Designing an Off-Grid Microgrid 101
Episode 4-b of ‘How to monetize a microgrid?”: building a grid from the bottom up
After introducing the off-grid energy space in the previous article (read here if you haven’t yet), let’s dive deeper now.
While I had written a whole article listing up my current understanding of the diverse topologies, business models and pricing schemes of off-grid microgrids, my dear co-author Vicky posed a new challenge: you should turn this into a 101 for building a microgrid!
I realized that doing so would turn my article upside down as I, like many others, started with topology and tech classifications first, followed up with business models, and left customer needs to the conclusion.
Let’s go through it as if we were to design our own grid, then the order should be something like this:
The customer always comes first
This looks obvious, but so many people normally skip this part. First of all, customers don’t want electricity, they want light, TV, a fridge, or a pump. Many companies, driven by impact KPIs, have overestimated the customers’ energy needs (or their ability to grow their energy needs) which lead to over-dimensioned grids that couldn’t generate the expected revenues.
Furthermore, understanding cultural differences is crucial. The grid infrastructure is different, population density is different, payment behaviours are different.
- Who is your customer and what are the customers’ needs and wishes?
- Do they want to own their system? Or perhaps they already have one?
- Would they prefer to pay as per their energy consumption or through a fixed tariff per day? Which payment system is flexible enough and yet still has the needed predictability to work for both parties? (this depends a lot on the type of income, whether it is constant or not)
- What amount and for what purpose do they need energy? Do you need to provide appliances?
Understand regulatory restrictions
Once you have got an idea of what the customer wants, double-check whether regulations allow it or if you have to make a compromise. Country policies vary greatly and what makes sense in one country does not in another. For example in Bangladesh anyone can generate and sell electricity without a license (off-grid) if the generation capacity is below a certain threshold, however, this is not the case in Cambodia, where everyone selling electricity needs a licence.
- What are the regulations for selling electricity? Is it allowed? Are tariffs imposed?
- Are subsidies available?
- What are the plans for grid expansion and would it impact the model?
Business model
Depending on what the customer wants (and whether it is allowed by regulations), I tend to think there are generally four business models you can choose from depending on two major questions — also shown in the matrix below:
- Whether customers wish to own the system (pay to own) or already own it.
- Whether paying a fixed fee per day or a variable, consumption-based fee makes more sense for them (the latter requires metering technology).
The answer to the previous two points are very much country dependent, but having clarity on the differences should help you choose the business model. There is no one best solution, it really depends on the environment, which is why the first two steps are crucial.
Let’s look at each more in detail. Here is a summary diagram with examples.
- Energy as a Product
The SHS industry started off by selling a product outright and later introduced installment payments (PAYG) to address a wider customer base. They were the first off-grid solutions available from companies like M-Kopa or BBOXX who built everything by themselves (i.e., vertically integrated) and obtained big investments to expand faster. This was mostly enabled by widespread adoption of mobile money. In countries like India where this wasn’t the case, money collectors would do that job (Simpa acquired by Engie).
People became familiar with this very simple model of paying by the day, regardless of whether the sun was shining or not and independent of energy consumption. There was no need for metering in this scheme. This is a radically different business model from the developed world, where people are used to paying only for what they consume. However, in off-grid areas, people typically have little choice and so they often end up paying a high price for a small amount of energy. Further to that, because the energy is limited, having highly efficient appliances matters. The mindset is different, it is not that customers are trying to avoid a higher bill, rather they simply realise that they can watch 4x more TV if their TV set consumes 50W instead of 200W. Over time, appliances have become highly efficient, often powered with direct current. Some 21” colour TVs consume a mere 6.5W!
2. Leasing Energy
This is very similar to the previous model except that the customers never own the energy system or appliances, they simply lease them for as long as they need. As you pay by the day it does not require metering. A leasing model allows replacing or upgrading as customers’ needs change (sharing economy). For the company it is logistically more challenging than the ownership model but, if successful, allows companies to hang onto their customers for longer. Nevertheless, in emerging countries most customers still seem to prefer ownership if they can afford it (though that may change once the advantages of leasing become clear and customers have sufficient trust in leasing models).
This model is common in India (implemented by companies like Zola Electric — formerly Offgrid Electric) because of its simplicity and because it has even further reach than Energy-as-a-product models?, providing for those who cannot afford their own SHS. Also, some Energy-as-a-Product type companies such as BBOX are moving to some form of service-like model, as can be seen below.
3. Energy as a Commodity
In microgrids, where assets are not owned by users but by the company or a third party, electricity is allocated and billed to users as a commodity, ie. per kWh. This is by far the most common business model for microgrids. In some cases, billing is not done purely in kWh but by a proxy, such as the number of battery recharges, in order to avoid “selling energy” and thus circumventing licensing requirements. Sometimes tariff schemes similar to mobile phone packages are used: pay $5 for up to 4kWh/month, $8 for up to 8kWh. Not only are such energy packages easy to explain but they also ensure a minimum demand (up to 4kWh for example).
