Business Model Canvas beyond startups – Part 3: Back-end

How can we use Business Model Canvas beyond its initial intended context of commercial startups? In particular, how best can we use it to explore the ‘back-end’ of the business-models – the internal and supplier-facing parts – for non-profit organisations and government, and for existing commercial businesses such as large corporations?

Business Model Canvas [BMC] is very popular in the startup community, and with good reason, because it works very well for that type of context – no doubt about that. Yet its beguiling simplicity and power there depend in turn on a swathe of often-undeclared assertions and assumptions – ones that often don’t work outside of a startup context. To use BMC for non-profits and large organisations, we need to surface those assumptions, assess their implications in each case, and adjust as necessary. Which, following on from the previous introduction-post, and the post on the business-model ‘front-end’, is what I’ll aim to explore here.

(See also the posts by Paul Hobcraft, referenced in that introduction-post, that formed the start-point for this exploration. For enterprise-architects, I’d also strongly suggest reading Nick Malik‘s post ‘The EA Metamodel Behind The Business Model Generation‘ – it provides a strong theoretical overview of the internal structure for BMC and how it links to business-strategy and business-motivation.)

Also, the BMC is perhaps a bit over-focused on the customer-facing ‘front-end’ of the business-model: by comparison, the ‘back-end’ – where all the work is done to make the ‘front-end’ happen – often seems barely skimmed-over, merely summarised. And yet a business-model can be made or broken on the ‘back-end’ just as easily as on the ‘front-end’: we do need to maintain some sense of symmetry and balance here.

So to elicit some of those assumptions, and build a more balanced view across the whole business-model, let’s explore the ‘back-end’ of the BMC, block by block:

I’ll go through the structure of the ‘back-end’ of the BMC in the same sequence as in the Business Model Generation book, starting with Key Resources.

6: Key Resources

The Key Resources block is shown on the centre-left of the Business Model Canvas. The book’s summary on p.34 for this block is that it:

Describes the most important assets required to make a business-model work.

The book categorises assets as follows:

  • physical - material, facilities, equipment
  • intellectual - brand, IP, partnerships, data
  • human - people and skills
  • financial - cash, line of credit, loans

Myself, I’m glad to see that this is much wider than many IT-oriented ‘enterprise’-architects would acknowledge: the latter often notice only data, applications and computer-based technology. The one definite oddity here is that brand is classed as ‘intellectual’: often the whole point of a brand is how it connects to human emotion (see Davidé Casali’s distinctions between logo, mark and brand for more detail on that).

In my own work I’ve avoided separating out ‘financial’ as a distinct asset category – it’s arguably more a form of data about purported ‘rights’ – but I’d agree that doing so is what most business-folk would expect. I’m concerned, though, that there’s no mention of ‘aspirational’-type assets such as reputation or trust – because without those assets in place, the business-model has no chance even to get started.

In any case, the list of ‘assets’ here must include relational (person-to-person) links to customer, supplier and employee – otherwise there’s no access to those ‘human’ assets – and also aspirational (person-to-abstract) links for brand, reputation and the like.

There also needs to be a clear cross-link between Key Resources and the core Value Proposition, to show why those resources are ’key’ to the operation of the business-model.

Continuing with that, we can move on to the book’s ‘key questions’ for this block:

  • What Key Resources do our Value Propositions require?
  • Our Channels?
  • Customer Relationships?
  • Revenue Streams?

…and also the other blocks not yet referenced: Key Activities, Key Partners, Cost Structure.

We need to watch for potential confusion here between means and content - much as we saw earlier with Channels versus Customer Relations. Strictly speaking, all content should be described in this section, and all means under Key Activities – but it’s not always easy or obvious to identify which is which. For example, for enterprise-architects, we might perhaps choose to place capabilities under this heading: they seem to fit better here, rather than under Key Activities, because they’re often described as ‘resources’ or ‘assets’, and also because they represent potential for action, rather than activity itself.

