Aequitas, the Roman goddess of fair dealing, appears on imperial Roman coins from the middle of the first century CE. She is usually shown holding a balance scale (the trutina) and a cornucopia. The scale stood for evenness, with one pan for what a buyer gives and one for what a buyer gets. The cornucopia stood for the abundance that follows when an economy’s weights and coins can be trusted.
The Directional Policy Matrix (DPM), which scores market attractiveness against competitive strength, helps decide which product lines to invest in, independent of revenue. However, it does not weigh the customer’s side of the scale: how well your product meets the customer’s needs compared with the next-best alternative they could buy. Without that comparison, your plan rests on your costs and your competitors’ catalogs, not on what your customer gets.
The problem the quadrant leaves open
In addition to guiding investment, each DPM quadrant carries a default play, such as invest to grow the product line (upper right quadrant) and harvest for cash (lower right quadrant). Those plays assume the customer sees the same advantage the quadrant implies. Until the Product Manager (PM) checks, that is a guess, and a wrong guess can bury a real advantage or hide a lost one.
Uncovering the customer’s perceived value of the product, including the support from Field Application Scientists and Field Service Engineers that comes with it, is what is missing from the DPM. How well the product meets or exceeds the customer’s needs, compared with the next-best alternative, is the third axis of the DPM: relative perceived value (RPV).
RPV is the test of your value proposition. Lanning and Michaels, McKinsey consultants who coined the term in 1988, described it as a statement of the benefits, tangible and intangible, that you promise each customer segment. The value proposition is what you say, and RPV is how a PM tests whether the customer believes the company’s value proposition against the alternative. Translating that value into an absolute price is a specialist job, and the classic method, economic value to the customer (EVC), is described by Forbis and Mehta. RPV stays on the customer’s side of the scale: ahead of, level with, or behind the alternative.
Ideally, a PM would build this RPV view during the product development process because new benefits are often uncovered along the way. This draft of the RPV can be tested with customers and used to set the product’s positioning and value proposition at launch. However, value proposition and positioning at launch often turn into a ‘one and done’ item on the launch checklist, and they are rarely discussed several years post-launch, even though the next-best alternative keeps moving as competitors enter.
How to assess relative perceived value
1. Name the alternative the customer would buy instead. Pick one customer segment first, because needs and weights differ by segment. This example uses academic core labs. The next-best alternative (Competitor A in the table below) is whatever the customer would use instead to meet the same need, which may be a competitor’s product, the one they use today, or your own previous generation.
2. Choose three or four criteria that reflect the customer’s needs, and weight them. Use the customer’s words, not your feature list. A feature is what you built, such as a one-tube master mix. A need is what the customer is trying to get done, such as a result before the afternoon. In an interview, a lab manager who says she cannot have a run fail the week before a grant deadline is telling you reliability matters. Ask which need she would give up first, and which one made her switch last time. The order and the stories become the weights. Convert the ranking into weights, then let step 5 tell you whether the exact numbers matter. Weights show how much each need matters to this segment and add up to 100%.
3. Score both products from 1 to 5 on each criterion, as in part 1: 1 means the product does not meet the need, 3 means it meets the need, and 5 means it clearly exceeds it. Score the alternative the same way, using win and loss notes, its reviews and publications, and what customers say when they compare the two. Multiply by the weights and add up.
4. Compare the totals. A gap of 0.5 or more is a real advantage or a real shortfall. A gap under 0.5 is parity, because the scores are judgments, not measurements. The 0.5 is a rule of thumb, not a published threshold, so adjust it to how noisy your scoring is.
5. Stress-test the weights. Rerun the totals with two or three plausible alternative weightings. If the conclusion holds, you can act on it. If it flips, the weights are the assumption to validate with customers first. See the example in the FAQ.
| Customer need (criterion) | Weight | Our kit | Competitor A | Gap |
|---|---|---|---|---|
| Time to result (setup plus run time) | 30% | 5 | 3 | +2 |
| Run-to-run reliability | 30% | 3 | 4 | -1 |
| Fit with existing cycler and past data | 15% | 2 | 4 | -2 |
| Field application support | 25% | 3 | 4 | -1 |
| Weighted score | 100% | 3.45 | 3.70 | -0.25 |
RPV table: illustrative qPCR reagent kit, academic core labs. Scores run from 1 (does not meet the customer’s need) to 5 (clearly exceeds it); 3 means the need is met. Weights are illustrative. In practice, set them from customer interviews, or before interviews have the PM, a field application scientist, and a sales rep score independently and average.
The RPV table shows customers rate the kit’s time to result faster than Competitor A’s, but they trust the alternative more, and the cycler fit and support gaps cancel out that advantage.
