, I had briefly talked about this case study and how we at Aryma Labs not only excel at OG MMM (CPG/FMCG) but also with implementation of MMM in other domains like Sports, QSR, Finance, Gaming, E-commerce and Pharma. I am really proud of the team for the work done on this project as it required out of the box thinking not only statistically but also from business problem stand point. Kudos Tannishtha Sen Febin P Babu Soham Giri and Venkat Raman. Also thanks Soham (Our in house Football Aficionado ⚽) for the beautiful work on the case study.