Share This Post


Why “Hyper-local” for groceries needs to move quickly…beyond just the App

(Image courtesy: The telegraph)

One of the leading dailies ran an article some weeks back saying- Groceries is slated to become the 3rd largest category online after Electronics and Fashion. And yes, the number of online grocery start-ups and the quantum of risk capital flowing in…seem to suggest an explosion for online groceries as a segment.

Looking at the consumer behavior, online grocery does seem to add value to the consumer. Put very crudely, grocery shopping is a chore- to be performed from hyper markets 1-2 times a month and from neighborhood kiranas many times a month for emergency or fresh purchases.  Moreover 70-80% of the items are common month-on-month. I fail to understand the “high” that someone might get from picking the same pack of Aashirvaad atta or Kellogg’s cornflakes month-on-month for years! And I find it somewhat appalling that these shopping trips are (forcefully) called family outings!!

I believed in the above so strongly that I ran my own online grocery store ( for 3.5 years (before shutting it down as we were unable to raise funding),

(Disclaimer- Given this, whatever I write is based on hands-on experience and not plain hypotheses. However, given that I latched on to a certain model, there could be an element of bias as well).

There are many models that are being tried right from online grocer having own warehouses (and even own bakeries), to using cash & carry stores/hypermarkets as sourcing points, to a completely hyper-local model where the stock and infrastructure of local kirana is used for deliveries in the neighbourhood. Independent of these models from a customer point of view there are 5 important values that need to be delivered- Range, Fill rates, Quality, Pricing & Logistics.

Range: The range of items or SKUs is very critical. Local Kirana stores have SKUs in the range of 3000-7000 depending on size of the store. Corresponding figures for Supermarkets and Hypermarkets are 9,000-12,000 and 18000+ respectively (excluding apparels & electronics). Moreover, the width differs. Many Local kiranas do not store fresh produce and even many super markets do not store items like fresh meat/fish. Range is one of the factors why we find supermarkets/hyper markets crowded on weekends/holidays- so crowded, in fact, that it takes 30-45 minutes just to complete the billing.

Fill rates: You ordered 50 items, and you got exactly the same 50 items- that will be a fill rate of 100% (I reiterate the words EXACTLY the SAME). Fill rates of 70-80% are considered very good from a Brick and Mortar point of view. Getting a fill rate of 100% is practically impossible unless the customer is shown a pre-selected inventory and asked to choose from the same set. There are 2 possible issues with the fill rates:  

  • To set fill rates high, the SKU range can become very limited.
  • A customer ordered 25 items from a store for home delivery but was delivered only 20 items. The immediate thought is if one has to go out for buying 5 items, s/he might have bought the remaining 20 as well. There are players currently who are offering money back/ reward points as a compensation for lower fill rates. While this could address the immediate customer dissatisfaction and a possible negative word of mouth, it clearly fails to address customer value from point of view of ordering groceries online. And, more importantly has a huge potential to impact the repeat rates (unless fill rates are so pathetic that the money back policy can become an “alternate income” for the customer!). Substitution fill rates like replacing a Red Gel toothpaste with a Blue Gel toothpaste can help in increasing the overall fill rates but there are limits to customer acceptance for substitutions.

Quality: Apart from the Branded food items, quality of Fresh produce is very subjective and requires management of a very complex supply chain. That is one of the reasons why you see very few kiranas stocking Fruits & vegetables and even fewer supermarkets storing Fresh meat/fish etc. It could be counter-productive to rely on judgment of quality of different order pickers from different stores in different localities unless these are fairly well controlled at source.For pulses, rice and grains, customer preferences do get built over a period of time. Unless, a customer knows that the quality of rice (as per her preference) is going to be more or less consistent, generating customer stickiness is going to be difficult. Delivering from X local store today and Y local store tomorrow is prone to variations in quality.

Pricing: The customer might have just watched a movie in a gold class and spent Rs 100 on caramel popcorn, but grocery as a category surprisingly brings out price-consciousness the most! The moment a customer steps in to a grocery store, pricing becomes important- “this market provides Rs 5 lesser discount for my favorite brand of atta”. And pricing is important especially for the staples. Consumers are usually in tune with the prices of about 20-25 items (e.g. staples, onion, potatoes, tomatoes and some other high frequency purchase items).The pricing power in groceries is through bulk purchasing and it is very apparent that the kind of pricing power that a hyper market will have will be higher than the others.

Logistics: The last mile delivery is a make or break in any organization that has delivery as a key component. In this the vicinity of delivery becomes extremely critical. Kiranas have the biggest advantage here as the handyman in the store doubles up a delivery boy to deliver even a pack of 6 eggs and bread. Try delivering this by sending a delivery boy on a 2-wheeler and even the cost of fuel is not going to get recovered. Order sizes become very critical in this regard.

Where Hyper-local stands in terms of the above parameters

Hyper –local solves the last-mile problem perfectly well. However, it runs the risk of falling woefully inadequate on the other four parameters (unless the foundation of hyper-local is a source that addresses all these).

Crystal Gazing

The way I see a hyper-local model evolving in groceries is through one of the following ways:

  1. There is a backward integration where the online grocer is not just an app provider but also facilitates the procurement part of the supply chain. A private label brand for staples, fresh non-veg and even fruits & vegetables which is in-turn supplied to the local Kiranas can address most of the above mentioned issues. Moreover, this approach can address the issue of lower margins and also provide a clear differentiator to the online grocer. 
  2. The online grocer ties-up with hypermarkets and uses them as sourcing points.  
  3. Large Brick & Mortar retail chains use their existing formats as hyper-local sourcing points and add online as part of the Omni-channel strategy. This could help in “sweating” a fixed asset more and at the same time help in rationalizing the store counts ultimately having a positive impact on profitability.

Building an online grocery model purely on the strength of an aggregation app has its serious limitations!

Twitter handle: @ssjunnarkar

Originally posted here.



Share This Post


  1. Hi Sushant,

    To summarize what you said, are you suggesting a grocery aggregation service? As in I order 5 things, based on my location you know the nearby kirana stores, the company should order the items in the closeby shops and have it delivered to my doorstep. Kind of like the food delivery startups like Swiggy?

  2. Hi Sridhar In my view the current focus of most of the grocery aggregation ventures is that they are focusing primarily on the last mile delivery solutions. Other important parameters of Range, Fill rate, Quality & Pricing need equal (if not more) focus. Thanks

  3. Hi Sushat,

    Well Said!!

    I’m also into grocery, fresh fruits and vegetable business & have successfully completed  2 months.
    For grocery we are working on hyper-local model and for fruits and vegetables we are doing self-procurement from day 1.
    We want to make changes in this model and would love to make it more vibrant. We are releasing version 2 which none of the biggies have attempted to indulge in.




  4. Hi Sunil,

    Glad to hear that. I will be keen to follow your venture and the changes in your model. Do keep me updated as and when you make the changes.

    All the best.


Comments are now closed for this post.

Lost Password