For most business-contexts, it’s probable that we would also need here to explore:

  • ownership of resources - possessed (‘owned’), leased, just-in-time purchase, futures, etc
  • sourcing of resources - in-source, out-source, right-source, dynamic-sourcing etc

For any ‘out of house’ sourcing-arrangements, we also need to ensure maintenance of the Value Proposition’s vision, values, service-level agreements and suchlike across all sourcing-boundaries – otherwise key intangible-assets such as relationships, reputation and trust may be placed at risk.

For an existing business, a key concern here will be about reuse and re-purpose of existing resources. Each existing resource is something that we won’t have to find for this business-model – but its use in the business-model could constrain or cannibalise other existing business-models. Likewise, if we have to add new resources, we can look for synergies with existing business-models – where they too might benefit from the new resources – yet we also need to be wary about how and where these new resources might conflict with existing ones. In essence, it’s the same kind of trade-off challenge that we saw with Customer Segments in the ‘front-end’ – and getting it right is not easy.

For government, NGOs and other non-profits, we’ll need to take much more care than a commercial business often would about our ‘intangible resources‘ – relationships, commitments, trust and suchlike – because in essence these are likely to be the real core of our ‘the business of the business’. Importantly, these don’t usually arrive via the same ‘supply-chain’ mechanisms as for physical-resource (‘things’) or virtual-resources (information) – we often need a broader form of value-stream mapping, combined with value-network mapping, that does track all forms of value in context as the respective value, not merely as things or information or money.

7: Key Activities

This block is shown in the upper centre-left of the Business Model Canvas, and is summarised on p.36 of the book as:

Describes the most important things an organisation must do to make its business-model work.

The book categorises Key Activities as:

  • production
  • problem-solving
  • platform/network

…which is a very incomplete subset, even for a summary of this type. An alternative approach might use a classic ‘functional business model’ categorisation:

Or, perhaps, and more in line with Business Model Canvas:

  • Manage
  • In
  • Do
  • Out
  • Support

From my own experience, I would recommend using Enterprise Canvas as another alternate base-map for such categorisation, but in practice it really depends on the industry and context: for example, the telecoms industry uses eTOM/Frameworx, the logistics industry uses SCOR, and so on.

As with Key Resources, all Key Activities would need to be anchored to the shared-enterprise vision via the Value Proposition. (Once again, note here the crucial distinction between Value Proposition and Offer - respectively, linkage to core vision and values, versus deliverable product and/or service.)

In the book, the ‘key questions’ here are much the same as those for Key Resources:

  • What Key Activities do our Value Propositions [Offers] require?
  • Our distribution-Channels?
  • Customer Relationships?
  • Revenue Streams?

Again as per Key Resources, this list needs to be extended to all of the other blocks in the Business Model Canvas: Key Resources, Key Partners and Cost Structure. And also much as before, we need to watch for potential confusion between means (activities) and capability (potential); and also explore sourcing-issues for all Key Activities – including enterprise-architecture complexity-themes such as mutual-interdependencies, variety-weatherrequisite-scatter and so on.

For an existing business, the same concerns apply as for Key Resources: opportunities for synergy, risks of conflict, and risks of cannibalisation or constraint on other existing business-models.

For government, NGOs and other non-profits, to parallel that necessary focus on ‘intangible resources’ in Key Resources, we’ll need to identify and track activities that act on those resources – or, perhaps even more important, to identify, track and resolve when and where they don’t act on those resources when they need to do so.

8: Key Partners

On Business Model Canvas, the block is shown over on the far left. The book summarises this block, on p.38, as:

Describes the network of suppliers and partners that make the business model work.

The book suggests four different types of partnerships:

  • strategic alliance between non-competitors
  • coopetition - alliance between competitors
  • joint venture
  • buyer-supplier relationship “to assure reliable supplies”

We might note also that an ordinary ‘hands-off’ or contract-only buyer-supplier relationship actually is also a type of partnership – just not as explicit in its form as a relationship of partners in business.

