Sales business case template for account executives: open with the prospect’s own screenshot

A sales business case template sets out what a buyer’s finance lead needs to approve a deal: the problem, its cost, the options, the plan and the proof. Put the prospect’s own problem on page 1, dated, with the screenshot or email that shows it, such as a basket reminder that never arrived. Obsession keeps what each prospect shows only to its customers, dated with a screenshot, and looks it up by API, MCP (the standard way AI agents call tools) or a table.

A sales business case template is the document a buyer’s finance lead reads before signing: the problem, its cost, the options, the plan and how the result will be proved. The template here is for account executives at B2B software companies whose product touches what their prospects’ customers see. That covers email and SMS, help desk and AI support, checkout and payments, and identity and onboarding. Its first page is the prospect’s own problem, dated, with the screenshot or email that shows it.

Most published templates use five sections. Dock’s runs from an executive summary and current challenges to the business impact and the investment, and is built for the chief financial officer. Mutiny’s puts the problem first, then the cost of inaction. Both frames are sound, and in both the seller writes the problem section. A first page that shows the problem happening at the prospect, dated, can be checked by their finance lead in a minute, with nobody presenting it.

Most of those people never meet the account executive. Forrester’s release also says 86% of B2B purchases stall during the buying process. A case that gets forwarded has to make its point on page 1, in evidence the reader can open.

The sales business case template, section by section

The template has 7 sections, written in the prospect’s words and numbers. Copy the table into a document and fill one row at a time; page 1 comes from your own sign-up at the prospect, and the numbers come from discovery.

The order follows how a finance reader tests a proposal. HM Treasury’s Green Book is the UK government’s guide to appraising spending, updated on 28 October 2026. It builds every business case on 5 cases: strategic, economic, commercial, financial and management. Sections 1 to 3 here are its strategic case and section 4 its economic case. Sections 5 to 7 cover its commercial, financial and management cases, cut to what a software purchase needs.

Page 1: the prospect’s own problem, dated

Page 1 holds one fact the prospect can check for themselves. It shows something that happened to a customer of theirs, which their own dashboards may not show, because a missing email never appears in a sent report. It has 6 parts.

• The fact, in one sentence a finance lead can repeat: “A basket left on larkspur.example at 09:12 on Tuesday 3 November got no reminder by 09:12 on Thursday.”
• The journey that produced it: what you did, as which customer and from which country, timed to the minute with the time zone.
• The evidence: the screenshot with its date visible, or the original email with its arrival time. The original email is the stronger of the two, because it carries its own time.
• The window: how long you watched, and that it has closed. Call an email missing only once the window you chose has closed, whether that is 24, 48 or 72 hours.
• What it means for their customers: the shopper who left $64 in the basket heard nothing back.
• What it doesn’t show: their sales figures, or why it happened. Say so on the page, and the rest of the page is trusted more.

Rerun the journey the day before the case goes out. If the problem has been fixed, say so and give the date: the buyer sees that you checked, and you open on the fix instead. A missing abandoned cart email is the clearest page 1 for an email vendor, and each vendor in the examples further down has its own.

Page 2: what it costs, in the prospect’s numbers

Page 2 turns the fact into money with three inputs: how often the moment happens each month, how much each occurrence used to bring in, and how long it stays broken. Each input belongs to the prospect. Ask for them in discovery, and use labelled example inputs only until they arrive.

For an email and SMS vendor, the moment is a checkout left with an email address on it, and the value is what the reminder used to earn. Klaviyo’s 2026 abandoned cart benchmarks draw on more than 110,000 Klaviyo customers. They put the average abandoned cart email at $6.77 in revenue per recipient, and $13.70 for the top 10% of brands. Multiply the prospect’s abandoned checkouts with an email by that figure, or by their own from the flow report before the reminder stopped.

Revenue per recipient is the revenue a flow is credited with, and some of those shoppers would have come back without a reminder. Say which figure the case uses, and let the finance lead apply their own view of how much is new revenue. A case that states its own assumptions is easier to approve than one that hides them.

Business as usual, then the options

Start the options with what happens if nothing changes. The Green Book says the business as usual option must always go forward to the shortlist, because “It acts as a benchmark against which other options are compared”. At the $100M+ example, business as usual costs $4,062,000 over 12 months if nothing else changes. That is $2,031,000 over 6 months and $1,015,500 over 3.

• Business as usual. The page 1 problem stays, and the section 3 cost keeps running.
• Do minimum. The smallest change that fixes page 1, such as restarting the one flow that stopped. The Green Book uses this option to show whether the bigger options add real benefits or only “gold plating”.
• Fix it in house. Who on their team would do it, how many weeks it takes, and what they would stop doing meanwhile.
• Another vendor. Describe the other product fairly, by what it does well, and name a competitor only where you can describe it from its own pages.
• Your product. The preferred option, with the reason it beats the do minimum on their own numbers.

The plan, and how the result will be proved

The plan names an owner and a date for each step. The steps are the contract, the integration, the first live flow or queue, and the reviews at 30, 60 and 90 days. Put their people’s names next to yours, so the plan reads as a joint one.