6. Draw the RPV as a bubble on the DPM. Plot the kit on the Directional Policy Matrix as before, and draw the RPV (third axis) as bubble color: green when it is ahead by 0.5 or more, red when it is behind by 0.5 or more, and grey when it is within 0.5. The kit sits in harvest (lower right) with a grey bubble. Part 1 scored the kit 2.9 on market attractiveness and 3.4 on competitive strength, on 1 to 5 scales split at 3.0, which puts it just inside harvest. Bubble size shows revenue, as in part 1.
| Market attractiveness | High53 | Develop offerFund only the specific gap holding you back, such as distribution or performance data, or exit. | InvestFund R&D, line extensions, and sales coverage to build and defend share. |
| Low1 | Evaluate or disinvestRationalize SKUs, move to distributor-only, or discontinue. | qPCR kitHarvestKeep new R&D minimal, protect margin, and use the cash elsewhere. | |
| Low13 | High5 | ||
| Competitive strength | |||
| Bubble color = RPV versus the next-best alternative: green ahead by 0.5 or more grey within 0.5 red behind by 0.5 or more | |||
Illustrative qPCR kit on the Directional Policy Matrix. Bubble color shows RPV against Competitor A.
A harvest investment plan assumes that customers will continue to buy the company’s kit, but since both kits are at parity for the RPV, the customers have little reason to stay if the alternative improves. In the example, the kit’s sales grew 22% over the last twelve months (moving annual total, MAT), but part 1 showed that growth came from the funding wave lifting every qPCR supplier. Its market share is flat, which fits the RPV result: customers see no reason to pick it over Competitor A.
Repeat these steps for each product line on your DPM, not only the ones in the harvest quadrant. Each bubble color points to a different next step.
| RPV reading | Bubble | What to do next |
|---|---|---|
| Ahead by 0.5 or more | Green | Customers see more value than the alternative. Put the needs where the product line scores highest first in the value proposition. Rescore when a new competitor enters or an existing competitor makes an improvement. |
| Within 0.5 | Grey | Customers see about the same value as the alternative. Look for a heavily weighted need where the product line could score higher, or for evidence customers are missing. Rescore after new customer interviews or a competitor launch. |
| Behind by 0.5 or more | Red | Customers see less value than the alternative. Decide if the gap is missing evidence (fix the content) or a real shortfall (take it to product development). Rescore after the fix or a competitor launch. |
What the customer values decides where you compete
Aequitas held a scale with two pans: what the buyer gives (price paid) and what the buyer gets (value, which RPV measures). In my experience, most companies weigh only the first pan, setting price from cost plus a margin and checking it against competitors’ catalogs. Weigh what the customer gets first, in their terms and by the needs they prioritize, and the pricing conversation has something to stand on.
Revenue alone would have funded more investment in the kit. Part 1’s DPM placed the product line in the harvest quadrant, and part 2’s RPV shows customers see no edge, so even the harvest decision rests on weaker ground than it looks. Each lens adds something the others miss, which is the argument for using market attractiveness, competitive strength, and relative perceived value together before you make an investment decision.
Before your next portfolio review, pick your top product lines. For each, write the three or four needs your target segment cares about, name the next-best alternative, and score the RPV of both. If you cannot score a criterion, you have found the voice of the customer (VOC) research to do first. Date each score so the next review can identify what moved.
FAQs
How do I score a criterion when customers will not give me a number?
Start with evidence you already own: win and loss notes, support tickets, and what field application scientists hear at the bench. Where that is thin, a grounded synthetic customer panel can give you rationale and a ranking of criteria. Grounded means AI personas built from what you know about customers, such as interviews, win and loss notes, and what the market actually did, not a model’s general guess about a lab manager. One study found that grounded synthetic respondents’ rank order of products matched roughly 90% of human test-retest reliability, though on personal care products, not lab tools (Maier et al., 2025). Run the grounded synthetic customers with the brand names blinded, randomize the order, and treat the RPV output as a hypothesis, not a score. Then confirm with five or more real buyers.
What if the weights are adjusted until the product looks good?
This is not uncommon. Rerun the totals under two or three plausible weightings. If the RPV differences hold, act on them. If they flip, as the kit’s do in the table, you can see the one assumption the decision depends on, and that is the customer research to do first. The stress test cannot fix a bias that everyone scoring shares, so confirm the weights with real buyers before you act on an RPV outcome.
| Weighting | Time / reliability / fit / support | Our kit | Competitor A | Gap | Reading |
|---|---|---|---|---|---|
| Base case | 30 / 30 / 15 / 25 | 3.45 | 3.70 | -0.25 | Parity |
| Speed-heavy | 60 / 15 / 10 / 15 | 4.10 | 3.40 | +0.70 | Ahead |
| Trust-heavy | 15 / 45 / 20 / 20 | 3.10 | 3.85 | -0.75 | Behind |
How do I use RPV when sales says the product sells itself?
Take it seriously, because sales hears the buyers who say yes. The question is which buyers sales hears from. Reps talk most to the accounts that already like the product, so their view skews toward the criteria where the product scores highest. Ask sales to score the same four criteria for the kit and Competitor A, then compare their scores with the win and loss notes. Where they agree, you have an advantage to protect. Where sales scores a need higher than lost deals do, you have found the gap to investigate.