In order to optimize the available energy, the grid operator will try to balance the loads and generations, potentially by giving different prices to different types of users.
Such models are widely used by microgrid operators (Winch, Sunkofa, Devergy, Okra, …), but they bear an often-overlooked risk: whenever you bill on consumption — in this case, a consumption that does not exist at the time of planning — you may be very tempted to overestimate it, as it makes your business model and story work attractively! Beware of doing so, this has resulted in many microgrids that are over-dimensioned and have never reached profitability.
4. Energy Trading
Moving one step further, we arrive at the P2P platforms where people can buy and sell energy, not just virtually as discussed in my previous blog post (LINK), but physically. This makes sense in particular when people already own an SHS and want to monetize their excess energy. Energy trading therefore allows distributed ownership and also encourages entrepreneurship. These kinds of models have received a lot of attention as the ultimate future of the electricity market in developed countries, but are still rather far off due to stringent regulations and backward compatibility with the existing systems. This is why I believe we will see these models evolve earlier and faster in emerging countries where infrastructure is still being built and regulations are often less limiting. SOLshare in Bangladesh has pioneered one model.
However, the technical and business challenges can be daunting and should not be overlooked. It is highly critical to check your assumptions about the customer: do people have sufficient energy to meet their needs and want to sell excess energy? How can a sufficient margin be created for both the microgrid operator as well as for the end-users to participate? This microgrid format requires a multi-layer business model with clever but easy to understand pricing — an open challenge.
Microgrid Topologies
Bringing it all together — Microgrid topologies can be grouped into SHS, centralised microgrids and distributed microgrids. So how do you choose which topology works best?
Actually, depending on your business model, one or two topologies will be most suited. This is shown in the completed matrix:
It is worth noting that microgrids are often represented by their topologies, which is the physical arrangement of the renewable generation (mostly PV) and storage, despite fundamentally different business models.
- Independent Solar Home Systems (SHS = a kit of panel, battery and appliances), enabled by clever financing solutions and new, cheaper technologies, SHS have been the front runners in off-grid solutions over the last decade. They are simple in terms of technology and business model as they can be sold as a product. However, the generation and storage of one SHS is limited, and thus, it results in wasted energy whenever the battery is full, but the panel is still generating during good weather.
- Centralized Microgrids are a more top-down approach, they dispatch centrally generated energy to multiple consumers: a microgrid developer would plan and build the generation and storage and connect houses. Central resources are usually easier to manage and maintain, but have an issue of financing and ownership. Contrary to SHS, which are largely unregulated, microgrids are usually regulated requiring licenses, or imposing tariffs (very much country dependent). The process of applying for such licenses can be very burdensome and time-consuming, making small microgrids financially unviable and difficult to scale. Hopefully, portfolio applications for microgrids will become more common and enable developers to reduce expenses.
- Distributed Microgrids are another approach to interconnect SHS and create a decentralized network. This has the advantage of scaling more organically. New systems can easily be added as, where and when the demand grows. It also opens the possibility of leveraging pre-existing SHS and pooling them in a more efficient way. Customers can both buy and sell energy. However, the regulatory and other hurdles can be daunting.
Hybrids: In between, you might have hybrids with some central and some distributed assets, but because of the technical and business complexity these examples are scarce. What does exist is hybrids with only partially connected systems (one or more SHS may be too far apart and not interconnected at first).
Historically, the market has evolved from least complex to most complex, although I believe all three approaches will co-exist as they each have their respective strengths and weaknesses. In the future, new topologies will arise when those microgrids get interconnected to the main grid (I won’t discuss that here).
Offgrid Energy Pricing
Now the big decisions are taken. But a tough one lies ahead:
How do you price energy in an off-grid context when you cannot refer to an energy market price, or even a fuel price?
No metering (price per day)
For SHS with PAYG, the most common attempt is simply to calculate the Levelized Cost Of Electricity (LCOE): take all the costs and spread them over the payback time (daily, weekly, or monthly installments). For leasing services, you spread the costs over the lifetime. Naturally when ownership is not transferred prices are lower, but the payback time is longer. The difficulty here is how to incorporate financing cost, default rates (customers who stop paying or pay late), and operational costs — which can make up more than half of the retail price.
Metering (price per kWh)
When billing is linked to the energy consumed, a similar LCOE calculation applied to microgrids can be done. The main difference is that the revenues are not by the day but by kWh consumed (not generated!). While generation is relatively easy to forecast, demand is hard to predict, especially for people who have never had electricity before. As mentioned earlier, overestimating demand has been a common issue with off-grid microgrids.
