The benefit of a flexible construction model is that it can now be updated with real data to offer insights into how the fund is performing - we have crystallized these workflows into Tactyc.
Last updated
Why is portfolio construction important? Not just because LP’s will ask for it!
Many managers believe portfolio construction to be a one-off activity. Construction models are primarily used in fundraising, when LPs review your assumptions and ask questions around reserves and check sizes.
Post-launch, the original construction model is frequently “thrown away”. Models are viewed as theoretical exercises - with limited practical use for active fund management.
This is a mistake, in our opinion.
At Tactyc we have seen that successful managers build construction models in such a way that they help GP’s guide actual fund performance. In this post, we’ll look under the hood to shed light on these best practices for constructing fund models that every manager can employ at their fund.
Elements of an effective construction model
At Tactyc we have helped hundreds of emerging and established managers build their construction strategy. We have also reviewed countless spreadsheet-based construction models and found the following common themes among effective construction models:
1. Clear model inputs and outputs
Many managers fall into the trap of confusing model outputs with model inputs. For example, some managers start off with “we need a follow-on reserve of 40%” or “we expect to invest in 40 companies” and then build a model based on those assumptions.
Instead, great construction models start off with more basic assumptions (i.e. what is the graduation rate you expect in your portfolio, what are your target entry ownerships) and use them to build up to reserve ratios, number of investments and expected portfolio performance.
Determine the smallest number of variables that are fixed and let the model tell you bigger picture metrics such as follow-on reserves, number of investments etc.
In Tactyc, we guide managers through a checklist of questions such as expected graduation rates, check sizes or target ownerships in each round - and calculate the reserve ratio and expected number of deals as outputs instead of inputs.
For example, the follow-on amount at the fund level is calculated based on:
Number of deals graduating (Graduation Percentage) x How often the fund will follow-on (Follow-On Participation %)
2. Flexible
Another common approach we have seen is to model a fixed portfolio of X investments with some failures and success rates as a representative portfolio.
The problem? This is a static model that represents one scenario and is not flexible enough to stress-test with other scenarios. If you need to change exit assumptions, follow-on reserves or target ownerships that might result in a completely different portfolio strategy - and it might require a complete rebuild of your original static portfolio.
In Tactyc, managers can build multiple fund scenarios (e.g. a higher reserve or lower reserve scenario) and compare them with each other side by side to evaluate which might be a better fit.
3. Real-world data
Finally, your model assumptions should be based on real-world data. If you’re investing in FinTech, to get a 10% ownership at entry in a Seed stage requires you to collect and find average valuations in the sector - and your expected check sizes should be based on this market data.
Similarly if you are expecting a 40% reserve ratio, you should collect graduation rates for your sector (how many companies on average graduate from Seed to Series A for example) - as that heavily influences the potential follow-ons your fund has access to.
This also enables you to update market assumptions around round sizes, valuations and graduation rates to quickly generate a new performance forecast for comparison if the market has moved significantly in the future.
In Tactyc, we ask every manager to decide on a sector profile where they can manage assumptions on round valuations, graduation rates and exit rates. We provide real-world benchmarks by industry to guide them through this process.
How does this guide actual fund performance?
The benefit of a flexible construction model is that it can now be updated with real data to offer insights on how the fund is performing. For example, if your fund is 30% deployed, you can update the model to include your actual deals - and let your model forecast the remaining 70% to be deployed as per your construction assumptions.
This arms the GP with:
Actual vs. Planned: Were our original valuation and check size assumptions too rosy? Has the market moved significantly since we launched?
Projected Returns: By incorporating actual investment data the model can now start projecting expected returns and give you a line of sight into potential DPI, TVPI and other return metrics.
Course Correction: How can the fund “get back on track”? Should we change our allocation or check size strategy going forward? Questions such as these can be answered once you have layered your actual data on top of your construction plan.
This is a powerful feedback loop that many managers use to drive outperformance. By having a constantly “alive” portfolio model that not only projects future returns but offers comparisons with original construction plans, GPs are armed with more data-driven insights on future capital deployment.
Closing Thoughts
At Tactyc, we have seen repeatedly that this workflow, while powerful - can be difficult to execute with spreadsheets. Excel workbooks can quickly get out of hand as GPs need to build and maintain multiple scenarios for comparison.
We have crystallized these workflows into Tactyc so any manager can construct, manage and strategize their fund without spending time on mechanically building complicated spreadsheets. Please reach out to anubhav@tactyc.io if you’d like to learn more or schedule a demo with us here.