As motivations for partnership-type relationships, the book suggests:

  • optimisation and economy of scale
  • reduction of risk and uncertainty [and also enabling of shared-opportunity]
  • acquisition of particular resources and activities

The practical problem is that gives us no real description of how partnerships actually work. For example, we’ll often need to bridge between boundary-of-control and boundary-of-identity; we need shared-vision and shared-values to cross that bridge; and we need mechanisms and measures to support joint reputation and trust. These are all essential here.

And as before, all of this needs a solid understanding of non-monetary forms of value as well as the more easily identifiable monetary forms. Other than in rare and extremely-simple cases, the business-model will not work without that understanding, and without careful modelling and tracking of non-monetary forms of value.

In the book, the ‘key questions’ to consider here are:

  • Who are our Key Partners?
  • Who are our key suppliers?
  • Which Key Resources are we acquiring from partners?
  • Which Key Activities do partners perform?

Unlike the ‘customer-facing’ side of the business-model, the book’s description of this ‘supplier-facing’ side is very sketchy and incomplete. This needs much more clarity and detail on themes such as interfaces, service-level agreements, risk- and opportunity-management, trust/reputation management, impacts on Cost Structures, and much, much more. A business-model can be made or broken on the ‘supplier-side’ just as much as on the ‘customer-side’.

For an existing business, probably the key challenge – much as with the other Canvas blocks – is in the trade-offs and balances between existing Key Partners and any new ones. Whilst ‘hands-off’ basic supplier-links can perhaps be made and broken-off with relatively little cost on either side, it takes time to build the deep-trust needed for fully viable and supportive partnerships: and as with all other parts of a new business-model, the risk here is that we may lose existing partnerships before we can bring others fully on-stream.

There are also other complexities that are perhaps too-easily glossed-over in the bare term ‘partners’. For example, in a business-consortium, a partner will often take on both ‘supplier’-type and ‘customer’-type roles – in other words, will in effect appear on both sides of the business-model. Other Key Partners may have supplier-like yet non-transaction roles: an alliance with a university research-unit, for example, or a standards-body. And as shown in Enterprise Canvas, we perhaps also need to distinguish between supplier-like Key Partner roles, versus investor-type Key Partner roles – because the effective value-flows connect different forms of value in different ways to different parts of the business-model:

For government, NGO or other non-profit, we’d probably say much the same – but louder, perhaps? We’d also need far more emphasis on the non-monetary forms of value-flow between the organisation and its Key Partners.

It’s perhaps also useful to view all stakeholder-groups as some variant on a theme of Key Partners – whether as ‘customer’, ‘supplier’, ‘investor’, ‘beneficiary’ or whatever. And it can be wise, too, to remember that the term ‘stakeholder’ can also be interpreted as ‘person who wields a sharp-pointed stake’ – with us as the potential target!

9: Cost Structure

This block is shown on the lower-left of the Business Model Canvas. In the book, this is summarised on p.40 as:

Describes all costs incurred to operate a business model.

Which is fair enough, in principle; yet in practice the book discusses only monetary costs – not costs in terms of the whole range of values in play in the overall business model. As a result, following the book’s descriptions alone may lead to a business-model description that is dangerously-incomplete.

Such costs can be calculated relatively easily after defining Key Resources, Key Activities and Key Partnerships.

This might perhaps be true for monetary-costs. For non-monetary costs, though, the respective metrics and transforms are rarely easy to model, or even to understand at all: this can often be one of the hardest challenges in designing and verifying the viability of a business-model. At the least, we need to cross-link back to Value Proposition in order to identify what the values are that are in play in this business-model.

(Even for monetary costs alone there may be complexities arising from non-linear and/or non-reversible transforms from other forms of value: for example, outsourcing can often carry a risk of longer-term cost in loss of ability to innovate – a cost which is ‘invisible’ unless we explicitly acknowledge it as a potential cost. The same applies also to well-understood yet difficult-to-model concerns such as opportunity-cost.)

The book suggests just two business-model cost-structure types:

  • cost-driven: aim for lowest-possible cost-structure (e.g. budget airline)
  • value-driven: emphasis on perceived-value such as premium service (e.g. luxury hotel)

In this context, ‘value’ appears to be viewed as an optional add-on – perhaps not the wisest move?