The proof reuses page 1. After go-live, run the same journey with the same steps. Put the two runs side by side: the basket left on 3 November with no reminder, and the one left after go-live with its reminder’s arrival time. The buyer can repeat the check themselves, which makes the result hard to argue with, and the 3, 6 and 12 month reviews repeat it again.

Sales business case examples for three kinds of vendor

The same 7 sections fit any vendor whose product touches what a prospect’s customers see. Only page 1 and the cost line change.

An email and SMS platform selling to an online brand

Page 1 is the basket at larkspur.example in the picture at the top. Its welcome series arrived in 4 minutes, so the list works, but no basket reminder came in 48 hours. The cost line is abandoned checkouts with an email × revenue per recipient. Do minimum is restarting the reminder; your product adds the browse and text flows the brand doesn’t send. The proof is a second basket after go-live, with the reminder’s arrival time.

A help desk and AI support vendor selling to a retailer

Page 1 is at corvelhome.example, a homeware retailer. A question about a late order, sent through the contact form at 10:02 on Monday 2 November, was answered at 17:14 on Tuesday. That is 31 hours, and its help page promises 24. An AI support vendor can use the chat assistant instead: it said returns are accepted for 60 days, and the returns page says 30. The cost line is contacts a month × the share answered after the promised time × the cost of the extra contact each late answer brings. The proof is the same question, through the same form, after go-live.

An identity and onboarding vendor selling to a savings app

Page 1 is at pennant.example. On a phone, the application stopped at the address page, the third of 6 steps, where the Continue button didn’t respond. No email followed in 48 hours to help the applicant finish. Walk it as yourself, and stop at the identity check. The cost line is applications started a month × the share lost at that step × the value of a funded account, from their finance team. The proof is the same application on the same phone after go-live.

Discovery call prep: collect page 1 before the call

Bring page 1 to discovery, and leave with the inputs for page 2. A dated problem gives the prospect a reason to open their own reports on the call, which is where the volumes, rates and values come from. Each step has a pass mark.

1. Pick the journey your product touches. Email and SMS: sign-up, browse and basket. Help desk and AI support: a question through the chat and the contact form. Identity and onboarding: the application, up to the identity check. Pass: 1 journey for each product you sell them.
2. Sign up as yourself, with a fresh address for each prospect, through the public form any customer uses. Pass: every message in that inbox comes from one company.
3. Do what a customer does, and stop where a customer pays or proves who they are. Pass: each step has a screenshot with the date and time visible.
4. Wait for the full window, then check spam and Promotions. Pass: the window has closed before you write “none arrived”.
5. Save the original emails as well as the screenshots. In Gmail, Show original keeps the arrival time with the message. Pass: every line on page 1 links its evidence.
6. Rerun the journey the day before you send the case. Pass: page 1 is dated within the last week.

On the call, say how you saw it: “I signed up to your store as a customer last Tuesday, and the basket I left got no reminder.” Then ask what the basket reminder earned before it stopped, and how many shoppers leave a checkout each month. The same fact can open the first email, as in our cold email first line examples, and it is one of the buying signals examples a sales development team can route. Read each company’s terms first, buy nothing you don’t intend to keep, and treat this as general information, not legal advice.

Where page 1 can come from

Most sales teams use similar data, from contact databases and intent tools to Exa’s search of the public web. Exa’s home page calls Exa “The world’s data for AI agents” and says it searches “the largest index of public web and private information”. Its page for GTM teams offers “70M companies, 1B+ people profiles, and the live web”, written to the CRM with a source link for each update. Exa’s pages don’t describe signing up at the companies it covers or keeping what they send their customers, which is the fact page 1 needs. Our comparison with Exa sets out what each holds. We make Obsession, so every other option is described from its provider’s own pages as of 11 November 2026.

Use Exa for what an account publishes, and the by-hand method for a few deals in their last weeks. If the deal is against a named competitor, our battlecard template keeps that side of the case current.

Where doing it by hand stops

Each prospect needs its own address, each journey takes about 20 minutes to run and write up, and your sheet starts the day you sign up, so it holds nothing from before. Nothing tells you when a fixed problem breaks again. Take an account executive with 30 open opportunities, checking 2 journeys at each and rerunning each before the case goes out. That is 30 × 2 × 20 minutes × 2 runs: 2,400 minutes, or 40 hours a quarter. Those are example inputs; put in your own count.

Obsession collects what companies show only to customers, step by step: their screens, emails and texts, member prices, trial limits, quotes and cancel offers, each dated with a screenshot. Teams and AI agents look up any company through the API or MCP, and a company not yet covered is added on request. For sales teams, prospect intelligence and account watch put a dated page 1 behind every business case in the pipeline. They keep it current for as long as you follow each account, so you can see what it does to win rate, deal size and sales cycle.

The sample holds a page 1 fact in full: one real store’s emails and screenshots, each with its time. It comes from the web behind the login, the part search engines don’t reach. Sales intelligence from behind the login shows the same view for every role on a revenue team, with an example for each.

Questions