MySales Labs

What they say… and what they don’t 🤫
(A reality check about AI forecasting in retail)

In retail tech, many claims the same thing:

“Our platform uses AI to forecast at SKU–store level with price, promotions, weather, seasonality, cannibalisation and more.”

Sounds perfect. But here’s what they don’t tell you:

Most platforms do not actually learn these effects at SKU–store level.
Instead, they learn one global or region-level coefficient (price elasticity, promo uplift, weather sensitivity, etc.) and then simply:

👉 apply it top-down,
👉 weight it by store data,
👉 and call the result “SKU–store forecasting.”

The forecast number becomes store-specific —
but the behavior behind the number is not. And that’s a big problem!

Price sensitivity differs dramatically by store type and neighborhood
Promo uplift depends on local habits & competitor intensity
Weather impact is extremely location-specific
Cannibalisation is always local — driven by assortment, shelf proximity, and shopper missions

You can’t solve these with a global elasticity and some multipliers.
You can’t “allocate down” true customer behavior.

At MySales Labs, we chose the harder but correct path:

👉 True bottom-up causal learning — per SKU, per store, per factor.

That means the model actually understands how customers in that store react to: price changes, promotions, weather, assortment shifts, cannibalisation, competition, local events

No shortcuts. No global averages. No fake granularity.

If you want to instantly separate real forecasting platforms from marketing fiction, ask this:
“Do you learn causal impacts per SKU–store, or do you apply top-level coefficients and allocate them down?
Can you show how?”

This question ends the storytelling very quickly. 😎

🚀 Forecasting retail demand today means analyzing millions of data points — every single day

Modern retail demand is shaped by hundreds of influencing factors at once:
seasonality, price, promotions, cannibalisation within the assortment, competitors’ prices and promos, weather, local events, store formats, range changes — all interacting differently for every SKU in every store.

Now multiply this by millions of SKU–store combinations, recalculated daily.
This level of analysis is simply impossible to do manually.
That’s where AI-powered forecasting in MySales comes in.

MySales models demand at the most granular level: per SKU, per store, per day.

AI automatically selects the best forecasting approach for each SKU–store pair, combining statistical models, machine learning, and neural networks where demand is complex or volatile.

What does this change operationally?
🔹 Store-level manual ordering is fully replaced
🔹 Forecasts and orders are recalculated daily, without human bias
🔹 Cannibalisation and competitive pressure are built into demand models
🔹 Ordering becomes consistent, scalable, and explainable
🔹 Store teams are freed from routine calculations

This is not incremental improvement.
It’s a shift from human-scale planning to machine-scale decision-making.

People don’t disappear — but manual ordering does.

And that’s how retailers unlock:
📈 higher on-shelf availability
📉 lower spoilage
💰 less capital locked in stock
⚙️ truly centralized control

💡 Why LightGBM Doesn’t Work for Retail Forecasting — and Why MySales Does

Some of our competitors tout their use of LightGBM for sales prediction. It sounds impressive — until you dig deeper into how retail really works.

Here’s why LightGBM falls short in retail forecasting:
1️⃣ No built-in seasonality — you have to manually add it, which is prone to error.
2️⃣ Overfitting hazard — it can’t distinguish between Christmas promotions and holiday seasonality.
3️⃣ False correlations — if you raise prices during high season, it might mistakenly think higher prices cause higher sales. 🤦‍♂️

Since 2014, MySales has focused on interpretable and reliable demand forecasting for retail and FMCG. We’ve built our own proprietary algorithms specifically for this domain — ones that understand seasonality, weather, price elasticity, promotional uplift, and even market trends together.

✅ We don’t just apply a generic open-source tool — we’ve engineered logic-based, trainable models that work in complex, real-world retail environments.
✅ Our forecasts are transparent — you can always see which factors drive the prediction.
✅ It took 10 years of real-world testing by clients to make MySales what it is today.

And the result?
We consistently outperform shallow plug-and-play solutions.
Simps don’t survive here.
Retail is tough. Forecasting should be smarter: https://chatgpt.com/share/68b59c21-1f90-800a-9b9f-bcfe36b53fc4

🔗 Learn how MySales can drive your revenue and margin: https://mysales-labs.com

MySales Labs Powerful tool for retail and businesses that sell at least one SKU per store per week. Optimise your stock, automate your ordering and accurately forecast your promo activities