How to read a Forrester TEI study for your business case
Sort a Forrester TEI study's inputs by their printed sources. Replace non-interview figures with your own and test the interview figures before using them in your case. Then rerun the line that carries most of the benefit before any ROI reaches your approvers.
A vendor has sent you a Forrester Total Economic Impact study with a large ROI, and someone wants the number in the business case. The figure belongs to a model, and the model is built from rows you can read. In the Red Hat study's benefit and cost tables, the Source column attributes inputs or shows how a row is calculated. Some inputs are attributed to interviews; others use a composite organisation, standard values, research figures or assumptions. In the Red Hat study read below, the developer time savings behind its largest benefit line rest on six inputs: two come from the interviews, and the other four can be replaced with figures from your own business.
What a TEI study is
Forrester describes the Total Economic Impact method on its methodology page. It has "four components to evaluate investment value: cost, benefits, flexibility, and risk". The same page says that "clients cannot purchase favorable opinions or results" and that "All TEIs are clearly marked as 'commissioned'".
The Red Hat Technical Account Manager study of March 2023 shows how such a study is built. Red Hat commissioned it. Forrester "interviewed four representatives with experience using TAMs" and combined their experiences into a composite organisation. The study adds that "Red Hat provided the customer names for the interviews but did not participate in the interviews." For that composite it reports risk-adjusted present values over three years of $4.99 million in benefits and $1.04 million in costs, and an ROI of 379%.
The study itself tells you what to do with the result: "Forrester strongly advises that readers use their own estimates within the framework provided in the study to determine the appropriateness of an investment in TAMs." For how a total cost of ownership and an ROI differ, and what approvers check, see the business case for a software purchase.
Read the Source column
In the Red Hat study, the benefit and cost tables include columns for Ref., Metric, Source and each year. The Source column attributes inputs or shows a calculation. It uses five labels for inputs:
1. Interviews: an input attributed to customer interviews. It may be a modelled estimate, not a measurement from your organisation.
2. Composite: a characteristic of the composite organisation, such as its number of developers.
3. TEI standard: a standard value the method uses, such as a salary.
4. Forrester research: a figure from Forrester's other research, such as the cost of an outage.
5. Assumption: a figure the model assumes.
Other studies print other labels. The Writer study of April 2025, for example, uses "TEI methodology". Work from the labels your own study prints.
Some rows hold a formula in the Source column, such as A1*A2. Those rows are calculated from the inputs above them. Change the inputs and let the calculated rows follow.
The line that carries the Red Hat case
The Red Hat study has four benefit lines. One of them, improved developer and IT productivity, is $3.3 million of the $4.99 million in risk-adjusted present-value benefits: about two thirds of the case.
Its first part, developer time savings, multiplies six inputs.
In year 3 the study's subtotal is $2,500,470, the product of all six.
The line's second part, IT team time savings, takes two interview-sourced figures: 7,500 hours a year spent resolving issues before a TAM and 252 hours with one. It values the hours saved at the A3 salary divided by 2,080, with the same 50% recapture. The study displays the hourly rate as $48 and the subtotal as $172,880 a year. To reproduce that subtotal, use the unrounded rate of $99,225 divided by 2,080, about $47.70. Multiplying by the displayed $48 instead gives $173,952. The line adds the two parts, $2,673,350 in year 3, before its risk adjustment.
| Input in the study | Value in the study | Source as printed | What you put in its place |
|---|---|---|---|
| A1 Number of developers employed | Value in the study 1,400 | Source as printed Composite | What you put in its place Your number of developers |
| A2 Time spent directly on applications and related projects | Value in the study 60% | Source as printed Assumption | What you put in its place Your share, from people who do the work |
| A3 Average fully burdened annual developer salary | Value in the study $99,225 | Source as printed TEI standard | What you put in its place Your fully loaded salary cost |
| A4 Portion of projects affected | Value in the study 15% | Source as printed Interviews | What you put in its place Keep as an interview-sourced estimate, then test it |
| A5 Time saved with TAM guidance | Value in the study 10%, 25%, 40% | Source as printed Interviews | What you put in its place Keep as interview-sourced estimates, then test them |
| A6 Productivity recapture | Value in the study 50% | Source as printed TEI standard | What you put in its place How much saved time your teams turn into other work |
A worked example, with invented figures
Suppose your company has 300 developers, they spend 45% of their time on applications, the fully loaded cost is $120,000, and you expect 30% of saved time to become other work. Keep the interview figures for year 3, 15% and 40%. The same multiplication gives 300 x 45% x $120,000 x 15% x 40% x 30% = $291,600. For the second part, use your own IT hourly cost and recapture assumption, and test the interview-sourced hours against your own team's work. Add the two parts and apply your own risk adjustment. Valuing time at salary rates is not automatically a cash saving: say what the released time would enable or which cost would actually be avoided.
Four of the six developer inputs now come from your own business. The other two are labelled as interview-sourced estimates from the study, pending a check with your own people. An approver can challenge any one of them without throwing out the case.
Risk adjustments go both ways
Forrester adjusts each line for risk. In the Red Hat study the four benefit lines are adjusted downward by 20% (developer and IT productivity), 15% (outages), 10% (technology fees) and 15% (security and compliance). The two cost lines are adjusted upward, by 5% (Red Hat fees) and 10% (ongoing management). The present values are discounted at 10%.
When you put in your own figures, decide your own adjustment for each line and write it down. A figure from your own systems may need less adjustment than an estimate from one interview; a new process may need more.
What the interview rows tell you
The rows labelled Interviews are inputs Forrester attributes to a small number of customer interviews. They may be modelled for the composite organisation. Forrester's risk adjustment is applied to the total of each line, not to these rows on their own. Use the interview-sourced figures as a starting point, test them with your own people and replace them where your evidence supports a different estimate.
The way to test them is to ask the people whose work would change, in a Business Value Assessment run as interviews. The Business Value Assessment template asks, for every number, which task it refers to, how often it happens and what the estimate is based on. Put the study's interview figures next to your own answers and record where they differ.
If you are the Sales Engineer handing the study to a customer, sort the rows with them and plan the interviews that test the figures they keep.
What to show your approvers
1. The study, who commissioned it and how many customers were interviewed.
2. The benefit line that carries most of the case, with its inputs and their printed sources.
3. Your figure in place of every input that is not from the interviews, with what it is based on.
4. The interview-sourced estimates you tested, what your own people said and any figures you changed.
5. The adjustment you applied to each line, upward for costs and downward for benefits.
6. The ROI recalculated on your benefit and cost estimates, risk adjustments and discount rate, next to the study's.
Sources
- Total Economic Impact Methodology. Forrester, undated.
- The Total Economic Impact of Red Hat Technical Account Managers. Forrester Consulting (commissioned by Red Hat), March 2023.
- The Total Economic Impact of Writer. Forrester Consulting (commissioned by Writer), April 2025.
Facts last checked on 6 November 2026.