Since more and cheaper energy is available during the day, it would make a lot of sense to change prices depending on energy availability. However in practice, the simplest and by far the most common model is to impose a constant kWh tariff, sometimes combined with a monthly fee. Both prepaid and postpaid exist and are often country/culture-dependent.
In the case of a P2P Trading Platform, the difficulty is double: both buy price and sell price must be determined. This is an article in and of itself and I might write more about this after concluding an experiment with Solshare who have implemented such a scheme. Meanwhile, these rules must be followed for it to work:
- Buy price: It must be cheaper to buy energy from neighbours than to buy an SHS yourself.
- Sell price: buying an additional SHS and selling all energy must be profitable.
Regulated tariffs
In many countries, kWh-energy prices are imposed or must be agreed upon with the regulators. This is highly country-specific and one of the main reasons why microgrids are spreading in some countries (e.g., Kenya, Rwanda, Tanzania, India, and the Philippines), but not others (e.g., Mozambique and Ethiopia). The availability of government subsidies and blended finance solutions are the other reasons. Both of these factors change quickly, which can be a risk or an opportunity.
Tadaaa!
Now you got your microgrid! Go and build it!
Joking aside, it is extremely hard to build a microgrid, to answer the above questions and to figure out all the details. And if you think it’s finally over, we have not even broached some topics, such as whether you would go AC or DC, or which communication infrastructure you could use (maybe in a future article, or feel free to send me questions).
Most likely when you get to the end you will need to start over and make several iterations. It may be just as hard for your investors to appreciate this. To quote a good friend and CEO of an off-grid microgrid company:
“Often the reality is that you won’t have enough time and resources and investor backing to go methodically over this process, so founders be aware that you may end up jumping from 1 (customer) to 4 (topology), and then from 2 (regulation) to 3 (business model) etc. It could more be along the lines of 1–5–1–3–4–2–5…!”
Nevertheless, I hope this guide has helped to look at it in a more structured way and that it became clear that everything depends on the environment — especially the customers and regulations. And perhaps a few of you feel inspired to get out there and take on the challenge.
Closing remarks
Many models are being tried out, and probably most of these projects will fade away. In reality, many of these projects were driven by social objectives of providing energy access — but when success is measured with the size or number of connections, it often leads to overly optimistic demand expectations which in turn results in over-dimensioned microgrids that cannot become financially sustainable, and therefore, also not scalable. Too many Proof of Concepts (PoCs) or pilots couldn’t live up to their promises, evaporating considerable amounts of money and also eroding enthusiasm for further investments?
That does not mean that grants or subsidies or investments are always bad. On the contrary, we are in need of technical exploration projects and visionary new business models trials. However, they should be seen as such: trials to fail fast and fail cheap (without shining a negative light on the entire industry).
Meanwhile, for microgrids with viable unit economics and projections, blended finance solutions are great for bridging the gap to scaling. Though private companies can reach modest returns even without subsidies (especially by diversifying their revenue streams), growth and scaling can be boosted with subsidies. Indeed, since private microgrid developers provide valuable infrastructure, they should be valued as such and receive similar subsidies for electrification as the main grid operator does. This should be the case particularly when they are more cost-effective, more environmentally-friendly and faster to deploy than conventional approaches. But this only works with realistic expectations and evaluation criteria, and a clear path to long-term sustainability.
Fortunately, as the market matures, viable business models and solid unit economics are becoming a more important requirement to obtain investment, and microgrids are gradually being considered as the most viable infrastructure for off-grid areas. With solar and battery prices decreasing more rapidly than any prediction anticipated [ref], market trends are in favour of microgrids and I firmly believe a bright future will be ahead.
[Countless times authored by Annette, spiced up by Vicky. We work together to choose the best content and elevate each other’s work. More on us and our approach here]
*** Update ***
Here are the quick links to the other articles in the series:
- A Quick Overview of Microgrids (and How to Monetize Them) — Episode 1 of “How to Monetize a Microgrid?” : Introduction the world of microgrids
- How to Monetize Microgrids Running on the National Grid? — Episode 2 of ‘How to Monetize a Microgrid?” : on-grid microgrids on existing infrastructure
- Why Pay for Private Infrastructure with an On-Grid Microgrid? — Episode 3 of “How to Monetize a Microgrid?” : on-grid microgrids with private infrastructure
- A Little Primer for Off-Grid Microgrids — Episode 4a of “How to monetize a microgrid?” building a grid from the bottom-up
- (Current one) Designing an Off-Grid Microgrid 101 — Episode 4b of ‘How to monetize a microgrid?”: building a grid from the bottom up