For cost-structures themselves, the book suggests the following characteristics:

  • fixed-cost: will (usually) be same regardless of scale (e.g. manufacturing-plant)
  • variable-cost: linked to volume, scale, consumables etc
  • economies of scale: costs reduce with increasing scale
  • economies of scope: re-use in other contexts

As usual, there seems to be a tendency to view ‘cost’ solely in monetary form (as also indicated by the ‘C$’ two-character tag for this block on Business Model Canvas diagrams). Failure to understand and include non-monetary costs will cause the business-model to fail.

The book lists the following as ‘key questions’ for this block:

  • What are the most important costs inherent in our business model?
  • Which Key Resources are most expensive?
  • Which Key Activities are most expensive?

Which is all well and good, but ‘expensive’ in what sense? For example, the use of a particular resource or activity may have low cost in monetary terms, but huge costs in reputational terms (e.g. ‘blood diamonds’, or use of child-labour). Such costs and their concomitant risks will often be rendered ‘invisible’ in a monetary-only mapping of costs – leaving the business-model at high risk of ‘unexpected’ failure from impacts of anticlient-activities and the like.

In parallel with those ‘key questions’ above, other concerns we would need to address here would include:

  • Which Customer Segments and Customer Relationships are most and/or least expensive? – and in what sense of ‘expensive’?
  • How would we identify the cost of a customer? – and ‘costly’ customers to avoid?
  • How do we link to Value Proposition, to identify what ‘cost’ is in this business-model?
  • How do we identify and address costs that cannot be resolved solely by monetary means? – e.g. time-cost, skills-availability / skills-shortage cost, many types of relational and/or reputational cost?

For an existing business, all of the above applies – the need to track non-monetary costs, and costs of transforms between different types of values and value-flows – yet all with the additional concerns about costs arising from the transition from old to new business-model.

We might also need to be wary that ‘hidden’ costs that were quietly absorbed by the old model may not be absorbed in the new model – and vice versa, of course. It’s only by deliberate ‘surfacing’ of such costs that we start to have choices about them – otherwise the business-model, again, may be placed at risk of ‘unexpected’ failure.

For government, NGO or other non-profit, probably the core point of the entire business-model is that it is not focussed on monetary profit – and hence, in turn, identifying and tracking non-monetary costs may well need a much higher prominence here. The more common danger, perhaps, is a failure to acknowledge the importance – in the broader culture, at least – of the monetary costs of operating any type of organisation, whether ‘for-profit’ or ‘not-for-profit’. There are just two pages in the whole of Business Model Generation that explicitly discuss non-profits (on p.264-5, in the Outlook section at the end of the book), and even those in essence consider only the financial aspects of business-models for non-profits – but it represents a very useful reminder that, even if in a sane economics money should not matter, in this culture and economics, unfortunately, it definitely does…

The other key point, perhaps, is to remember that the term ‘non-profit’ can easily be a dangerously misleading misnomer. That supposedly so-important distinction between ‘for-profit’ and ‘not-for-profit’ is merely a side-effect of an arbitrary choice, in our culture’s concept of ‘economics’, to regard money as the central measure of all value – and it’s not a wise choice at all. It leads, for example, to the delusion that only monetary-oriented ‘for-profit’ businesses are of value to the economy, and that so-called ‘non-profits’ are a kind of optional-extra that don’t really matter in the real scheme of things. (To be blunt, the opposite assertion is actually much more realistic.) Instead, we need to understand that every organisation is actually ‘for-profit’. The standard notion of ‘profit = return – costs’ applies to every organisation: the only difference between them is the forms of value that apply to ‘return’ and ‘cost’, and thence to the respective ‘profit’.

Once we get clear on that, it should then also become clear that the only thing that varies is the forms of value that flow through each type of business-model, and the transforms of value that take place within it. The same basic principles of business-models and business-model design apply to all types of organisations, at every scale.


Anyway, that’s it for now: best stop here. Over to you for comments on this overall series of posts, if you would?

