Strategic alliance vs joint venture, and the line between them
Strategic alliance vs joint venture, settled on structure not trust: the autonomy test in EU merger law, the deadlock question, and when a contract wins.
ReadMost account plans describe an account somebody already understands, then never open again. Bain’s survey of 870 B2B executives points at why, and at the one field almost every plan leaves out.
Most key account plans are a tidy summary of an account somebody already understands. Org chart, revenue history, a list of products the client does not buy yet, three objectives written in the language of a board paper. It gets built in January, presented once, and never opens again.
The test of an account plan is not whether it describes the account well. It is whether it changes how somebody spends next Tuesday. On that test, most plans fail, and they fail in a specific way: they record what you know instead of deciding what you will do.
Bain's brief on account time allocation, Is That Customer Worth Your Time?, puts the problem bluntly. From a survey of 870 B2B executives worldwide, fewer than 20 per cent said they had a data-driven, quantified understanding of their total market opportunity and untapped customer potential. Fewer than a quarter had an account management process that identified critical actions, such as cross-sell opportunities, based on an understanding of how customers make decisions.
The same brief makes the point that matters for your plan: coverage should be aligned to future potential rather than to last year's revenue. The phrase its authors use is that reps return to the same watering hole, overlooking the untapped pockets they have not had time to uncover. Small accounts often absorb a disproportionate amount of service without justifying the high-touch model they receive. Some large accounts are tapped out for growth and belong on a maintenance model.
So the first line of a real account plan is not revenue. It is an estimate of share of wallet, which means what this client spends in your category in total and what share of that comes to you. Bain's observation is that leading companies calculate this from hard data sources rather than from their reps' best estimates. If your number comes from a conversation with the account manager, mark it as a guess and move on, but mark it.
The brief describes a distribution company that combined a share-of-wallet view with an estimate of where reps spent their time, and found a close correlation between the two. A broad customer survey then confirmed that customers were acutely aware of how often they saw a rep, with more than 70 per cent saying increased interaction would lead them to buy more. The company found its best opportunities in medium-sized accounts with real growth potential and little field coverage, accounts that had historically been too small to earn attention.
That is uncomfortable, because it means the plan's most important number is one nobody writes down: how many hours a month this account gets, and whether that allocation matches its potential rather than its history. Put the hours on the page. An account plan without a time budget is a wish, because time is the resource you are actually allocating, not money.
Everything above fits on a single side. If it needs a second page, the second page is reference material rather than plan.
Six fields, and three of them force an admission rather than a description. That is the difference between a plan and a summary.
Annual account plans fail for a structural reason. The plan is written when you know least about the year, and the review arrives when it is too late to act. Ninety days is short enough that the actions are concrete and long enough that something can actually land in a B2B cycle.
The quarterly question is also better. Not "are we on track against the objectives", which invites narrative, but three facts: what did we do, what changed in the relationship map, and what did we learn about their decision process. The last one is where Bain's finding about understanding how customers decide gets operationalised, and it is the one most often skipped.
If your accounts are in Saudi Arabia, be careful about the growth rate your plan quietly assumes. The General Authority for Statistics published Structural Business Statistics for 2025 on 1 November 2026, reporting that business-sector operating revenues rose 2.3 per cent to SAR 5,628 billion while employee compensation rose 6.0 per cent to SAR 624.1 billion. You can see the release on GASTAT's news page.
Read those two numbers together. Revenue growth in the low single digits alongside compensation growth at nearly three times that rate describes a market where clients are under margin pressure even while headline activity holds up. An account plan that assumes your client will fund growth out of expanding budgets is making an assumption that the aggregate data does not support. A plan built on displacing an incumbent's spend, or on taking work the client currently does in-house, is better grounded in that environment than one built on new budget appearing.
These are aggregate figures covering the whole business sector, so they describe the weather and not your client. Use them to pressure-test the assumption, not to forecast the account.
The predictable next question is where the plan lives, and the predictable mistake is answering it with software before the plan has proved it changes anything. A document that six people maintain by hand is fine for a handful of accounts. The case for building something only arrives when the time budget and the quarterly facts are being tracked often enough to be worth automating, and that case needs scoping honestly rather than enthusiastically. Rivl's note on what "real-time" actually means for a sales dashboard is a useful corrective here, because most of what an account plan tracks changes monthly and does not need to be live at all.
Three things this does not solve.
A plan does not create access. If the relationship map has four names and you know one of them, the plan has told you the truth and solved nothing. The work is the introductions, and no document shortens it. What the map does is stop you believing that one friendly contact constitutes a relationship, and the buying committee piece covers how that misreading plays out.
Share of wallet is often unknowable. In some sectors there is trade association data, and in many there is nothing, and then your figure is a judgement. That is acceptable, as long as the plan says so on its face. A guess labelled as a guess still helps you rank accounts. A guess presented as a measurement makes every decision downstream of it falsely confident.
And a small team may not need this at all. With six accounts you can hold the whole picture in your head, and the plan is overhead. The point at which it starts earning its keep is when more than one person touches an account, or when nobody can say from memory what is happening in the third largest one. Below that, write the next ninety days on a page and skip the rest. The same threshold logic applies to account based selling for small teams, which was designed by companies with budget for it, and to how a small team splits the three jobs in the first place.
Send us three things: your three largest accounts by revenue, a rough estimate of what each spends in your category in total, and who in your team touches them. We will come back with a read on which of the three is actually at risk, where the untapped share sits, and whether a plan is the thing you need or just the thing that feels productive.
Book a Free Consultation