Tagged with: , , , , ,
Posted in Business, Complexity / Structure, Enterprise architecture
4 comments on “Business Model Canvas beyond startups – Part 3: Back-end
  1. Ian Glossop says:

    Hi Tom,

    Have you ever come across the “Cambridge Manufacturing Leaders’ Programme Audit Model”?

    Very similar, conceptually at least, to your Enterprise Canvas Model – except more academically justified.

    May I suggest some reading:

    Hillier, W., (2001), “The Manufacturing Business Audit”, Manufacturing Leaders Programme, University of Cambridge.

    Phaal, R., Farrukh, C.P. & Probert, D.R., (2001), “A Framework for Supporting the Management of Technological Innovation”, The Future of Innovation Studies Conference, Eindhoven Centre for Innovation Studies, Eindhoven University of Technology, (NL).


    • Tom G says:

      Hi Ian

      “Have you ever come across the “Cambridge Manufacturing Leaders’ Programme Audit Model”?”

      No, I haven’t – thanks for the pointers on that.

      “Very similar, conceptually at least, to your Enterprise Canvas Model – except more academically justified.”

      Hmm… Perhaps unlike you, I tend to sense a ‘red-rag time’ when I see the words ‘academically justified’, mainly because of a lot of painful first-hand experience about the quality and/or usability (or lack thereof) of much of what claims to be ‘academic’. I’ve seen way too many cases where people had quoted as ‘absolute truth’ academic-papers they’d clearly misread or misinterpreted – often wilfully so for ‘political’ reasons. Following the trails of citations back to original sources often highlights fascinating examples of the children’s game of ‘Chinese Whispers’, where purported ‘facts’ used in policy-documents and suchlike too often turn out to be almost the exact opposite of the results of the original research. And so on, and so on, and so on…

      Just to add to the ‘academic’ mess, most academic-style research work in business and EA and the like attempts to hold to a ‘scientific’-style paradigm, which in turn depends on an exact repeatability of context – which very applies in the real-world maelstrom of real-world enterprise and business. Which means that the best we can do with ‘academic’ assertions is to treat them as possibly-useful guidelines – it’s not a good idea to think of anything they say as being ‘fact’ (other than in contexts such as politics and technology hype-cycles, where its strong natural tendencies towards circular-reasoning are all too easily ignored…).

      Hence why I much prefer to work as much as possible from first-hand first-principles observation, using careful testing of occasional ‘academic’ ideas for possible utility, and a lot of cross-checks for their concomitant logic-traps such as circular-reasoning. Yes, there’s a significant risk of ‘reinventing the wheel’ that arises from doing that, but the human and other costs are usually a lot less than going via the ‘academic’ route. The utter shambles created by most so-called ‘economics’ should be proof enough of the wisdom of avoiding the literally-lethal self-delusions of ‘academically justified’…

  2. Ian Glossop says:

    > even if in a sane economics money should not matter, in this culture and economics, unfortunately, it definitely does

    Money, is just the conventionalised and codified information-mechanism for the quantification and exchange of power – the power to produce outcomes. As such it matters enormously.

    The cognitive failure in economics, is not located in the concept of “money” – but in the failure to accurately identify and quantify all exchanges or exercises of power – and focus on only the exchange of only material goods and resources (and the neglect of ‘externalities’). The different forms of ‘value’. There is, for example, no credible method in economics for quantifying customer-base ‘trust’ in the firm’s brand – and calculating the cost of ‘brand-weakening’ actions.

    • Tom G says:

      Ian -

      We agree on the notion and importance of social concepts of power; as to how they intersect with the manifold delusions of ‘money’ and its ilk, let’s just say it’s best if we agree to disagree?

      “There is, for example, no credible method in economics for quantifying customer-base ‘trust’ in the firm’s brand”

      Agreed. Except I’d take it a step further, and say that in most of present-day ‘economics’ “there is no credible method” for quantifying anything other than its own circular-reasonings and self-delusions. But I’m perhaps a bit extreme, I guess… :-|

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Books by Tom Graves
Ebooks by Tom Graves